The five-factor model of personality and work behaviour in Greece by Ioannis Nikolaou

The five-factor model of personality and work behaviour in Greece

A thesis submitted to the University of Manchester Institute of Science and Technology for the degree of Doctor of Philosophy

Ioannis E. Nikolaou B.Sc. (Hons.) Psychology (University of Crete, Rethymno, Greece) M.Sc. Organisational Psychology (UMIST)

March, 1999

Manchester School of Management, UMIST

Declaration

No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university, or other institution of learning

Acknowledgements The present study would not have been possible without the invaluable assistance of the participating Greek companies which provided me with access to their employees and managers. Especially the “citizenship” of some of their staff proved once more the existence of the famous Greek “philotimon”. Special thanks are also due to the State Scholarships Foundation of Greece (IKY) for its financial support all these years. Without its generous scholarship, my postgraduate studies would probably remain a dream. Further, my special gratitude is due to my beloved family and my close friends and “relatives”, Sotiris, Sofia, Dimitris, and Marika. Their continuous support all these years was more than precious helping me in very difficult moments. My supervisor Professor Ivan T. Robertson is gratefully acknowledged for his guidance, his patience with me, his intellectual insights, and the opportunities he offered me in becoming the researcher and organisational psychologist I am today. He remain an enlightening example of a serious teacher, researcher and scientist in the years to come, a role model. I hope that our productive co-operation will continue in the future. Finally, last but certainly not least, this thesis is dedicated to Maria, whose companionship, love and continuous support all these years, helped me carry out this research. Her patience, while I was away or around her, her encouragement and her whole existence provided me with much more than any man is looking for from his partner. I am really looking forward to the years to come in order to make our dreams come true.

Αφιερωμένο στη Μαρία μου, τη γυναίκα της ζωής μου όλης

Abstract The emergence of the five-factor model (FFM) of personality has proven to be a very significant step forward both for research and practice in personnel psychology. A number of studies have recently examined its criterion-related validity, and numerous tests, following its structure, have been developed and used from practitioners and human resource departments around the world for selection and assessment purposes. Nevertheless, its relationship with work-related outcomes such as job satisfaction, and employees’ work competencies has not been fully explored. The present study explores the application and validity of the five-factor model in occupational settings in Greece, examining its relationship to employees’ job satisfaction, organisational citizenship behaviour and work competencies. Although similar attempts are usually largely atheoretical, the present study incorporates socialcognitive theory and the role of personality processes in the examination of personality dispositions, as expressed by the FFM, in occupational settings. A number of bivariate and multivariate hypotheses were developed, focusing on the moderating-mediating effects of variables such as job satisfaction, organisational citizenship behaviour and work competencies on the validity of the FFM. Two hundred and twenty seven employees from various organisations and positions participated in the study completing a Greek questionnaire of the five-factor model, and a measure of job satisfaction. Their immediate supervisors completed three questionnaires assessing participants’ overall job performance along with their participation in citizenship behaviours and their generic work competencies. Results indicated that well-established notions in personnel psychology (e.g. the uniform validity of conscientiousness or neuroticism and the relationship between personality or satisfaction with organisational citizenship behaviour) do not apply to employees working in Greek small and medium enterprises. Some of the most significant results of this study were the moderating effect of job type on the criterion-related validity of openness and agreeableness, the strong links identified between personality and job satisfaction, and the importance of citizenship behaviours for employees’ performance assessments. These results are discussed in terms of the FFM literature taking into consideration the strong effect of Greek culture. The limitations of the study are also discussed in detail, along with a series of theoretical and practical implications for research and practice using the five-factor model of personality.

Table of contents CHAPTER 1. INTRODUCTION ________________________________________ 1 1.1. 1.1.1 1.1.2 1.1.3 1.2. 1.3. 1.4.

VALIDITY ISSUES ________________________________________________ 2 PREDICTIVE CRITERION-RELATED VALIDITY __________________________ 3 CONCURRENT CRITERION-RELATED VALIDITY ________________________ 4 VALIDITY GENERALISATION AND META-ANALYSIS _____________________ 5 MODERATOR-MEDIATOR VARIABLE ISSUES ___________________________ 6 RESEARCH ON SELECTION AND ASSESSMENT IN GREECE _________________ 9 AIMS OF THE PRESENT STUDY _____________________________________ 10

CHAPTER 2. PERSONALITY ACCOUNTS _____________________________ 13 2.1. 2.2. 2.3. 2.3.1 2.3.2 2.3.3 2.3.4 2.3.5 2.3.6 2.3.7 2.3.8 2.4.

THE TRAIT APPROACH ___________________________________________ 13 THE FIVE-FACTOR MODEL _______________________________________ 16 PERSONALITY AND WORK PERFORMANCE ___________________________ 20 BARRICK AND MOUNT (1991)____________________________________ 21 TETT, JACKSON AND ROTHSTEIN (1991)____________________________ 23 ROBERTSON AND KINDER (1993) _________________________________ 26 SALGADO (1997) _____________________________________________ 27 HOUGH ET AL. (1990) __________________________________________ 30 COMMENTING ON THE META-ANALYTIC STUDIES OF PERSONALITY _______ 31 APPLYING THE FFM IN OCCUPATIONAL SETTINGS ____________________ 33 LEVEL OF ANALYSIS IN THE PERSONALITY- JOB PERFORMANCE RELATIONSHIP35 CONCLUSIONS__________________________________________________ 40

CHAPTER 3. JOB SATISFACTION____________________________________ 42 3.1. 3.2.

PERSONALITY AND JOB SATISFACTION ______________________________ 43 SATISFACTION AND JOB PERFORMANCE _____________________________ 45

CHAPTER 4. PERFORMANCE AT WORK _____________________________ 47

CHAPTER 5. ORGANISATIONAL CITIZENSHIP BEHAVIOUR __________ 50 5.1. 5.2. 5.2.1 5.2.2 5.2.3 5.3. 5.4.

THE DIMENSIONS OF OCB AND PROSOCIAL BEHAVIOURS. ______________ ANTECEDENTS AND CONSEQUENCES OF OCB_________________________ JOB SATISFACTION AND OCB ____________________________________ FAIRNESS AND OCB ___________________________________________ PERSONALITY AND OCB________________________________________ MEASURING OCB ______________________________________________ CONCLUSIONS__________________________________________________

52 54 54 56 60 63 66

III

CHAPTER 6. WORK COMPETENCIES ________________________________ 67 6.1. 6.2. 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 6.2.6 6.2.7 6.3. 6.4. 6.5. 6.6. 6.7. 6.8. 6.9.

DEFINING COMPETENCIES ________________________________________ COMPETENCIES CLASSIFICATION __________________________________ GENERIC OR ORGANISATION-SPECIFIC COMPETENCIES? ________________ COMPETENCIES AS INPUTS OR OUTPUTS? ___________________________ GENERAL OR SPECIFIC COMPETENCIES? ____________________________ INDIVIDUAL OR TEAM / ORGANISATIONAL COMPETENCIES? _____________ HARD OR SOFT (E.G. TRANSFORMATIONAL) COMPETENCIES? ____________ CORE OR PERIPHERAL COMPETENCIES?_____________________________ WHAT IS THE APPROPRIATE LEVEL OF ANALYSIS?_____________________ BOYATZIS’ MODEL OF EFFECTIVE PERFORMANCE _____________________ SCHRODER’S WORK _____________________________________________ THE MODEL OF THE MANAGEMENT CHARTER INITIATIVE (MCI) ________ OTHER GENERIC COMPETENCIES LISTS______________________________ ASSESSING COMPETENCIES _______________________________________ APPLICATIONS OF THE COMPETENCY APPROACH ______________________ CONCLUSIONS__________________________________________________

68 74 74 75 75 75 75 76 76 77 80 83 85 89 90 91

CHAPTER 7. PERSONALITY DISPOSITIONS AND PERSONALITY PROCESSES________________________________________________________ 92 7.1. 7.2. 7.3. 7.4.

THE COGNITIVE-AFFECTIVE PERSONALITY SYSTEM ____________________ 92 USING CAPS TO EXPLAIN LINKS BETWEEN PERSONALITY AND WORK RELATED VARIABLES ____________________________________________________ 97 MODERATING-MEDIATING RELATIONSHIPS AND CAPS ________________ 102 CONCLUSIONS_________________________________________________ 105

CHAPTER 8. RESEARCH QUESTIONS _______________________________ 106 8.1. 8.2. 8.3. 8.4.

DEMOGRAPHIC CHARACTERISTICS ________________________________ PERSONALITY DIMENSIONS PREDICTING WORK-RELATED OUTCOMES ____ RELATIONSHIPS BETWEEN THE WORK-RELATED VARIABLES ___________ SUMMARY OF HYPOTHESES: _____________________________________

106 107 111 112

CHAPTER 9. METHOD AND PILOT STUDY __________________________ 114 9.1. PROCEDURE __________________________________________________ 9.2. MEASURES ___________________________________________________ 9.2.1 THE PERSONALITY QUESTIONNAIRE ______________________________ 9.2.2 THE JOB SATISFACTION MEASURE ________________________________ 9.2.3 THE OCB MEASURE __________________________________________ 9.2.4 THE JOB PROFICIENCY INDEX ___________________________________ 9.2.5 THE GENERIC WORK COMPETENCIES QUESTIONNAIRE (GWCQ)________ 1st stage ___________________________________________________ 2nd stage ___________________________________________________

114 116 116 117 118 118 118 119 120

3rd stage ___________________________________________________ 4th stage ___________________________________________________ 5th stage ___________________________________________________ 6th stage ___________________________________________________ 7th stage ___________________________________________________

121 123 124 133 134

CHAPTER 10. RESULTS ____________________________________________ 145 10.1. SAMPLE DESCRIPTION __________________________________________ 10.2. RELIABILITY AND FACTOR STRUCTURE OF THE MEASURES _____________ 10.2.1 THE OCB MEASURE_________________________________________ THE COMPETENCIES MEASURE _________________________________ 10.2.2 10.2.3 THE JOB PROFICIENCY INDEX __________________________________ 10.3. PRELIMINARY ANALYSES ________________________________________ 10.3.1 GENDER-AGE DIFFERENCES ___________________________________ 10.4. EXAMINATION OF RESEARCH QUESTIONS ___________________________ HYPOTHESES 1, 3___________________________________________ 10.4.1 10.4.2 HYPOTHESIS 2 _____________________________________________ HYPOTHESES 4, 5___________________________________________ 10.4.3 10.4.4 HYPOTHESES 6, 7___________________________________________ 10.4.5 HYPOTHESIS 8 _____________________________________________ 10.4.6 HYPOTHESIS 9 _____________________________________________ 10.4.7 HYPOTHESES 10, 11_________________________________________ HYPOTHESIS 12 ____________________________________________ 10.4.8 10.4.9 HYPOTHESES 13-16 _________________________________________ 10.4.10 HYPOTHESIS 17 ____________________________________________ 10.4.11 HYPOTHESIS 18 ____________________________________________ 10.4.12 HYPOTHESIS 19 ____________________________________________ 10.4.13 HYPOTHESES 20, 21_________________________________________

145 148 148 151 152 153 156 159 162 163 166 168 169 170 171 172 173 176 176 176 177

CHAPTER 11. DISCUSSION _________________________________________ 181 11.1. 11.1.1 11.1.2 11.1.3 11.2. 11.2.1 11.2.2 11.3. 11.4.

PERSONALITY AND ORGANISATIONAL OUTCOMES ____________________ PERSONALITY AND JOB PERFORMANCE ___________________________ PERSONALITY AND ORGANISATIONAL CITIZENSHIP BEHAVIOUR ________ PERSONALITY AND WORK COMPETENCIES _________________________ ANTECEDENTS AND CONSEQUENCES OF JOB SATISFACTION _____________ PERSONALITY AND JOB SATISFACTION ___________________________ CONSEQUENCES OF JOB SATISFACTION ___________________________ THE VALIDITY OF THE PERFORMANCE MEASURES ____________________ LIMITATIONS OF THE STUDY – THEORETICAL AND PRACTICAL IMPLICATIONS ________________________________________________ 11.5. CONCLUSIONS _________________________________________________

181 181 186 189 192 192 196 198 200 204

REFERENCES _____________________________________________________ 206

Appendix 1 Sample items of the TPQue (Tsaousis, 1999; 1996) _______________ Appendix 2 The Job Satisfaction Scale (Warr, Cook & Wall, 1979)_____________ Appendix 3 Measure of Organisational Citizenship Behaviour (Smith et al., 1983) Appendix 4 The job proficiency index (Robertson and Gibbons, 1996) __________ Appendix 5 The questionnaire booklet used in the UK pilot study ______________ Appendix 6 The employeesâ&#x20AC;&#x2122; questionnaire booklet (cover letter and job satisfaction measure) of the study _______________________________________ Appendix 7 The supervisorsâ&#x20AC;&#x2122; questionnaire booklet (cover letter, OCB measure, job proficiency index and the competencies measure) of the study _______

222 223 224 225 226 230 232

List of tables Table 1. Types of validity (adapted from Robertson, 1996, p. 135) ________________ 2 Table 2. Potential situational moderator variables in the personality-job performance relationship (Schneider & Hough, 1995) _____________________________ 7 Table 3.The “Big-Five” Trait Factors and Illustrative Scales (adapted from Costa & McCrae, 1985)_________________________________________________ 17 Table 4.The five scales of NEO PI-R, their facets and brief definitions ___________ 19 Table 5. Meta-analytic results for personality dimensions and criteria (pooled across occupational groups) (adapted from Barrick & Mount, 1991, p. 15) _______ 22 Table 6. Meta-analytic results for all samples combined and for various subgroups __ 24 Table 7.Meta-analytic results for the “Big-Five” dimensions on findings reported in confirmatory studies ____________________________________________ 25 Table 8. Results from Robertson & Kinder’ s meta-analytic study (adapted from Robertson, 1993) _______________________________________________ 27 Table 9. Salgado' s meta-analytic results pooled across criteria and occupations ____ 28 Table 10. Salgado' s meta-analytic results across criteria ______________________ 29 Table 11. Meta-analytic results for criterion-related validity of integrity tests with applicant samples (adapted by Ones et al., 1994b) ___________________ 36 Table 12. True score correlations between composites of the FFM and composites of integrity tests (based on Ones et al., 1994b)_________________________ 37 Table 13. Mount and Barrick' s (1995) meta-analytic results for global-specific criteria______________________________________________________ 38 Table 14. A multidimensional approach to job performance ____________________ 48 Table 15. Smith et al.’s (1983) OCB measure _______________________________ 63 Table 16. Elements of competencies (adapted by Sparrow, 1997, p. 348)__________ 73 Table 17. Boyatzis’ managerial competencies (1982) _________________________ 78 Table 18. Summary of McBer’s results (1982) ______________________________ 79 Table 19. Schroder’s High-Performance Managerial Competencies (HPMC) (Schroder & Cockerill, 1990) ____________________________________________ 81 Table 20. Key roles and their associated units of competence for level I in management (adapted by BTEC & CNAA, 1990) ______________________________ 83

VII

Table 21. Key roles and their associated units of competence for level II in management (adapted by Torrington & Hall, 1995) _____________________________ 84 Table 22. The personal competence model (Management Charter Initiative, 1993) __ 85 Table 23. Dulewicz supra-competencies (adapted by Dulewicz, 1989)____________ 86 Table 24. Examples of competencies as part of assessment centres ______________ 87 Table 25. Woodruffeâ&#x20AC;&#x2122;s list of generic competencies (adapted by Wooduffe, 1993) __ 88 Table 26. Methods of measuring competencies (Smith & Robertson, 1992)________ 89 Table 27. The five cognitive-affective units in the personality mediating system (adapted by Mischel and Shoda, 1995, p. 253) ______________________ 94 Table 28. Examples of specific content in five categories of personality variables (adapted by McCrae & Costa, 1996) ______________________________ 96 Table 29. Comparisons of the validity check, lie scale and social desirability scores between the normative sample of TPQue and the present study.________ 117 Table 30. The five competencies clusters initially identified ___________________ 120 Table 31. Clusters and items of the competencies measure used in the UK pilot ___ 121 Table 32. The rating scale of the competencies measure ______________________ 123 Table 33. Means, standard deviations and inter-correlations of the competencies scale (UK pilot) __________________________________________________ 125 Table 34. First factor analysis of the UK pilot ______________________________ 126 Table 35. Pattern matrix of the UK pilot __________________________________ 126 Table 36. Inter-correlation matrix of the 5 factors of the UK pilot ______________ 127 Table 37. Definitions of clusters and items of the competencies measure used in the Greek pilot _________________________________________________ 132 Table 38. Means, standard deviations and inter-correlations of the competencies scale (Greek pilot) ________________________________________________ 134 Table 39. First factor analysis of the Greek pilot ____________________________ 135 Table 40. Pattern matrix of the first factor analysis (Greek pilot) _______________ 136 Table 41. Inter-correlation matrix of the 5 factors of the Greek pilot ____________ 137 Table 42. Hierarchical regression analysis of the competencies scales (Greek pilot) 138 Table 43. Second factor analysis of the competencies scale (Greek pilot) ________ 139 Table 44. Inter-correlation matrix of the 3 factors of the Greek pilot ____________ 140 Table 45. Final factor analysis of the competencies scale (Greek sample) ________ 141

VIII

Table 46. Inter-correlation matrix of the final factor analysis of the Greek pilot ___ 141 Table 47. Clusters and items of the competencies measure retained for the main study ______________________________________________________ 142 Table 48. Stepwise regression analysis of the three competencies scales _________ 142 Table 49. Means, standard deviations and inter-correlations of the competencies scales______________________________________________________ 143 Table 50. Job families of the sample _____________________________________ 146 Table 51. Job families and centrality of interpersonal interaction _______________ 148 Table 52. Factor analysis of the OCB measure (excluding item 6) ______________ 149 Table 53. Factor analysis of the competencies measure_______________________ 151 Table 54. Internal consistency (Cronbach’s α) of the study’s measures __________ 153 Table 55. Descriptive statistics of the study variables ________________________ 153 Table 56. Range restriction examination of the study’ s scales _________________ 155 Table 57. Gender-age differences________________________________________ 156 Table 58. Scales’ inter-correlation matrix _________________________________ 160 Table 59. Inter-correlation matrix between the 30 TPQue sub-scales and the performance measures ________________________________________ 161 Table 60. Regression analysis of the block of big-five to overall job performance__ 162 Table 61. Hierarchical regression analysis of the 30 sub-scales of big-five to overall job performance _____________________________________________ 163 Table 62. Big-five and performance measures for occupations involving interpersonal interactions _________________________________________________ 163 Table 63. Big-five and job satisfaction ___________________________________ 166 Table 64. Hierarchical regression analysis of the big-five dimensions on job satisfaction _________________________________________________ 167 Table 65. Regression analysis of the block of big-five to job satisfaction _________ 168 Table 66. The moderating effects of job satisfaction to the conscientiousnessneuroticism job performance relationship _________________________ 169 Table 67. The mediating effect of job satisfaction to the agreeableness-job performance relationship for occupations involving interpersonal interactions _______ 170 Table 68. The moderating effect of OCB-conscientiousness to the conscientiousnessjob performance relationship ___________________________________ 171

Table 69. The mediating effect of OCB-altruism to the agreeableness-job performance relationship for occupations involving interpersonal interactions _______ 172 Table 70. The contribution of job satisfaction to OCB-conscientiousness over conscientiousness ____________________________________________ 173 Table 71. The contribution of job satisfaction to OCB-altruism over agreeableness_ 173 Table 72. Personality predictors of action-leadership competencies _____________ 174 Table 73. The mediating effect of people-orientation competencies to the agreeablenessjob performance relationship for occupations involving interpersonal interactions _________________________________________________ 174 Table 74. The effect of agreeableness on overall job performance above and beyond the effect of competencies in occupations involving interpersonal interaction 175 Table 75. The effect of agreeableness on overall job performance above and beyond the effect of competencies and citizenship in occupations involving interpersonal interactions ______________________________________ 175 Table 76. The incremental effect of OCB to the satisfaction-performance relationship _________________________________________________ 177 Table 77. Inter-correlation matrix of performance scales _____________________ 177 Table 78. Hierarchical regression analysis of the citizenship and competencies measures (OCB entered first) ___________________________________ 178 Table 79. Hierarchical regression analysis of the citizenship and competencies measures (competencies scales entered first)_______________________________ 178 Table 80. Hierarchical regression analysis of the citizenship and competencies measures on overall job performance_____________________________________ 179 Table 81. Hierarchical regression analysis of the three competencies scales on overall job performance _____________________________________________ 180 Table 82. Regression analysis of the block of competencies to overall job performance ________________________________________________ 180

List of figures Figure 1. The stages of systematic selection (adapted by Smith & Robertson, 1993, p. 5). ___________________________________________________ 1 Figure 2. A model of moderating-mediating effects in the personality-job performance relationship __________________________________________________ 12 Figure 3. Boyatzis’ model of effective job performance _______________________ 69 Figure 4. McBer’s model of identifying managerial competencies _______________ 77 Figure 5. Integrated theory of HPMC and KAI (Schroder & Cockerill, 1990) ______ 82 Figure 6. Categories of personality variables as a metatheoretical framework for personality theories (adapted by McCrae & Costa, 1996, p. 73) _________ 95 Figure 7. The social-cognitive affective units and aspects of the personality-work related behaviour ____________________________________________________ 97 Figure 8. Personality, cognitive-affective units and work-related variables ________ 98 Figure 9. Participants’ gender___________________________________________ 145 Figure 10. Participants’ age ____________________________________________ 145 Figure 11. Scree-plot of the OCB (two factors requested) _____________________ 150 Figure 12. Scree plot of the competencies measure __________________________ 152 Figure 13. The moderating effect of job type on the openness-job performance relationship ________________________________________________ 165 Figure 14. The moderating effect of job type on the agreeableness-job performance relationship ________________________________________________ 166

Introduction

Chapter 1. Introduction Personnel selection and assessment remains a very important issue in organisational life. Companies not only invest millions in choosing the appropriate people for filling in their vacancies, but also spend valuable working hours of current employees acting as interviewers, or test administrators (Schmitt & Chan, 1998). Thus, the process of procedure eliciting the best candidates for the position.

personnel selection and assessment can and should be an elaborate and sophisticated

Nevertheless, this is not always the case. A lot of companies, especially small and medium size enterprises, still follow intuitive methods in their selection and assessment

AL LO

procedures based on the â&#x20AC;&#x153;expertiseâ&#x20AC;? of their personnel manager or make recruitment decisions based on less valid and reliable methods, such as a single traditional interview, resulting in inappropriate recruitment decisions. However, a well-respected number of companies both here in the UK and overseas, especially in the USA,

following, or at least attempt to follow, the guidelines of research, use reliable methods

and psychometrically sound instruments with very satisfactory results for their

workforce (Shackleton & Newell, 1997).

The following figure describes the stages of the traditional selection paradigm. The stages of systematic selection (adapted by Smith & Robertson, 1993, p. 5).

Figure 1.

Job analysis

Criterion development

Personnel specification

Dealing with candidates Choice of selection method Choice of candidates Establishing accuracy of selection Feedback

Introduction

The importance of each of those stages has been discussed with detail in a number of textbooks regarding personnel selection and assessment (e.g. Schmitt & Chan, 1998; Smith & Robertson, 1993). Here, in line with the present research, attention will be focused on the role of validity for personnel selection and assessment, and the importance of moderating-mediating effects affecting validity, which provided the insight for and are the main focus of this study.

The first issue has legitimately attracted much attention since the justification of the

validity of any selection and assessment method is, or should be, the first priority of an

organisation in the recruitment/assessment process. On the other hand, the

AL LO

moderating/mediating role of various work-related variables that influence the validity of specific recruitment and/or assessment tools has not attracted considerable attention. In the following sections, those two issues, which provided the starting points of the present study, will be discussed in more detail, along with the role of research on

1.1.

Validity issues

selection and assessment in Greece, where this study was conducted.

The requirement for a selection and assessment tool or method to demonstrate

acceptable validity is not simply essential but should be considered as the most important aspect of any personnel selection and assessment measure. Robertson (1996)

defined validity as “the extent to which a measurement procedure gives appropriate

information for the quality under investigation (p. 135)”. He discussed four types of

validity:

Table 1. Types of validity (adapted from Robertson, 1996, p. 135)

Face validity: the extent to which a measuring procedure seems “at face value” to assess the qualities in question. Not really a form of statistical validity at all. Content validity: the extent to which the items or procedures in a measuring instrument sample the domain of interest. Construct validity: the extent to which the measure gives accurate information about the psychological construct that it is designed to measure. Criterion-related validity: the extent to which the measure produces scores related to scores on a measure of some relevant criterion. In the case of personnel selection, the relevant criteria are usually indicators of job success, such as supervisors’ ratings of job performance.

In personnel selection and assessment the first and the last types are of most importance, the first because the measures have to be acceptable by both the candidates

Introduction

and the management, and the latter because the instrument has to be appropriate in selecting the best candidate for the criteria used, e.g. job performance, tenure, promotability, trainability, etc. Robertson (1996) claims that information for the instrument’s construct validity is also significant since it is very important for personnel specialists to understand the psychological construct assessed by the instrument. However, in selection and assessment, the criterion-related validity remains the most

important type of validity since it is concerned with “prediction” of the applicants’ behaviour (Smith & Robertson, 1993). There are two main types of criterion-related

AL LO

1.1.1 Predictive criterion-related validity

validity:

From a scientific point of view, this is the most desirable type of validity (Smith & Robertson, 1993), concerning the extent to which the measure predicts the candidates’ scores on some future assessment. With this type of validity, scores of the individuals

on a specific predictor are collected prior to any hiring or promoting decision. Those

scores should not be used for any decision. On the contrary, they are kept separately

from any other data gathered through the process, and are correlated with measures of specific criteria after a decision is made and the selected employees have settled down

at their posts, usually after two-three months. The resulting coefficient demonstrates the

criterion-related validity of the method for the specific sample. In its pure form the predictive criterion-related validity requires an organisation to

hire all applicants, even those who seem unsuitable, a requirement very few

organisations are in a position to fulfil. Subsequently, predictive studies of this kind

often suffer from restriction of range, the phenomenon occurring in a validation study,

when the rejected applicants are not included, and only the successful applicants for whom available data exist, both in predictor and criterion, are considered. A direct result of the restriction of range is a drop to the correlation coefficient, making it look smaller than it really is (Smith & Robertson, 1993). There are two types of restriction of range; the one just described is in the predictor measure. The second one concerns the criterion data, where data are not available for all the successful candidates either because they were promoted or transferred, or simply because they were poor performers and either voluntarily or not, became redundant.

Introduction

Another problem with predictive validity is the so called attenuation of the criteria or the unreliability of the criteria. If the criteria used have low reliability then the validity coefficient between the predictor and criterion will drop radically. Even a small drop in the reliability of the criterion can result in a very significant effect in the predictive validity of the predictor. Fortunately, statistical formulae have been

1.1.2 Concurrent criterion-related validity

developed and may be applied for correcting both types of the restriction of range.

The difficulties of the predictive criterion-related validity have led most researchers

and practitioners in adapting the concurrent validation approach. In essence, concurrent

AL LO

validity involves collecting the scores on the predictor and the criterion at the same time from a sample of people already working in the organisation. Then, correlating the two sets of scores the validity coefficient is derived. The advantage of the concurrent validation design is that it is a simple, straightforward, quick and cheap process,

compared to the predictive design. Nevertheless, the effects of the restriction of range

both in the predictor and criterion are in this case more severe, spuriously reducing the

validity of the method (Smith & Robertson, 1993). In addition, the fact that job incumbents are used in contrast to job applicants has a direct effect, since the two

groups will not show similar motivation or need to do well (Robertson, 1996).

Robertson (1996) argues that until the late 1970s the results of validation studies for various personnel selection methods were not very encouraging, as a result of those

methodological artefacts, leading personnel researchers to require local validation

studies as a check that the method adapted has acceptable validity. Unfortunately, this

type of approach had also a very important drawback; that most of these validation

studies were making use of rather small samples, often smaller than 100 people, resulting again in an underestimation of the validity of personnel selection methods. Subsequently, it looked like research in personnel selection and assessment had â&#x20AC;&#x153;hit the glass ceilingâ&#x20AC;?. However, all these changed in the early 1980s with the pioneering work of two American psychologists, Jack Hunter and Frank Schmidt in the field of validity generalisation or meta-analysis.

Introduction

1.1.3 Validity generalisation and meta-analysis Schmidt and Hunter developed procedures in the late 1970s to demonstrate that much of the variability in observed validity coefficients is due to various methodological artefacts, such as sampling error (i.e. small sample sizes) and differences across studies in the degree of range restriction and criterion unreliability (Schmitt & Chan, 1998). Subsequently, they developed meta-analytic methods to

aggregate studies of different predictors and criteria, concluding that most methods have nonzero validity for the prediction of important organisational outcomes and also that

AL LO

accounting for most of the variability in validity coefficients.

most of these methods generalise quite well across situations, with sampling error alone

Thus, meta-analysis is a set of statistical procedures for quantitatively cumulating the results of many independent validation studies. In such analyses, effect size data and data regarding study characteristics are coded from the original research studies (Schmitt & Chan, 1998). “They involve methods for estimating the amount of sampling

error in a set of studies and hence calculating a more accurate estimate of the likely upper and lower limits of the population validity coefficient in question” (Robertson,

1992, p.7). Meta-analytic procedures also allow for validity coefficients to be adjusted

for restriction of range and unreliability of measurement, and provide a basis for and job families.

estimating the extent to which validity coefficients may be generalised across settings

Summing up, meta-analysis can reduce the impact of other study artefacts such as

attenuation of study results produced by error of measurement in the dependent

•

(Hunter & Schmidt, 1989):

variable

•

attenuation of study results produced by error of measurement in the independent variable

•

attenuation of results produced by dichotomization of the dependent variable

•

attenuation of results produced by dichotomization of the independent variable

•

variation in results produced by range variation on the independent variable and attrition artefacts on the dependent variable.

Introduction

Hunter and Schmidt (1989) argued that since most psychological studies have used samples consisting of far less than 1500 participants, they are subject to massive chance variation in results. They say that it is this chance variation that produces the most significant contradictions between studies. For most researchers in the field of psychology, the statistical significance test was supposed to solve the problem of sampling error although statisticians have noted for years that the significance test

works only when the null hypothesis is true (Hunter &, 1989).

The consequences of meta-analysis were overwhelming for the field of personnel

selection and assessment. A number of selection methods, previously thought as having

AL LO

very low or even non-existent validities, such as the situational interviews or personality tests, demonstrated moderate to high criterion-related validities, whereas cognitive ability which was expected to predict moderately organisational outcomes, such as job performance or trainability, showed surprisingly high validity coefficients which was generalisable across occupations (Schmidt & Hunter, 1998). In addition, especially in

the case of cognitive ability, meta-analysis contradicted the hypothesis of single group validity, i.e. the hypothesis that cognitive ability tests are valid in predicting work

performance for majority workers only and not for minority employees (Hunter &

Schmidt, 1989). A number of meta-analytic studies, which will be discussed later in more detail, demonstrated moderate validity coefficients for specific personality traits,

Moderator-mediator variable issues

1.2.

justifying the use of personality tests in recruitment and assessment.

The examination of moderating and mediating effects in validation research has not

attracted the attention expected, considering the moderate validity coefficients of some

selection methods. Stone (1988), in a review article of moderatorsâ&#x20AC;&#x2122; research, claimed that their importance stems both from theoretical and practical considerations. At a theoretical level, moderators are implied in every theory, or model, arguing that a relationship between two variables may be depended on the value of a third moderating variable, whereas in a practical level, most empirical research requires examination of not only the main or simple effects of variables, but also the effect of their interaction in explaining variance of the dependent variable.

Introduction

Stone (1988) defined a moderator variable as a variable that interacts with another variable in predicting scores on and accounting for the variance in a criterion of interest, or a variable that systematically causes the relationship between two other variables to change. In the field of personnel selection, moderators’ research has had a major impact on the differential validity of psychometric tests examining whether variables such as race, gender, or age moderate the criterion-related validity of the instrument under subgroups under consideration (Aguinis & Romero-Stone, 1997).

question, or in other words if the instrument predicts performance equally well for the

As far as validation research in personality is concerned, Schneider and Hough

AL LO

(1995) suggested that it should concentrate more on the examination of non-linear relationships. They proposed a theoretical framework incorporating a number of personal, situational and criterion moderators, suggesting that, especially regarding the personality-job performance relationship, researchers should take under consideration the effect of those moderators in their attempt to explain possible effects of personality

described in the following table:

characteristics to occupational outcomes. A few examples of situational moderators are

Table 2. Potential situational moderator variables in the personality-job performance relationship (Schneider & Hough, 1995) Task consistency Amount of task structure Time on task Skill variety, task identity, task significance, autonomy, feedback Specificity Complexity Difficulty Rewards Ambient conditions Time of day Workplace layout Presence versus absence of physical handicap Level of physical attractiveness Role ambiguity Role overload Role conflict (table continued)

Task characteristics

• • • •

Goal characteristics

Physical environment characteristics Physical characteristics of individual Work role characteristics

• • • • • • • • • • • •

Introduction

• • • • • • • •

Social environment characteristics

Organisational characteristics Miscellaneous

Personality of co-workers (supervisors, peers, subordinates) Management style of supervisor Cohesiveness of work group Amount of social support Team work versus individual work Organisational values Organisational reward systems Organisational structure Physical illness

•

However, Schneider and Hough (1995) claim that not all moderators will moderate every personality-performance relationship. Some moderators are likely to exert more

AL LO

influence than others. They also argue that as more theories are developed and more moderator variables are simultaneously taken into account, “old validity barriers will go the way of the sound barrier” (Schneider & Hough, 1995, p. 120). Finally, they conclude that “it is up to the researchers to develop and test theories specifying

linkages among personality, situational, and job performance constructs if our field is

to advance” (Schneider & Hough, 1995, p. 120).

The examination of mediating effects in the field of personnel psychology is a more

complex issue. A variable is considered to be a mediator in organisational research when it accounts for the relation between the predictor and the criterion (Baron &

Kenny, 1986). In other words, if a predictor correlates significantly with a criterion, but

the researcher believes that this relationship is due to another third variable, and indeed

after statistical examination, when the effect of the third variable is taken into account, the correlation between the independent and the dependent variable drops to low, non-

significant levels, then the researcher is justified to claim that this third variable

mediates the relationship between the independent and the dependent variable. Subsequently, tests of mediation have to be theoretically driven; the researcher is required to specify causal models in order to explain how mediation processes occur (James & Brett, 1984). A number of variables have been examined in personnel psychology as potential mediators of various work-related variables. Robertson et al. (1991) found that candidates’ reactions to assessment procedures in a personnel selection setting mediate the link between post-assessment decision and employees’ attitudes and intentions, such

Introduction

as organisational commitment or job withdrawal and/or career withdrawal cognitions. Dekker et al. (1996) examined the mediating role of benefits offered by the organisation in the relationship between work force size and some aspects of job satisfaction, especially those which the organisation has the control. Singh et al. (1994) claimed that occupational burnout mediates the negative effects of role stressors on job outcomes for a sample of customer service representatives, whereas, Vecchio (1998) could not

identify mediating effects of leader-member exchange (LMX) scores in the relationship hypothesised role for LMX in the appraisal process.

Research on selection and assessment in Greece

AL LO

1.3.

between objective performance measures and supervisor ratings, failing to support a

The present study was almost entirely carried out in Greece. The main reason for that was that similar research interventions rarely occur in this country, which has a relatively short history of academic research in psychology, with the first independent

department established only in 1987, and where, even today, there is only one

permanent academic post in occupational psychology. Subsequently, the researcherâ&#x20AC;&#x2122;s

wish to assist to the development of the field in his home country was the major drive in deciding to investigate the present research questions in Greece, along with a desire to

investigate whether well-established relationships in the field of personality in

occupational settings, previously identified in American and West-European cultures, are repeated in a country like Greece, which only recently has started emerging as an

Union.

equal member of the other West-European countries and members of the European

Research on selection and assessment in Greece is quite limited. Papalexandris

(1992a, 1992b) provided the most recent discussion on the role of human resource management in Greece, claiming that Greek HRM is in a stage of rapid development and change, as a result of European Unionâ&#x20AC;&#x2122;s pressure. Nevertheless, these changes usually occur in multinational companies operating in Greece, rather than the small family run and owned enterprises which dominate the Greek economy (Papalexandris, 1992a). Kantas et al. (1997) investigated the recruitment tools used by managers in Greek small and medium enterprises, claiming that less reliable and valid measures, such as unstructured interviews, personal recommendations and examination of CVs,

Introduction

are still largely used. Similar results were obtained by Eleftheriou (1997) exploring managerial selection in Greece, including recruitment agencies as well, which were not included in the first study. Finally, Nikolaou (1995) also explored the application of personality and cognitive ability testing in a financial services organisation in Athens, Greece. These attempts are quite scarce however, highlighting the importance of additional

research in the field of personnel selection and assessment in Greece. The present study is an attempt to fill in this gap regarding the use of personality testing in selection and

assessment by exploring the criterion-related validity of a personality questionnaire

AL LO

based on the five-factor model of personality (Tsaousis, 1999) and its application in work settings in Greece.

1.4.

Aims of the present study

The initial idea of the present study came from a wish to examine in a more detailed

and theoretically driven way, the recent, moderately satisfactory results of meta-analytic

studies investigating the criterion related validity of personality tests. These meta-

analyses, along with the emergence of an, almost unanimously, accepted taxonomy of

personality characteristics such as the five-factor model, both of which will be examined later in more detail, restored up to a point the researchersâ&#x20AC;&#x2122; and practitionersâ&#x20AC;&#x2122;

confidence in the use of personality tests in personnel selection and appraisal.

Nevertheless, following the suggestions of researchers such as Schneider and Hough

(1995), it is suggested through out the present research that the role of personality in personnel psychology will be more easily understood and accepted if it is considered in

relation to and along with other work-related variables. Researchers in the field have

now understood, after two decades of arguments in the person-situation controversy, that their ongoing interaction and the dynamic processes involving both stable personality characteristics and situational or other variables can explain more about their potential effects compared to the examination of personality traits independently of situation and vice versa. Subsequently, variables such as job satisfaction, organisational citizenship behaviour and generic work competencies were thought to contribute to explaining some aspects

Introduction

of the personality-job performance relationship. The choice of those specific variables for examination in the present framework was theory-driven. As far as job satisfaction is concerned, very few studies have examined if specific personality traits are valid predictors of satisfaction at work. Nevertheless, research evidence from a number of studies, which will be discussed later, suggest that job satisfaction is related to two at least personality traits, positive and negative affectivity,

which in big-five terms are linked to extroversion and neuroticism respectively. In addition, if satisfaction is related to performance, irrespectively of which causes what,

and also if particular personality traits, as expressed by the five-factor model, are valid

AL LO

predictors of job performance, then an initial claim could be that a conscientious employee, for example, will perform better when s/he feels satisfied.

A similar relationship would be expected if extra-role activities are taken into account, such as organisational citizenship behaviour. In this case, previous research has

shown that both specific personality traits, such as agreeableness and conscientiousness,

and job satisfaction are valid predictors of citizenship behaviours. Subsequently, the

interaction of specific personality traits with job satisfaction is expected to increase supervisors’ ratings of employees’ contextual performance, whereas the interaction of

personality traits with citizenship behaviours is also expected to increase the effect of personality to job performance. Additionally, employees’ engagement in extra-role

supervisor ratings.

activities is expected to have a direct effect on employees’ job performance, as rated by

The role of generic work competencies was also examined. Competencies,

representing the ability factor in the present model, are directly related to superior work

performance, but are also expected to relate with specific personality dimensions, an interaction anticipated to result to superior work performance. These are a few examples of the ideas guiding our research, illustrated in the

following figure, which only intends to briefly describe the concepts reviewed in the present research, rather to describe a model of effective work behaviour.

Introduction

Figure 2.

A model of moderating-mediating effects in the personality-job performance relationship Organisational Citizenship Behaviour

Personality Five-factor model Conscientiousness Agreeableness Openness to experience Neuroticism Extroversion

AL LO

Job Performance

Generic work Competencies

• • • • •

Job satisfaction

The lack of a strong theoretical framework of personality in the personnel selection and appraisal research was making the present attempt more difficult. Subsequently, the

focus of attention was turned to the work of Mischel and Shoda (1995, 1998) and their

attempt to reconcile personality traits with dynamic processes. Their work was

considered very useful into describing and explaining the effect of personality to work settings, and the interaction of specific personality traits with the variables under

research, providing the necessary theoretical angle of the study, as will be explained in

the relevant chapter. In the following sections, the recent literature related to each of those variables is

summarised and the theoretical framework of Mischel and Shoda is presented and explained in terms of the present research and its use in personnel selection and appraisal.

Personality accounts

Chapter 2. Personality accounts The study of personality has always been an integral part of psychology. This should not be a surprise since one of the many definitions given to psychology is the â&#x20AC;&#x153;science of the mind, science of personalityâ&#x20AC;? (Oxford Thesaurus, 1994). The examination of personality processes was of major significance for all three major psychological

theories, ever since psychoanalysis was dominating the field, followed by behaviourism and more recently social-learning theory.

However, this was not always the case, as far as the application of personality at work settings is concerned. For years, especially during 1960s and 1970s, the role of

AL LO

individual differences at work, especially between academics, was not simply disputed but severely criticised, although personality scales were used quite extensively by I/O psychologists employed in industry. The main reason for this disregard could be the publication of two studies by Guion in 1965, where after thorough review of the

literature he came up with the conclusion that the use of personality measures in

employment decisions is probably inappropriate (Schneider & Hough, 1995).

However, this view started changing slowly in the early 1980s when a taxonomy first

identified in the 1960s, was described more accurately and examined in detail. This was the Five-Factor Model (FFM) of personality, otherwise known as Big-Five. However,

before discussing the FFM in detail, it is necessary to say a few things about the

2.1.

personality theory where Big-Five has its origins in; the trait approach to personality.

The trait approach

The trait approach is very much based on characterisations and attributions that

people make for others in their everyday communication. When people are asked to describe someone, we do not describe him/her using psychoanalytic concepts such as conscious, or id, ego etc. nor we use behavioural terms such as conditioning or process, for example. Usually we rely on trait terms. We can say that one person might be aggressive or calm, reserved or extrovert etc. One of the most prominent researchers in the field of personality, Hans Eysenck (1995) has defined traits as â&#x20AC;&#x153;relatively enduring descriptive characteristics of a person... predispositions to behaviour with cross-situational consistency, that is,

Personality accounts

leading to similar behaviour in a variety of situations” (p. 40-41). Thus, according to Eysenck, a trait is a relatively stable personality characteristic that makes the person respond consistently in various circumstances, in line with and as a result of this particular trait. Pervin (1993), similarly to Eysenck, defined trait as a broad predisposition to behave in a particular way. He commended that traits are the fundamental building blocks of

the human personality, where McCrae and Costa (1990) defined traits as dimensions of

individual differences in tendencies to show consistent patterns of thoughts, feelings and

actions.

AL LO

Gordon W. Allport was probably the first researcher in the personality field who introduced the concept of “trait”. He believed that traits are the basic units of personality and that they actually exist, based in the nervous system. He distinguished three types of traits: cardinal, central and secondary (Pervin, 1993).

A cardinal trait expresses a disposition very pervasive and outstanding in a person’s

life that almost every act is influenced by that trait. Generally, people have few such

traits. Central traits express dispositions that cover a more limited range of situations than cardinal traits. Finally, secondary traits represent dispositions that are the least

noticeable, generalised and consistent. According to Allport “a trait expresses what a

person generally does over many situations, not what will be done in any one situation”

(Pervin, 1993, p. 279).

Approximately at the same time with Allport, but certainly in another context, the work of Carl Jung took place. Jung was a Swiss psychiatrist and one time follower of

Freud (Eysenck & Eysenck, 1985), who continued the work of many predecessors such

as Hippocrates, Galen, Kant and Wundt on personality types. He saw the main cause of typological differences in the extroverted or introverted tendency of the libido, i.e. the tendency of the individual’s instinctual energies to be directed towards the outer world or toward his own inner mental states. The work of Jung on personality types led to the development by Myers and Briggs of a personality questionnaire called Myers-Briggs Type Indicator (MBTI) which is widely used in work psychology. However, this approach is not the only one to the field of personality types (e.g. Holland’ s hexagonal model, RIASEC).

Personality accounts

Returning to the discussion of the trait approach, Hans J. Eysenck’s work on personality traits has probably been the most influential in the field. Eysenck was born in Germany, but he came to England early to escape Nazi persecution. Based at the University of London he has worked both with normal and pathological populations. Eysenck was one of the hardest critics of psychoanalysis, since he was very strict in his standards for scientific pursuits, also placing great emphasis on conceptual clarity and

measurement. He emphasised the need to develop a personality theory that could be

tested and be open to disproof along with the importance of establishing biological

foundations for the existence of each trait (Eysenck & Eysenck, 1967). He argued that

there are three basic dimensions to personality and labelled them introversion-

AL LO

extroversion, neuroticism and psychoticism. Eysenck developed a series of questionnaires measuring these traits.

Another significant trait theorist was Raymond B. Cattell. He was also English but he did most of his work to the United States. Cattell started from the same point as

Eysenck concerning the existence of a number of basic traits in personality but he believed that the number of three traits only is not adequate. He argued that the majority

of personality variance may be explained by the existence of 16 factors-traits and he

developed a questionnaire called Sixteen Personality Factor (16 P.F.) Questionnaire (Cattell et al., 1994) to measure these traits.

The approaches of Eysenck and Cattell to trait theory of personality were dominating

the field for very long time. However, none of them should be considered “right” or

“wrong”, rather a narrow and broad approach respectively. Eysenck (1991) has argued

that consequent factor analyses of the Cattell’ s 16 primary factors lead to three superfactors closely resembled to his own three factor solution. Nevertheless, as is the case

for the FFM, which will be discussed next, there is a general consensus as far as two of

these dimensions is concerned; extroversion and neuroticism, which consistently are replicated across different approaches, studies, and measures adopted. The most recently developed approach to trait theory, which is also adopted in the present study, as already said, is the five-factor model or “Big-Five”.

Personality accounts

2.2.

The Five-Factor model “The five-factor model has its roots in the so-called lexical hypothesis of Galton that personality traits are captured in the words that people use to describe one another, and are therefore encoded in dictionaries” (Schneider & Hough, 1995, p. 77)

Galton, at the end of last century, catalogued the human personality traits by examining Roget’s Thesaurus (Schneider & Hough, 1995). Allport and his colleagues

continued Galton’s project developing a far more comprehensive list of about 18.000

such terms and dividing them into four alphabetical lists, the first of which included

approximately 4.500 terms that had been classified as stable traits (Goldberg, 1990). It

AL LO

was this particular list, adding and subtracting several hundred trait terms, that Cattell used as a starting point for his own research in lexical basis of traits, coming up with his sixteen factors.

However, the first researcher who derived a five-factor solution was Fiske, who

obtained self-ratings and ratings from psychologists and peers on 128 men, rated on 22

trait descriptors adapted from Cattell’s database. He obtained five factors in all three

cases (psychologists, peers, and selves) and labelled them Social Adaptability, Emotional Control, Conformity, The Inquiring Intellect and Confident Self-Expression

(Schneider & Hough, 1995).

The FFM, as it is known today, first emerged in a study that Tupes and Cristall

conducted in 1961 for US Air Force, using 35 trait descriptors, based again on Cattell’s

database. They factor analysed the data from 8 studies and reported that they found five

strong and distinct factors across seven of the eight studies. They called them Surgency,

Agreeableness, Dependability, Emotional Stability and Culture. Similar results were obtained from other researchers (e.g. Norman, 1963; Digman & Takemoto-Chock,

1981; cited in Schneider & Hough, 1995). Tupes and Cristall’s study, however, has been recognised as very significant since it was the first time that the five-factor model emerged the way we know it nowadays. But unfortunately, because of its military nature, this study was not widely known until 1992, when it was reprinted in the Journal of Personality. All the previous researchers have used all or some of Cattell’s trait descriptors resulting to the criticism that these studies are replicating rather than independently

Personality accounts

validating the five-factor model. Goldberg (1990) responded to this critique by using a new and unusual comprehensive set of trait terms, across studies, samples, factoranalytic procedures and ratings (i.e. self and peer). He emerged with the same five factors from this new list of variables. He called these factors Surgency, Agreeableness, Conscientiousness, Emotional Stability and Intellect. Probably the most comprehensive research on the five-factor model has been carried

out by McCrae and Costa (e.g. 1982, 1983, 1987, 1990, 1992, 1993, 1997). They have

labelled the five factors as follows: Neuroticism, Extroversion, Openness to Experience,

Agreeableness and Conscientiousness. In the following table, the definitions of these

AL LO

factors and the characteristics of the high and low scorer are presented. These definitions have been accepted and extensively used by most researchers in the personality field (e.g. Tett, Jackson & Rothstein, 1991; Barrick & Mount, 1991; etc.), although variations of this model also exist such as Hogan' s six dimensions (Hogan, 1986), and Hough' s "big-nine" (Hough, 1992). In addition, this categorisation is

followed in the present study, as well (Tsaousis, 1996, 1999).

Characteristics of the High Scorer

Table 3. The “Big-Five” Trait Factors and Illustrative Scales (adapted from Costa & McCrae, 1985) Trait Scales

Characteristics of the Low Scorer Calm, relaxed, unemotional, hardy, secure, self-satisfied

Sociable, active, talkative, personoriented, optimistic, funloving, affectionate

Assesses quantity and intensity of interpersonal interaction; activity level; need for stimulation; and capacity for joy.

Reserved, sober, unexuberant, aloof, taskoriented, retiring, quiet

Curious, broad interests, creative, original, imaginative, untraditional

OPENNESS TO EXPERIENCE Assesses proactive seeking and appreciation of experience for its own sake; toleration for and exploration of the unfamiliar. AGREEABLENESS

Conventional, down-toearth, narrow interests, unartistic, unanalytical

Soft-hearted, goodnatured, trusting, helpful, forgiving, gullible, straight-forward

Assesses the quality of one’s interpersonal orientation along a continuum from compassion to antagonism in thoughts, feelings and actions.

NEUROTICISM Assesses adjustment vs. emotional instability. Identifies individual prone to psychological distress, unrealistic ideas, excessive cravings or urges, and maladaptive coping responses. EXTROVERSION

Worrying, nervous, emotional, insecure, inadequate, hypochondriacal

Cynical, rude, suspicious, uncooperative, vengeful, ruthless, irritable, manipulative (table continued)

Personality accounts

Organised, reliable, hardworking, self-disciplined, punctual, scrupulous, neat, ambitious, persevering

CONSCIENTIOUSNESS Assesses the individualâ&#x20AC;&#x2122;s degree of organisation, persistence and motivation in goal-directed behaviour. Contrasts dependable, fastidious people with those who are lackadaisical and sloppy.

Aimless, unreliable, lazy, careless, lax, negligent, weak-willed, hedonistic

Mount and Barrick (1995) reported further evidence on the validity of the FFM by citing numerous studies that have investigated the convergent validity of the Big-Five,

i.e. the relationship between alternative theoretical perspectives and inventories and the FFM. They also argued about the universal application of the FFM, since studies across

AL LO

different languages and cultures (Japanese, German, Dutch, Filipino, French-speaking Canadians, Finnish, Polish, Australian, Israeli, Chinese and Greek) came up with similar five factors, with the exception, in a few cases, of openness to experience, which seems to be less replicable across cultures.

Costa and McCrae have developed a questionnaire specifically designed to measure

these five factors. Originally their questionnaire was called NEO-PI (Neuroticism,

Extroversion, Openness- Personality Inventory; Costa & McCrae, 1985), and was

measuring all five dimensions but only for three of them (i.e. N, E, O) their facetssubscales were also measured. Later (Costa & McCrae, 1992) they developed the NEO

PI-R specifically designed to measure all five factors along with their facets. In the

following table, the five factors are presented along with their facets and the definition

for each of them. The same structure is followed in the present study, since the personality questionnaire used here (Tsaousis, 1996, 1999) replicates the same five

factors as NEO PI-R along with its thirty subscales.

Personality accounts

Table 4. The five scales of NEO PI-R, their facets and brief definitions Neuroticism Facets It is related to emotions such as nervousness, tension, and worry. It has to do with how prone people are to be fearful of various situations. It is concerned with how aggressive people are and how bitter they can be when someone hurts them. It measures the tendency to experience feelings such as guilt, sadness, hopelessness and loneliness. It measures the degree of shyness in people and how often they experience feelings such embarrassment and shame. It refers to the ability of the individuals to control their impulses, desires, cravings and urges. It is related to the capability of individuals to deal with stress.

N1: Anxiety N2: Angry Hostility

N3: Depression N4: Self-consciousness

N5: Impulsiveness

AL LO

N6: Vulnerability Extroversion Facets E1: Warmth

This facet is related to issues of interpersonal intimacy. It measures how sociable people are and whether they like other peopleâ&#x20AC;&#x2122;s company. It is related to the confidence an individual feels about him/herself and whether he/she is afraid to take responsibilities in his/her life. It concerns how energetic people are and how much they favour participating in activities with rapid tempo. This facet deals with characteristics such as pleasure seeking, boldness, adventure, etc. The tendency to experience positive emotions such as pleasure, happiness, optimism, etc.

E2: Gregariousness E3: Assertiveness

E4: Activity

E6: Positive emotions

E5: Excitement-seeking

Openness to experience facets O1: Fantasy

O3: Feelings

O2: Aesthetics

The characteristics of active fantasy and vivid imagination form the core of this facet. This aspect has to do with the extent that people appreciate various kinds of art and beauty. The tendency of the individuals to experience feelings and emotions as an important part of their life. This facet touches on the subject of how open to new experiences people are. Whether, they like to try new activities, visit new places, taste strange foods, etc. This is related to intellectual curiosity, i.e. the tendency of the individual to consider new ideas or the willingness to accept unconventional ways of thinking. It is concerned with the promptness of the individuals to reconsider their own political, social, or religious ideas.

O4: Actions

O5: Ideas

O6: Values

Agreeableness Facets A1: Trust A2: Straightforwardness A3: Altruism

The tendency of the individuals either to trust or distrust the people they are dealing with. The characteristics of frankness, sincerity and forgiveness form the core of this facet. It measures whether the individual has a genuine concern for others and a willingness to assist anyone who needs help. (table continued)

Personality accounts

The characteristics of compliance, forgiveness and peacefulness are the basic features of this facet. It measures whether people measure themselves as humble and modest or not. This facet assesses how sympathetic and empathetic people are towards others.

A4: Compliance A5: Modesty A6: Tender-Mindedness Conscientiousness Facets

It deals with how effective and capable people are in encountering with needs of life. It assesses whether the individual likes to keep everything in order, and organise his/her life or not. It concerns with how consistent people are with their ethical principles and moral obligations. This scale deals with whether people are ambitious and striving for success, or lazy and lackadaisical. It assesses the ability of people to take on tasks and carry them through to completion despite boredom or other distractions. It determines whether the individuals have the tendency to think carefully before they proceed with an action or not.

C1: Competence C2: Order

C3: Dutifulness

AL LO

C4: Achievement Striving C5: Self-Discipline C6: Deliberation

“Big-Five” is now considered as a respectable and powerful framework in personality field. McCrae and John (1992) regard the five-factor model as a “turning

point for personality psychology”, where Digman (1990) cites an enthusiastic comment

from Hogan concerning “Big-Five”:

“This is an area in which personality psychologists have a major finding to report. We now can specify with some confidence the structure of vocabulary that observer use to describe [persons]-put another way, we have a replicable model of the structure of personality from the viewpoint of an observer...There is now considerable agreement that observers’ impressions can be expressed in terms of about five dimensions” (p. 436).

2.3.

Personality and work performance

The relationship between personality characteristics and actual performance at work,

has been of major importance both for researchers in the field of personality, providing evidence of the criterion-related validity of personality measures, and for practitioners, justifying the use of personality questionnaires in personnel selection and assessment. Two major developments brought new insight for research. These were the

emergence of the FFM since it could provide the necessary taxonomy to investigate

Personality accounts

relationships between specific personality dimensions and aspects of work performance, and the use of meta-analysis. The role of meta-analysis in personnel selection and assessment was dealt with detail in the introduction. As far as the use of personality testing in selection and assessment is concerned, a number of meta-analytic studies have investigated the validity of specific personality traits, using the Big-Five taxonomy, and their most significant results are

following.

2.3.1 Barrick and Mount (1991)

Barrick and Mount (1991) were the first to publish a meta-analysis on the

AL LO

relationship between the Big-Five personality traits and job performance. They examined the criterion-related validity of these dimensions to three job performance criteria (job proficiency, training proficiency, and personnel data) for five occupational groups (professionals, police, managers, sales and skilled/semi-skilled).

The researchers expected that conscientiousness and emotional stability (or

neuroticism), would be valid predictors of all job performance criteria for all jobs.

Conscientiousness because it assesses personal characteristics such as responsibility,

persistence and goal-directed behaviour, which are important attributes for accomplishing every-day work tasks in all jobs, where on the other hand, employees

exhibiting neurotic characteristics would tend to be less successful than more

emotionally stable individuals in all occupations studied because neuroticismâ&#x20AC;&#x2122; s

characteristics tend to inhibit rather than facilitate task accomplishment (Barrick &

Mount, 1991).

They also expected that Extroversion and Agreeableness would be valid predictors in

those occupations that involve frequent communication or co-operation with others, such as in sales and management. Furthermore, they believed that Openness to Experience would be a valid predictor of training proficiency because it assesses personality characteristics associated with positive attitudes towards new experiences,

such as learning experience. Barrick and Mount (1991) obtained sample-weighted mean correlations (corrected for unreliability) which were ranging from .04 for traits related to Openness to

Personality accounts

Experience to .22 for traits related to Conscientiousness. Table 5 summarises the results of their study.

Job proficiency Training proficiency Personnel data Mean (across occupations)

Total N

Number of r’s

Obs r

12,396 3,101 6,477

89 17 33

.06 .15 .06 .08

.10 .26 .11 .13

11,635 3,283 5,644

87 19 29

.04 .04 .05 .05

11,526 3,685 4,474

80 19 26

.04 .06 .08 .04

.06 .10 .14 .07

12,893 3,585 6,175

92 17 32

.13 .13 .11 .13

.23 .23 .20 .22

55 14 22

-.02 .14 .01 .03

-.03 .25 .01 .04

Job proficiency Training proficiency Personnel data Mean (across occupations)

Agreeableness Job proficiency Training proficiency Personnel data Mean (across occupations)

Job proficiency Training proficiency Personnel data Mean (across occupations)

Openness to experience

9,454 2,700 3,785

Job proficiency Training proficiency Personnel data Mean (across occupations)

Conscientiousness

.07 .07 .09 .08

Emotional stability

Criterion type Extroversion

AL LO

Table 5. Meta-analytic results for personality dimensions and criteria (pooled across occupational groups) (adapted from Barrick & Mount, 1991, p. 15)

Note: Obs r = uncorrected criterion-related validity ρ = corrected criterion-related validity

Generally, their hypotheses were confirmed although there were some variations.

Conscientiousness was a valid predictor for all job performance criteria across

occupational groups, but not Neuroticism, as Barrick and Mount expected. In addition – partially supporting their hypotheses - Extroversion only but not Agreeableness was a

valid predictor for two occupations, managers and salesmen. Finally, as predicted, Openness was found to be a valid predictor of training proficiency only but not for the other two criteria, job proficiency and personnel data. Recently Barrick and Mount attempted to replicate these results in a meta-analysis of service-oriented occupations involving either teamwork or dyadic interactions between employees and customers or clients (Mount et al., 1998). In line with the results of other meta-analyses (Organ & Ryan, 1995; Hough et al., 1990) and studies examining personality and service orientation (Hogan et al., 1984), they made the hypothesis that

Personality accounts

conscientiousness, agreeableness and emotional stability will be positively related to job performance, and also that agreeableness and emotional stability will be stronger predictors of performance in jobs involving teamwork than those involving dyadic service interaction. Their results showed that all of the FFM constructs had non-zero correlations with both ratings of overall job performance and with specific ratings of the quality of interaction with others, with stronger correlations for conscientiousness (.26),

agreeableness (.21) and emotional stability (.18), confirming their first hypothesis. Their second hypothesis was also confirmed regarding the moderating effect of the type of

interpersonal interaction in the relationship between agreeableness, emotional stability based on the FFM in predicting job performance. 2.3.2 Tett, Jackson and Rothstein (1991)

AL LO

and job performance, providing further evidences of the utility of personality constructs

Tett et al., (1991) raised four issues concerning previous review studies on the

criterion-related validity of personality measures.

1. First, the issue of distinguishing between correlations derived using an

exploratory versus a confirmatory approach to trait selection. It is expected that correlations, which are derived only from an empirical basis, will yield lower

correlations compared to studies that build their hypotheses on a sound

theoretical approach.

2. Second, some of the previous reviews included exploratory studies reporting only

significant findings, thus inflating the mean validity of personality measures.

3. Third, previous reviews were not contributing to their aggregation only one correlation, which even if it is averaged within the study, it still may be biased,

but more especially, and this could be a serious problem, if the specific sample is not representative of the population.

4. Finally, the fourth issue concerns the averaging of absolute values of validities, because, since both positive and negative correlations are expected due to the peculiarity of the personality construct, they would cancel each other out in the aggregation.

Personality accounts

Their research was in the same vein with Barrick and Mount’s (1991), though broader in its scope, since they were not examining only the relations between “Big Five” and job performance. They were also examining the effect of the type of research strategy followed i.e. confirmatory as opposed to exploratory research approaches, the use of job analysis to select personality traits versus no such use of job analysis, whether the study under research is published in a scientific journal or it is an unpublished

dissertation, etc. The main results of their meta-analysis are summarised in the following table.

Table 6. Meta-analytic results for all samples combined and for various subgroups N-wt’d Corrected z diff(4) mean r(2) mean r(3) All samples 13,521 97 .160 .237 Exploratory 4,467 51 .081 .121 Confirmatory 9,054 46 .198 5.97* .294 No job analysis 8,621 39 .196 .290 Job analysis 433 7 .253 1.68* .375 Professional 3,871 47 .163 .241 Non-professional 9,067 43 .160 .238 .09 Managerial 1,233 16 .163 .241 Non-managerial 11,284 75 .167 .248 .17 Recruits 4,853 12 .200 .296 Incumbents 8,542 83 .138 .205 2.46* Civilian 8,467 88 .133 .197 Military 5,054 9 .205 2.87* .304 Articles 10,153 61 .184 .273 Dissertations 3.368 36 .087 .128 4.85* (1) (2) (3) Number of studies sample weighted mean r mean r corrected for both predictor and criterion unreliability (4) Statistically significant at p < .05 K(1)

AL LO

Total N

Subgroup

The results of table 6 demonstrate that correlations were stronger 1) in confirmatory

vs. exploratory studies, 2) where job analysis was used to select predictors, 3) in samples using recruits vs. incumbents, 4) in military vs. civilian samples, and 5) in published articles vs. dissertations. Furthermore, Tett et al. (1991) did not find stronger correlations for samples using professionals (vs. non-professionals), managers (vs. nonmanagers), and also (not reported in table 6) for studies using subjective (vs. objective) performance criteria, and for studies carried out by researchers (vs. company studies). The researchers argued for the generalizability of personality scales validities across

Personality accounts

samples, since for all subgroups, the lower bound estimated values of the 95% confidence intervals were greater than zero (Tett et al., 1991). The corrected estimate of the overall relation between personality and job performance obtained in this study was .24. Tett and his colleagues argued that two were the most significant findings of their study. First, the mean validities derived from confirmatory studies were considerably greater than those derived from exploratory

studies, denoting the importance of careful and theory-driven matching between specific personality facets and relevant criteria; and second the necessity of carrying out job

analysis before identifying personality characteristics appropriate for a particular

AL LO

position. The importance of job analysis in personnel selection settings, for use with any selection method, is well established and naturally expected before telling the attributes that will be necessary for the successful applicant. However, in the present metaanalytic study, this finding demonstrated even further the value of confirmatory validation. In the following table, the results related to the studies based on

confirmatory approach for the dimensions of the “Big-Five” are presented.

Table 7. Meta-analytic results for the “Big-Five” dimensions on findings reported in confirmatory studies N-wt’d Corrected mean r(2) mean r(3) Neuroticism 900 10 -.150 -.223 Extroversion 2,302 15 .104 .155 Openness 1,304 10 .184 .272 Agreeableness 280 4 .220 .326 Conscientiousness 450 7 .121 .179 (1) (2) (3) Number of studies sample weighted mean r mean r corrected for both predictor and criterion unreliability Total N

K(1)

Dimension

The results of table 7 show that Tett at al. (1991), although they came up with

important results regarding the validity of the FFM, report outcomes which are quite different from Barrick and Mount’s (1991). The reasons for these discrepancies will be discussed later in the chapter. Tett et al. (1991) found that agreeableness and openness to experience had the highest effect sizes, where as you may recall, Barrick and Mount (1991) argued that conscientiousness is the most valid predictor across performance criteria and occupational groups. However, it should be noted once more, that the results presented in table 7 are only for the confirmatory based studies, whereas Barrick and Mount (1991) did not make this distinction. Finally, another significant result of Tett et

Personality accounts

al. (1991) is that tenure is a significant moderator in the validity of personality scales, with longer tenured employees displaying higher correlations between personality measures and supervisor ratings. 2.3.3 Robertson and Kinder (1993) In this meta-analytic study, the criterion-related validity of a particular series of personality questionnaires was examined; the Occupational Personality Questionnaire

(OPQ; Saville & Holdsworth, 1990). The OPQ is a 30-scale series of questionnaires (ipsative vs. normative, concept vs. factor model) which was specifically developed for

use at work settings. Using 20 validation studies conducted since 1984, the researchers

AL LO

explored the relationship between the OPQ personality scales and 12 specific criterion areas, following a â&#x20AC;&#x153;confirmatoryâ&#x20AC;? approach, in the sense that hypotheses concerning specific personality-job competence relationships were used to drive their meta-analysis (Robertson, 1993).

Initially, the researchers asked experienced users of the OPQ to hypothesise on

expected links between OPQ constructs and the set of twelve criteria they used (for

example, persuasiveness, planning and organising, business sense, etc.). Two metaanalytic procedures were followed. During the first one, only the personality constructs

that were expected to correlate with one or more of the twelve competences (i.e.

performance criteria) were used, and were then averaged within and across studies. The second procedure involved the use of raw data, from each of the twenty studies, to

create a composite variable of the personality variables judged a priori to be predictive

of the relevant criterion. The composite variable was created by bringing together the

unit weighted scores of the personality constructs hypothesised to associate with each

competence, and subsequently the predictor-criterion correlations that were computed were averaged first within study and then across studies (Robertson, 1993). Furthermore, Robertson and Kinder (1993) also investigated the incremental validity of the personality constructs after the effect of cognitive ability has been taken into account, a hypothesis which was consequently supported by their results. The main results of this study are demonstrated in table 8; they are in line with the results obtained by Tett et al., (1991) for studies using confirmatory research strategies.

Personality accounts

Table 8. Results from Robertson & Kinderâ&#x20AC;&#x2122; s meta-analytic study (adapted from Robertson, 1993) OPQ Personality scale

Job competencies (e.g. analysis, planning and organising, decision-making)

Composite personality variables based on a-priori hypotheses linking the variables with job competencies

Uncorrected (sample-size weighted) mean validity coefficient From .09 to .33

Examples

.20

AL LO

Resilience

.09

Untraditional Change-orientated Optimistic Relaxed Tough-minded Emotionally controlled Conceptual Critical Data-rational Forward-planning

Adaptability

Criterion

.32

Analysis

The results of this meta-analysis showed that, if relevant personality constructs are

matched with appropriate criteria, using a priori developed hypotheses, based on job analysis for instance, their criterion-related validity of personality tests may increase up

to a well-accepted level for use in applied settings. At this point, their results were similar with Tett et al. (1991). However, Robertson (1993) argued that his meta-analysis

with Kinder was different from Tett et al.â&#x20AC;&#x2122;s in the sense that they investigated specific

hypotheses by using meta-analysis to cumulate data from individual studies, where Tett

et al. investigated whether confirmatory research strategies produce better validities

than exploratory. Furthermore, commenting on Robertson and Kinder' s study, Salgado (1996) showed that because of a computational error, a part of their results was wrong.

He carried out a reanalysis of the data used by Robertson and Kinder (1993) and came up with more supportive results, as far as the criterion-related validity of OPQ is concerned, along with evidences on the generalizability of OPQ across the criteria used by Robertson and Kinder (1993). 2.3.4 Salgado (1997) This is the most recent meta-analysis examining relationships between personality and job performance using the FFM taxonomy. Starting from the point that the two

Personality accounts

major meta-analyses examining the criterion-related validity of Big-Five (i.e. Barrick & Mount, 1991, and Tett et al., 1991) included only studies conducted in United States and Canada, Salgado questioned the validity generalisation of these meta-analyses across countries and cultures (Salgado, 1997). He attempted to address two issues: first to explore the relations between personality factors as expressed by the FFM and work performance in Europe by carrying out a meta-analysis including validity studies not

covered in the previous meta-analyses and which were conducted in Europe. In

addition, he wanted to compare the validity generalisation results of the American

studies with Europe' s. From this point of view, Salgado' s meta-analysis is of major

AL LO

importance for the present study.

In line with the meta-analyses reported earlier, he expected that conscientiousness and neuroticism would be valid predictors of performance across jobs and criteria, whereas extroversion and agreeableness would be valid predictors for occupations involving interpersonal contact and openness would be a valid predictor of training

criteria.

The main results of Salgado' s meta-analysis are reported in table 9.

Table 9. Salgado' s meta-analytic results pooled across criteria and occupations

Personality dimension K N r rc 90% CV Ď Emotional Stability 32 3,877 .06 .13 .19 .10 Extroversion 30 3,806 .05 .08 .12 -.08 Openness 18 2,722 .04 .06 .09 .01 Agreeableness 26 3,466 .01 .01 .02 -.07 Conscientiousness 24 3,295 .10 .15 .25 .13 Note: K = number of correlations, r = mean observed validity, rc = corrected criterion-related validity, Ď = validity corrected also for construct validity, CV = credibility value.

His results, according to table 9, confirm his hypothesis that conscientiousness and

emotional stability generalise their validities across criteria and occupations, as the 90% credibility values show, and that the other three factors have low validities. However, these three factors were valid predictors of specific criteria, as Salgado had hypothesised, and it is shown in the following table:

Personality accounts

Table 10. Salgado' s meta-analytic results across criteria 90% CV r rc Ď Emotional Stability Rating 22 2,799 .08 .12 .18 .05 Training 6 470 .11 .18 .27 .27 Personnel 3 575 .07 .08 .12 .12 Extroversion Rating 22 2,799 .06 .09 .14 -.09 Training 4 383 .01 .02 .03 .03 Personnel 4 624 .07 .08 .12 -.03 Openness Rating 11 1,629 .00 .01 .02 -.06 Training 4 477 .11 .17 .26 .26 Personnel 3 616 .06 .07 .11 .11 Agreeableness Rating 19 2,574 .00 -.01 -.02 .04 Training 5 415 .12 .19 .31 .21 Personnel 2 495 .01 .01 .02 .02 Conscientiousness Rating 18 2,241 .10 .16 .26 .12 Training 3 324 .15 .24 .39 .39 Personnel 3 730 .05 .07 .11 .11 Note: K = number of correlations, r = mean observed validity, rc = corrected criterion-related validity, Ď = validity corrected also for construct validity, CV = credibility value. Criterion

AL LO

Thus, we may see from this table that indeed emotional stability and

conscientiousness generalise across criteria with validities ranging from .12 up to .27 for

emotional stability and .11 up to .39 for conscientiousness. Openness is a valid predictor

for training criteria, as Salgado expected, but unexpectedly for personnel data as well, although the studies contributing the personnel data are very few. Agreeableness,

although not hypothesised, is a valid predictor for training criteria with an Ď of .31,

whereas extroversion was not a valid predictor for any criterion. Salgado (1997) also carried out analyses across occupations, and his results

confirmed his hypotheses partially. Both emotional stability and conscientiousness were valid predictors across occupations, and extroversion for occupations with interpersonal requirements (such as managers, and police, although surprisingly not for salesmen), whereas agreeableness was a valid predictor for professionals, skilled labour, and managers. Salgado replicated these results in extensive version of this meta-analysis (Salgado, 1998) including army studies, as well.

Personality accounts

The results of this meta-analysis were in the same vein with Barrick and Mountâ&#x20AC;&#x2122; s (1991) meta-analytic findings. Conscientiousness showed the highest estimated true validity and was generalisable across jobs and criteria (Salgado, 1997). Furthermore, emotional stability was also generalisable across occupations and performance criteria, contrary to Barrick and Mount' s findings, but in agreement with a large study carried out by Hough et al. (1990), which will be discussed later. In addition, the validity of

openness as predictor of training criteria was again demonstrated, and the same applies

for the validity of extroversion and agreeableness regarding occupations involving

interpersonal contact. Summing up, as Salgado (1997) says, this study demonstrated

once more the validity of FFM as a predictor of job performance, especially as far as

AL LO

conscientiousness and neuroticism is concerned, for countries of the European Union, as well. 2.3.5 Hough et al. (1990)

This large-scale study examining the criterion-related validity of a number of

personality constructs is included here, although it is not a meta-analytic study, because

of the significant results it came up with. Hough et al., (1990) developed a questionnaire to measure 6 constructs that were very close to the five-factor model. These were

surgency, adjustment, agreeableness, dependability, intellectance, and affiliation. They administered their questionnaire in three contexts: a concurrent criterion-related validity

study, a faking experiment, and an applicant setting.

Their results showed that (a) validities were in .20s (uncorrected for unreliability or

restriction in range) against targeted criterion constructs, (b) respondents successfully

distorted their self-descriptions when instructed to do so, (c) response validity scales

were responsive to different types of distortion, (d) applicantsâ&#x20AC;&#x2122; responses did not reflect evidence of distortion, and (e) validities remained stable regardless of possible distortion by respondents in either unusually positive or negative directions. A worth-mentioning characteristic of Hough' s (1992) work is that in her research in the application of personality at work, she identified nine work-related personality constructs. Four of them (dependability, agreeableness, adjustment, and intellectance) were directly related to four of the big-five dimensions (conscientiousness, agreeableness, emotional stability, and openness, respectively). She also included

Personality accounts

achievement, as a separate dimension, a construct which in the traditional FFM as described by Costa and McCrae (1985), is part of conscientiousness, and in Hogan's sixdimension model is part of the ambition scale forming with sociability the respective to extroversion pole of his model. In addition, Hough' s taxonomy differs from Big Five by distinguishing affiliation and potency from extroversion, and also by including

masculinism and locus of control as independent work-related personality traits. 2.3.6 Commenting on the meta-analytic studies of personality

The most significant result of these meta-analyses is probably that “when a

theoretically driven research strategy is adopted there is considerable support for the

AL LO

existence of links between personality constructs and performance-related behaviour in organisations” (Robertson, 1993, p.193).

The five studies discussed earlier have their differences but similarities as well, and also their limitations. Barrick and Mount (1991) have limited their study in a few

occupations and criteria. On the contrary, Tett et al.’s (1991) meta-analysis was broader

linkages (Tett et al., 1994).

in its scope examining the overall validity of personality independently of particular

Ones et al. (1994a) argued that the most significant outcome of Barrick and Mount's meta-analysis was that conscientiousness is a valid predictor of job performance across

jobs and criteria, as did Salgado (1997) as well. On the other hand, Tett et al. (1991) in

their meta-analysis found that agreeableness was the construct with the highest validity.

Ones et al. (1994a) suggested that the differences in the results between these two metaanalytic studies concerning the effectiveness of the five-factor model were mainly due

to 3 reasons:

1. First, the comprehensiveness of the studies since the number of studies reported from Tett et al. (1991) and the sample sizes as well, were much smaller than those used in Barick and Mount (1991) for predicting job proficiency. Ones et al. (1994a) explained that by saying that Tett et al., (1991) restricted their “Big-Five” analysis only to those studies using a confirmatory strategy in the selection of the appropriate personality constructs. However, Ones et al., (1994a) argued that it is unnecessary to use a confirmatory approach in the selection of studies included in a meta-analysis when using it in combination with a personality taxonomy.

Personality accounts

2. Second, Tett et al. (1991) have not clearly demonstrated the procedures they followed to assign personality scales from specific inventories to personality constructs. Thus, the differences may be due to the fact that different personality scales were assigned to the “Big-Five” constructs from Tett et al. (1991) compared to Barrick and Mount (1991). 3. Finally, the third reason may be the existence of moderators. Barrick and Mount

(1991) summarised the validity of the “Big-Five” constructs across a number of

different occupations. On the other hand, Tett et al. (1991) reported validities across

job categories without summarising their results by job category. Furthermore, they

AL LO

also did not indicate the extent to which their results may be influenced from the existence of various moderators, that they found significant in their analysis, (e.g. incumbents vs. recruits; articles vs. dissertations; use of job analysis or not). In addition, Ones et al. (1994a) found a number of technical errors in the meta-

analysis of Tett et al. (1991) such as the use of absolute values and absolute

correlations, which produced a potential large upward bias in the estimated population

correlation. Also they argued that they used by mistake the median correlation instead of using the mean correlation based on the sampling distribution of the correlation.

Tett and his associates replied immediately to this critique (Tett et al., 1994). As far

as the comprehensiveness of the studies is concerned, they stressed the superiority of the confirmatory approach arguing that the significance of confirmatory findings can not be

substituted by using a personality taxonomy. As regards the assignment of the

personality scales to general categories, Tett et al. (1991) like Barrick and Mount

(1991), also used expert judgements but sought consensus rather than independent

ratings. However, they admitted that it was possible that scales were assigned to the “Big-Five” in slightly different ways in the two studies, but they also argued that this does not explain adequately why the mean validities reported by Tett et al. (1991) were higher than those reported by Barrick and Mount (1991). Finally, as far as the third argument is concerned, about the non-use of job families as moderators, Tett et al. (1994) argued that they preferred not to average validities across occupations, since the importance of personality, even across similar jobs, may vary significantly.

Personality accounts

In addition, Tett et al. (1994), reanalysed their original data, following the recommendations suggested by Ones et al.â&#x20AC;&#x2122;s (1994a) concerning the technical errors of their meta-analysis, and found slightly lower corrected mean validities for the personality dimensions. However, they were all in the same pattern, with only one exception, the comparison between studies based on job analysis or not; where although based, this difference was no longer statistically significant.

the job analysis based studies still showed higher validity than the non - job analysis

Mount and Barrick (1995) concluded that the previous meta-analyses suggest that

certain personality constructs seem to be valid predictors of job performance. However,

AL LO

with effect sizes on the range of .20 to .30 they questioned the usefulness of such predictors given the availability of other predictors with higher validities. Nevertheless, a suggestion like this should be considered in the light of the bandwidth-fidelity dilemma in personality research, which will be discussed later.

2.3.7 Applying the FFM in occupational settings

The emergence of FFM has provided human resources professionals with a very

useful taxonomy of personality dimensions, a number of which has proven to be useful in predicting various criteria at work, such as overall job performance, training, etc., as

the previous meta-analyses have shown.

Barrick and Mount have been two of the most "conscientious" researchers in the

research of applications of the FFM at work settings. Using as a starting point the

outcomes of their meta-analysis, they conducted a number of studies examining the

effect of autonomy and goal-setting between dimensions of the FFM and performance. In a series of studies they attempted to investigate the moderating role of a number of

important work-related variables on the relationship between personality dimensions as described by the FFM and job performance. They first studied the moderating role of autonomy (Barrick and Mount, 1993). Initially they found that, in agreement with their meta-analytic findings, both conscientiousness correlated significantly with supervisory ratings of job performance (r = .25, p < 0.01, corrected r = .35) and extroversion as well (r = .14, p < 0.05, corrected r = .20). As far as the moderating effects of autonomy is concerned, they found that managers with high scores on conscientiousness and extroversion performed better in jobs with high autonomy compared with those

Personality accounts

managers in jobs with low autonomy, whereas in the case of agreeableness they found, contrary to their expectations, that managers with low scores on this personality dimension performed better in jobs with high autonomy, compared with those managers in jobs low in autonomy. These results have very important implications for managerial positions, since it seems that in jobs where a great deal of autonomy is essential, managers who are organised, reliable, persevering (high conscientiousness), sociable,

active, talkative (high extroversion), and less co-operative, cynical, and suspicious (low agreeableness), have the necessary characteristics to perform effectively.

Further than that, they tested the relationship between conscientiousness -

AL LO

extroversion and job performance and the effect of two moderating variables (autonomous goal setting and goal commitment) in a sample of sales representatives (Barrick et al. 1993). They found that sales representatives high in conscientiousness are more likely to set goals and to be committed to these goals and that conscientiousness but surprisingly not extroversion was positively related to sales volume and supervisory

ratings of job performance. Using a similar sample of sales representatives Mount et al. (1994) examined the validity of observer ratings (supervisor, co-worker, and customer)

and self-ratings of personality measures. They found that observersâ&#x20AC;&#x2122; ratings of

conscientiousness and extroversion were valid predictors of performance ratings and also accounted for significant variance in the criterion measure beyond self-ratings

alone.

Schmit and Ryan (1993) however questioned the generaliseability of the FFM, since

in the majority of studies conducted using Big Five instruments, the samples consisted

of incumbents or students. Thus they conducted a very interesting study examining the stability of the 5-factor structure in a job applicant population - as part of a personnel

selection procedure. Their purpose was to examine whether the factor structure depends on the purpose of the test administration. On account of that, they also had a second sample of students. Their results suggested that the five-factor structure fit the studentsample data better than it fit the factor structure of data from the job applicant sample. The first factor of the applicant sample included loadings from four of the five dimensions of Big Five. Schmit and Ryan (1993) claimed that this factor should be called ideal-employee factor and it included items describing an applicant as productive, highly likeable, conscientious, courteous, thoughtful, considerate, organised, committed,

Personality accounts

active, high spirited, not helpless, and not argumentative. Nevertheless, although it seems that people can actually distort their replies to personality questionnaires if they like, the evidences are conclusive that this distortion does not attenuate the predictive validity of personality constructs (Barrick and Mount, 1996). In line with Schmit and Ryan' s study, Mount and Barrick (1995) described their thoughts for the existence of a functional personality at work, which may be captured by a conscientiousness-plus

factor, incorporating four of the five dimensions of the FFM, i.e. conscientiousness, agreeableness, emotional stability and extroversion. However, these ideas bring us to

the discussion of the bandwidth-fidelity dilemma in personality research.

AL LO

2.3.8 Level of analysis in the personality- job performance relationship Robertson (1993) argues that there is an issue concerning the use of Big Five as a basis for categorising personality constructs, as far as their application at work settings is concerned. He argued that especially for applications to the field of personnel

selection, personality constructs of greater specificity might be better predictors. As it

has been discussed earlier, the meta-analyses carried out by Barrick and Mount (1991),

Tett et al. (1991) and Salgado (1997) limited their analyses on a "higher-order" level, whereas Robertson and Kinder (1993) and Hough et al. (1990) made use in their studies

of narrower facets of the personality constructs. This distinction between broad vs. narrow personality dimensions is known in the field of personality psychology as the

"bandwidth-fidelity dilemma".

A number of arguments in favour of each side have been cited. A lot of researchers

have argued that the Big Five dimensions of personality have great bandwidth and too

much information is lost when data are aggregated to the level of the FFM. On the other

hand, other researchers and practitioners not only support the appropriateness of broad personality traits as better predictors of work performance, but also argue that the best predictor of job performance is a linear combination of a number of broadly defined traits.

This need for the appropriate level of analysis in personality research has attracted some attention during the last few years. The Journal of Organisational Behavior devoted in 1996 a special issue on the subject of bandwidth-fidelity. Ones and Viswesvaran (1996) took a very strong position in favour of the use of broad personality

Personality accounts

traits for use in selection and assessment. Initially, they claimed that broad personality traits have much higher reliabilities than their respective subscales, and as an example they brought the cases of 16PF and NEO PI-R, where the coefficient alpha reliabilities for the subscales are in the range of .40 to .70 and for the factor scales are all above .90. They also focused their attention on global measures of job performance since they argued that specific job components are highly inter-correlated and because of that,

broadly defined personality traits should be used for selection and classification purposes, since they are better predictors of overall job performance, rather than

narrower traits. However their stronger argument about the usefulness of broad personality traits in selection originated from a number of studies examining the

AL LO

criterion-related validity of integrity tests.

Integrity (or honesty) tests are classified into two categories (Sackett et al., 1989). Overt integrity tests, where the respondents are directly questioned about any dishonest, illegal activities they were involved in the past, and covert-personality based tests where dimensions,

such

trustworthiness, and sociability.

reliability,

conscientiousness,

adjustment,

personality

counterproductive behaviours are attempted to be predicted by composite measures of

Ones et al. (1993) conducted a very comprehensive meta-analysis of all the published integrity tests, employing various performance criteria, including supervisor

ratings of overall job performance, organisational production records, and various

table.

measures of counterproductivity. Their main results are summarised in the following

Table 11. Meta-analytic results for criterion-related validity of integrity tests with applicant samples (adapted by Ones et al., 1994b) Criterion

Predictor

Supervisor ratings of overall job Overt & Personality based performance tests Broad counter-productive behaviours Overt tests Detected theft Overt tests Broad counter Personality-based tests Note: Ď = true validity, K = number of correlations

.41

7,550

.39 .13 .29

5,598 2,434 93,092

10 7 62

Personality accounts

Ones at al.' s meta-analysis came up with very satisfactory results as regards the predictive validity of integrity tests with effect sizes ranging up to .41. Nevertheless, as far as the bandwidth-fidelity dilemma is concerned, the following two findings require more attention (Ones et al., 1994b). First, that overt integrity tests, which could be argued that they tap a narrow facet of conscientiousness (e.g. delinquency), predict better a broad criterion, such as counter-productive behaviours, rather than a narrow

criterion (e.g. theft), and second, that both overt and personality-based integrity tests

predict overall job performance much better than a narrower criterion such as

counterproductive behaviours, providing evidences on the breadth of the integrity

AL LO

construct.

Ones et al. (1994b) examined, using meta-analytic procedures, the relationship between the FFM and integrity, in an effort to examine the construct validity of integrity tests. Using linear composites of both the five dimensions of the FFM and integrity tests, they came up with the following very interesting results, presented in the

following table:

Extroversion

Openness to experience

Agreeableness

Conscientiousness

.59

-.18

.26

.78

.87

Emotional stability

Integrity tests

Table 12. True score correlations between composites of the FFM and composites of integrity tests (based on Ones et al., 1994b)

This table shows that a linear composite of conscientiousness has an impressive overlap with integrity tests, followed by agreeableness and emotional stability. Similar

results are found when composites are calculated separately for overt and personality-

based integrity tests. Summing up, the results of these studies show the significance of the use of integrity testing in personnel selection. Ones et al (1994b) argue that the addition of an integrity test in a personnel selection battery, along with a mainstream measure of the FFM may improve its predictive validity by more than 100%. The evidences above, according to Ones and her colleagues (1993, 1994b, 1996), provide more that enough support about the resolution of the bandwidth-fidelity dilemma in applications of personality testing at work, in favour of the use of broadly defined personality dimensions. Ones et al. (1994b) move their arguments even further,

Personality accounts

making a comparison between the history that led to the acceptability of the "g" factor as the best predictor of job performance, claiming about an analogous scenario in personality domain, as regards integrity. More supportive evidences about the use of broad personality traits in selection are provided by Mount and Barrick (1995). In a follow-up of their original meta-analysis (Barrick and Mount, 1991) they examined in more detail the validity of the

conscientiousness dimension of Big-Five along with two of its components,

achievement and dependability, across a number of specific criteria. Using as a starting

point their previous meta-analysis, but increasing the total sample by 50%

AL LO

approximately, they were expecting, along with the bandwidth-fidelity dilemma, that broad dimensions (i.e. conscientiousness) would be better predictors of broad criteria (e.g. overall job proficiency), whereas the best predictors of narrower criteria (e.g. effort, employee reliability, etc.) are expected to be narrower dimensions (i.e. achievement, dependability). Their results provided partial support for the bandwidth

approach in personality, as the following table is showing:

Table 13. Mount and Barrick' s (1995) meta-analytic results for global-specific criteria Achievement N Mean ρ 19301 .19 .33 46137 .22 .38

Dependability N Mean ρ 29455 .17 .30 50436 .22 .38

Conscientiousness Criterion N Mean ρ Global 35381 .18 .31 Specific 55526 .23 .40 Note: ρ = true validity

According to table 13, the use of a global personality dimension like

conscientiousness is, contrary to researchers' expectations, the best predictor of specific

criteria, with stronger effect sizes than the narrower dimensions of achievement and dependability, although this difference is quite small. Moreover, all three personality

dimensions are better predictors of performance, when specific rather than global criteria are used. However, the opposite side, in favour of the fidelity approach, has strong evidences as well, arguing about the importance of using a construct-oriented approach to match narrower personality traits to specific dimensions of work performance that have been found to be job relevant.

Personality accounts

A construct-oriented approach involves identifying a personality taxonomy, a job performance taxonomy, and a nomological link between them (Schneider and Hough, 1995). The researchers advocating the construct-oriented approach move along a similar line with the meta-analytic studies in personality research supporting the importance of a confirmatory approach in selecting relevant personality dimensions and performance criteria for each position. Thus, they argue that the choice of the appropriate level of

analysis in the personality domain and also the selection of the most relevant criteria

should be the results of either an empirically driven search of the literature or,

preferably, the result of a job analysis.

AL LO

Schneider et al. (1996) criticise studies that follow a broadside approach in personality research, i.e. studies that average correlations for all personality dimensions across all criteria (Schneider and Hough, 1995), arguing that this procedure is rather uninformative, leading to lower validity coefficients, even for personality dimensions that a priori someone would expect, at least in a conceptual basis, to relate to specific

criteria. Schneider et al. (1996) argued that both Mount and Barrick' s (1995) and Ones et al' s (1994b) meta-analyses fall into this category. Especially, as far as the latter study

is concerned, they argued that integrity is actually a constellation of narrow personality

traits rather than broad as Ones and her colleagues suggested. They cite a number of studies claiming that a number of the most prominent integrity tests along with service

orientation measures, which have also been argued that they have high criterion-related

validity and also consist of lower-order personality traits, actually were built by first

specifying narrow personality traits that were expected to relate to specific criteria, attributing thus their high validity to the use of facet-level scales specifically selected to

maximally predict the criteria of interest.

Moreover, Schneider et al. (1996) disputed the evidences provided by Ones and

Viswesvaran (1996) about the unidimensional concept of job performance and that is better predicted by broad personality traits in validation studies, insisting into a

hierarchical taxonomy of performance at work, where specific "performances" are predicted by narrowly defined traits, following thorough job analyses. Concluding, Schneider et al. (1996) take a very strong position in favour of the use of narrow traits in personality assessment, especially in selection and assessment settings.

Personality accounts

In a similar vein with them, Nyfield (1994) explored the benefits of a "differential" approach in the criterion-related validity of personality tests versus the use of Big Five. As a criterion measure she used a number of scales from the Inventory of Management Competencies, developed by SHL. In line with studies cited earlier supporting the fidelity approach, Nyfield (1994) showed that a constellation of narrowly defined personality traits, as represented by OPQ, are better predictors of specific performance similar results for both managerial-professional and lower level jobs.

criteria, than the Big Five dimensions derived by OPQ. Furthermore, she came up with

Paunonen (1998) in the most recent study on the issue, raised the question of the

AL LO

relative predictive superiority of the lower-narrower traits as opposed to higher-broader factors. He first showed that the subscales of the Jackson' s Personality Questionnaire were superior predictors than the broad five dimensions of the NEO-FFI, for the criteria used. Most interestingly, he found that, in an additive sense, the lower traits increased the predictiveness of the broader factors more than the factors increased the

predictiveness of the traits, in two independent studies using different domains of the lower level traits, although, in our view, questions should be raised as regards the

criteria selected for use in these studies.

Summing up the section of the appropriate level of analysis in personality research, we will cite Hogan and Roberts (1996) who argued that there is no real trade-off crisis

in personality research, rather a need for matching appropriate personality dimensions to

relevant criteria. The conclusion of all these studies is that the HR specialist, judging

individually in each case, must choose both the appropriate level of analysis if

personality instruments are going to be used as predictors, and the suitable performance

criteria as well.

2.4.

Conclusions

"The Big Five is firmly established as a useful and widely-accepted general

framework for the description of broad personality traits. All five factors appear to have predictive validity in occupational contexts, ... although some psychometric issues remain to be resolved" (Matthews, 1997 p. 489). This proposition by Matthews concluding the most recent review of the application of the FFM in occupational settings, sums up the usefulness of the FFM for I/O

Personality accounts

psychology. Big Five can be very useful if appropriately used. I/O psychologists know from their very first courses in personnel selection that job analysis guides their subsequent steps in selection and assessment. If the results of job analyses show that the incorporation of a personality measure is appropriate, then the FFM can provide them with a very useful taxonomy of personality traits. Big Five is not the answer to all the

AL LO

value for personnel selection have to be acknowledged.

problems that selection specialists faced in the past. Nevertheless, its importance and

Job satisfaction

Chapter 3. Job satisfaction Employees’ well-being and satisfaction is a major concern for every responsible employer, in addition to being probably the most frequently studied variable in the field of organisational behaviour. Job satisfaction concerns employees’ feelings about their jobs or about specific aspects of their jobs (Spector, 1997). It implies that a need is

fulfilled, a motive is achieved and a goal is met (Thierry, 1998). This definition of job satisfaction, one of numerous similar definitions, identifies the

two main approaches followed in the study of job satisfaction. The attitudinal and the cognitive approach. The attitudinal approach has traditionally been the most frequent

AL LO

method of measuring work satisfaction, although more recently, the focus of attention has been turned to the cognitive, underlying processes of work satisfaction. Another issue related to the definition of job satisfaction is whether it should be considered a global feeling or a constellation of attitudes about various aspects of the job.

Researchers employing the former approach usually assess job satisfaction with a single

item, assessing employees’ overall job satisfaction, or if using multiple items, by

averaging the scores of individual items in a total score. The facet approach, with the

use of multiple item questionnaires, assesses employees’ satisfaction/dissatisfaction in various aspects of the job, such as pay, promotion, working conditions, supervision, etc,

eliciting different scores for each of the various aspects of the job.

Thierry (1998) describes three main approaches regarding the use of job satisfaction

in work settings. Satisfaction as the result of a behavioural cycle, reflecting the employee’s evaluations of the outcomes produced in relation to his/her needs, motives,

values or goals that are important to the person. Satisfaction as a component of the

controlling and regulating system, where the evaluation of the results may cause the introduction of changes, either to improve or maintain the conditions leading to (dis)satisfaction. Finally, the third approach is satisfaction as a cause of behaviour, emphasising behaviours that result from (dis)satisfaction, such as absenteeism, turnover, or employee involvement/participation, contextual behaviours, etc. Thierry (1998) argues that these three approaches can mainly be categorised as motivation issues without designating any one particular motivation theory.

Job satisfaction

The relationship between specific motivation theories and job satisfaction is one of the many studied in the vast literature of employeesâ&#x20AC;&#x2122; well-being. The most significant of them include job characteristics, personality, occupational stress, person-environment and person-vocation fit, whereas as far as the effects of job satisfaction is concerned, its relationship with job performance, citizenship behaviour, absence, turnover, burnout, etc., has been investigated. Here, the relationship between job satisfaction and

personality dispositions and between job satisfaction and work performance will be

discussed. The relationship of satisfaction with citizenship behaviours will be discussed

Personality and job satisfaction

AL LO

3.1.

later in more detail.

The majority of the research in this field has concentrated on two personality dispositions, positive and negative affectivity. Negative affectivity is a disposition characterised by a tendency to experience aversive emotional states. People high in

negative affectivity tend to be distressed, agitated, pessimistic, and dissatisfied, whereas

people low in negative affectivity tend to feel more secure and satisfied with their

environments. On the other hand, positive affectivity is seen in high levels of energy, excitement and enthusiasm; individuals scoring high in positive affectivity are more

active and view their environments more positively, than low scorers (Warr, 1996).

Scores in those two measures are moderately (negatively) correlated with each other, around â&#x20AC;&#x201C;0.25 (Cropanzano et al., 1993), indicating that they are not the opposite poles

of the same concept. Hence, an individual may be high on both, low on both, or high on

one dimension and low on the other (George, 1992).

Warr (1996) suggested that since people high in negative affectivity tend to

experience negative feelings, then it should be expected that in work settings they will also feel similarly, leading to low job satisfaction, where the opposite will be true for employees scoring high in positive affectivity. He cites a series of studies examining the relationship between the traits of positive or negative affectivity and affective reactions at work. Subsequently, as George (1992) claimed, positive and negative affectivity are the dispositional underpinnings of job satisfaction. A number of researchers such as Arvey (Arvey et al., 1989) and Bouchard (1997) have also attempted to identify

genetical influences on job satisfaction. Based on studies of monozygotic twins reared

Job satisfaction

apart they have reached to the conclusion that approximately 30% of the observed variance in general satisfaction is due to genetic factors, suggesting that job satisfaction may be partly determined by enduring characteristics of people. In terms of the trait-factor analytic approach, few studies have examined the relationship between specific personality traits and satisfaction at work. The dispositions of positive and negative affectivity are related to the core of extroversion

(Diener, 1996; Tokar et al., 1998) and neuroticism respectively (Robertson and Callinnan, 1998). Cropanzano et al. (1993) found that lower negative affectivity and

higher positive affectivity predicted global satisfaction among 198 employees at a

AL LO

pathology laboratory, Necowitz and Roznowski (1994) also found that negative affectivity predicted some aspects of lower job satisfaction. Tokar and Subich (1997) found that the block of the Big Five personality dimensions contributed significantly to the prediction of job satisfaction, where higher extroversion and lower neuroticism were unique predictors to higher job satisfaction. Furnham and Zacherl (1986) in a study

examining the relationship between Eysenckâ&#x20AC;&#x2122;s personality traits and job satisfaction, found that extroversion correlated positively with overall job satisfaction and

satisfaction with pay, where neuroticism showed negative correlations with some overall satisfaction.

aspects of job satisfaction, but a statistically significant correlation was not found with

If indeed job satisfaction can be predicted by characteristics inherent to the person,

such as personality, then the potential to increase employeesâ&#x20AC;&#x2122; job satisfaction by altering

job characteristics and work design, as traditionally believed until recently, is limited

(Robertson and Callinan, 1998). Approaches, such as job enrichment and work rotation, have to take under consideration the influence of at least neuroticism and extroversion

in successfully improving employeesâ&#x20AC;&#x2122; satisfaction. In an attempt to understand the relationship between personality and job satisfaction,

and more specifically why negative affectivity is related to work satisfaction, Spector (1997) claimed that people high in neuroticism/negative affectivity experience high levels of all sorts of negative affect at work, including job satisfaction, perceiving negatively the job situation they are involved, or even because they make bad occupational choices resulting in job dissatisfaction. Another attempt to explain the

Job satisfaction

relationship between personality traits and job satisfaction was made in the field of “subjective well-being”. Diener (1996) argued that trait psychology has to learn a big lesson from subjective-well being, a concept, which although is very highly correlated to neuroticism and extroversion showing weak correlations with short-term situational variables, can not and should not be explained solely on the effect of heritable personality traits, or on situational effects exclusively. In line with his conclusions, job

satisfaction may be heavily influenced by individual personality characteristics, but under no circumstances should be considered that themselves are enough to explain

affective or cognitive reactions at work, independently of situational variables, such as

3.2.

AL LO

job characteristics.

Satisfaction and job performance

The relationship between satisfaction and job performance was examined very early in the research of job satisfaction, from the years of the Human Relations movement,

departing from the “common-sense” idea that satisfied employees will be more

productive at work, compared to less satisfied colleagues. This belief led to the

introduction of initiatives attempting to improve job characteristics (Hackman & Oldham, 1975) or increasing employees’ duties and introducing more variety in their

jobs, such as job enrichment and job enlargement.

Iaffaldano and Muchinsky (1985) in a meta-analysis of studies examining

relationships between job satisfaction and job performance (usually supervisor ratings)

found an average coefficient of .25, with stronger coefficients for intrinsic satisfaction (e.g. nature of the work, variety at work) than for extrinsic satisfaction (e.g. rate of pay,

supervision). Petty et al. (1984) reported similar findings for overall satisfaction but also

noted that the effect sizes were stronger for professional and managerial employees than for others (.31 and .15 respectively). These associations are quite common in studies examining links between satisfaction

and job performance, but nevertheless, do not indicate any direction of causality. It is not necessarily the case that a satisfied employee is a good performer. The opposite could be true as well; high performers might be more satisfied as a result of their effective work (Warr, 1996). Thierry (1998) also claims that it is not necessary to relate overall job satisfaction with a general index of job performance because it is more

Job satisfaction

useful to relate specific facets of job satisfaction with particular components of job performance. Another issue, which will be discussed later in more detail when discussing the organisational citizenship literature, is the notion that job satisfaction is more likely to lead to extra-role behaviours as compared to in-role job performance. Organ (1988a), who was the first researcher in the field of organisational citizenship, argued that studies

examining relationships between satisfaction and performance do not usually find

strong correlations because job performance is traditionally defined exclusively in in-

role terms, without taking into account extra-role performance. He argued that a

AL LO

satisfied employee will be keener to participate in extra-role activities rather than simply improving his/her in-role performance, following a cognitive process (Organ & Paine, in press) of the kind â&#x20AC;&#x153;I am satisfied thus I may participate in voluntary activitiesâ&#x20AC;?. Summing up, the role of job satisfaction in a model of personnel appraisal may be

crucial. If indeed satisfaction is partially determined by personality dispositions, such as

neuroticism or extroversion, then it should be expected to interact with personality

dispositions in predicting job performance, further than having a direct relationship with in-role or contextual performance. Thus, as will be mentioned later when the hypotheses

of the study are presented, a conscientious, or emotionally stable employee, for example, who is expected to perform well at work, would also be expected to receive

improved ratings of job performance if he/she feels satisfied from aspects of his/her

work. If this statement is true, the effects are self evident for employee selection and

assessment. Rather than spending massive amounts of money in training programmes of

doubtful effectiveness, human resource departments could improve their selection methods in order to reflect the influence of job satisfaction, for example by

incorporating measures of affectivity.

Performance at work

Chapter 4. Performance at work Employee’s performance has been one of the most important and widely used dependent variables in research and practice in organisational psychology. Especially in personnel selection and assessment, the measurement of job performance is an essential part of every approach. The various methods used to choose employees are based on the

assumption that the people selected will perform considerably better than their counterparts in the performance measures used to validate the selection method.

Job performance is often neglected from those doing research in selection and assessment simply defined as the variable to be predicted, the dependent variable

AL LO

(Schmitt & Chan, 1998) or the outcome/product of the employees’ efforts. An important issue related to our understanding of the concept of job performance is whether it is a single “general” factor or it comprises of a number of dimensions.

Campbell (1994) describes the unidimensional approach to job performance as the

“classical” approach to personnel research, since he claims that it has been used for the most part of the century and when people refer to the term they clearly imply the

existence of one general factor. Nevertheless, he argued that the unidimensional model

of job performance has been a major source of the criterion problem “because the search for reliable, uncontaminated, and objective indicators that significantly reflect

the general factor, or ultimate criterion, has generally been a failure” (Campbell, 1994,

p. 34-35). On the other hand, the unidimensional approach has recently received more

attention, since the application of meta-analytic methods has shown that a general factor underlies most common performance measures (Viswesvaran, 1996 cited in Arvey &

Murphy, 1998).

The alternative approach accepts that job performance is composed of a number of

clearly distinguishable components, such that some employees could perform well on one performance component, but relatively speaking, not as well on others (Campbell, 1994). In a longitudinal research carried out for the American army, the so called Project A, Campbell (1994) proposed a “theory of job performance” that hypothesises that job performance consists of eight major dimensions, each of which consists of several more-specific features. These dimensions are presented and briefly described in the following table:

Performance at work

Table 14. A multidimensional approach to job performance

4. Demonstrating effort 5. Personal discipline 6. Facilitating peer and team performance

7. Supervision/ leadership

3. Written and oral communication task proficiency

2. Non-specific task proficiency

It captures the core substantive or technical tasks central to a job. They are the job-specific performance behaviours distinguishing the content of one job from another. In many organisations individuals are required to perform tasks that are not specific to their particular job but are required by everyone in the organisation. Many jobs require the individuals to make formal oral or written presentations, and the proficiency to which one can write or speak are essential part of performance. In every job, the employees have to commit themselves to the performance of work tasks, exert a high degree of effort and persist in that effort. It is characterised by the degree to which negative behaviour, such as alcohol and substance abuse, law or rules infractions, and excessive absenteeism, is avoided. The degree to which the individual support his/her peers, helps them with job problems, and acts as trainer, if necessary. It encompasses all the behaviours directed at influencing the performance of subordinates through face to face interaction and influence. The distinction with the previous factor is peer leadership and supervisory leadership. It includes the major factors in management, that are distinct from direct supervision, including tasks that help to manage, report or define what an organisation does without direct interaction with subordinates, as would be the case for supervisory tasks.

AL LO

1. Job specific task proficiency

8. Management/ administration

Whatever is the case, if job performance is a general or a multidimensional construct, it is necessary to have appropriate and psychometrically sound measures of the relevant

constructs. In most occupations it is usually impossible to use objective measures of job

performance, thus leading researchers to adopt reputational measures, i.e. employeesâ&#x20AC;&#x2122;

performance is assessed by their immediate supervisor(s), which is the most common

approach, by peers, subordinates, or even using self-ratings.

Supervisory ratings have attracted most attention because they are easy to develop

and simply to use. In addition, supervisorsâ&#x20AC;&#x2122; assessments are usually considered as the most important source of information for employeeâ&#x20AC;&#x2122;s behaviour and performance at work in most organisations. Although they have often been criticised for a wide variety of errors, such as the similar-to-me effect, leniency, central tendency, and halo effect, the overall picture is not as bleak as might be expected (Campbell, 1994; Schmidt et al., 1992). Likewise, Harris et al. (1995) examined the psychometric properties of performance ratings collected for research versus administrative purposes and they

Performance at work

found that administrative-based ratings were more lenient, that ratings obtained for research purposes demonstrated significant correlations with a predictor, and that the two types of error correlated significantly (.58). Viswesvaran et al. (1996) carried out a meta-analysis of the interrater and intrarater reliabilities of job performance ratings. Their findings showed that supervisory ratings appear to have higher interrater reliability than peer ratings (.52 for supervisors, .42 for peers) and that intrarater

estimates of reliability (e.g. coefficient alphas) tend to be higher than interrater estimates. Finally, another meta-analysis (Harris & Schaubroeck, 1988) of the

relationship between peer-supervisor ratings, self-peer ratings and self-supervisor

ratings showed that while peer-supervisor ratings are moderately correlated (.62), self

AL LO

ratings were not highly correlated with either peer (.36) or supervisor ratings (.35). The nature of job performance at the end of the 20th century is expanding. Although it still remains a very important issue in personnel selection and assessment, the significant changes in the characteristics of most occupations and a move towards more

flexible definitions of work roles and jobs has resulted into viewing occupations as more dynamic and interchangeable, and performance in a more broad and adaptable

way, defined with less precision. Arvey and Murphy (1998) suggested that the new

definition of job performance at work settings should also incorporate the concepts of contextual performance and personal competencies. They argued that contextual

performance or citizenship behaviour is viewed as an important part of the new

performance concept, essential in most jobs of the contemporary, competitive work

environment, whereas personal competencies are required to perform various roles and tasks rather than a narrow view of specific tasks and duties inherent in fixed jobs and

work roles. These two issues, which are also essential parts of the present study, will

take up the following sections.

Organisational Citizenship Behaviour

Chapter 5. Organisational Citizenship Behaviour The term Organisational Citizenship Behaviour (OCB) first emerged in the literature of work and organisational psychology in the early 1980s, when Dennis Organ with his colleagues (Bateman & Organ, 1983; Smith et al., 1983) in an effort to explain the satisfaction-causes-performance relationship, suggested an alternative form of job

performance, they called citizenship behaviour. They argued that the low correlations usually found between job satisfaction and work performance, may lie in part in the craftsmanship (Bateman & Organ, 1983).

AL LO

Organ (1988a, p. 4) defined OCB as following:

meaning given to performance, usually defined as quantity of output or quality of

"OCB represents individual behaviour that is discretionary, not directly or explicitly recognised by the formal reward system, and that in the aggregate promotes the

effective functioning of the organisation.”

By discretionary, Organ describes the type of behaviour which is not enforceable by

the organisation or a requirement of the role or the job description, but “rather a matter of personal choice, such that its omission is not generally understood as punishable”

(Organ, 1988a, p. 4). In other words, OCB taps the actions the employees engage themselves to, which are not directly required by the management but are welcomed,

since they assist in the improvement of the image and running of the organisation.

Subsequently, in their first studies, Organ and his colleagues argued that low

correlations, if any, usually found between job satisfaction and performance are due to

the fact that OCBs are not taken into account when measuring job performance. Thus, they suggested that job satisfaction should lead to discretionary activities like OCB

rather than “hard” activities usually captured by job performance measures (Konovsky and Organ, 1996). A few examples of citizenship behaviours include, guiding a new colleague, exhibiting punctuality and attendance beyond expected norm, assisting peers and supervisor with their duties, making constructive suggestions for improvement, etc. Organ (1994b) argued that his first efforts in trying to explain the relationship between OCB and job satisfaction, which actually started with the publication of a paper

Organisational Citizenship Behaviour

in 1977 (Organ, 1977), were based on social-exchange theory and the rules of reciprocity and equity. “When people are satisfied, presumably they have received treatment they appreciate and would choose to return the favour. To reciprocate, they would choose those behaviours of which they are capable, that are not constrained by the situation, that are volitional, and that they have reason to believe their benefactors would

appreciate” (Organ, 1994b, p. 54).

Deluga (1994) argues about the role of social exchange and equity theories in work

settings and citizenship behaviours, by claiming that according to the former, the

AL LO

management provides its employees with financial rewards and support while in exchange the employees bring in their effort, and expertise, whereas according to the latter employees and supervisors feel more satisfied and are more productive when the ratio between the benefits received and the contributions made is similar to the ratio of

colleagues within or outside the firm. Subsequently, as it will be discussed later in more

detail, OCB may be attributed as a function of fairness (Organ, 1988b).

A similar to OCB concept appeared in literature by Brief and Motowidlo (1986) called prosocial organisational behaviour. Their definition is very close to Organ’ s

definition of OCB, describing prosocial organisational behaviour as the behaviour

performed by a member of an organisation towards another member, or group of the same organisation, with the intention of promoting the welfare of the target towards

where the action is directed. They considered prosocial behaviour as a concept closely

related not only to organisational citizenship but also to organisational commitment and

whistle-blowing which is the act of revealing improper organisational practice to

someone able to correct it, usually inside the organisation. Based on these definitions, it is obvious that OCB and prosocial behaviour in general

can be a very important aspect of employee’s performance. Although not specifically required and described by job descriptions, OCBs are very significant for organisational and individual progress. The next section describes the various dimensions of citizenship behaviours.

Organisational Citizenship Behaviour

5.1.

The Dimensions of OCB and prosocial behaviours.

Most of the studies examining the structure of OCB agree that it is a multidimensional concept bearing at least two dimensions, a “personal” and an “impersonal” construct. A few studies (e.g. Bateman & Organ, 1983) used an unidimensional construct, other studies used a four (e.g. MacKenzie et al., 1991) or a five (e.g. Tansky, 1993) dimensions structure but all of these researchers agreed at least

about two of these dimensions. These are altruism and conscientiousness (or generalised compliance).

Altruism taps the personal form of OCB. It refers to behaviour, which is directed

AL LO

towards an individual in face-to-face situations aiming at helping him/her (e.g. assisting someone with a heavy overload, or orienting new colleagues at work). Organ (1988a) argues that altruistic behaviours include helpful actions rendered spontaneously on the spot directed not only towards colleagues, although this is the most frequent use, but also towards outsiders (e.g. customers, clients, suppliers) as long as these actions have

organisational relevance.

The impersonal form of OCB called either conscientiousness or generalised compliance is indirectly helpful to other people within the firm (peers, supervisors, or

subordinates), assisting to the better running of the organisation in general, such as being punctual, giving advanced notice if unable to attend, etc. Mostly it includes

instances where the employee carries out certain role behaviours well beyond the

minimum level required from the position (Organ, 1988a).

Apart from the obvious distinction between altruism and conscientiousness, the

personal and the impersonal dimension, “one can intuitively appreciate that the people

whom we think of as most conscientious are not always the most altruistic and vice versa; … altruism and conscientiousness, while overlapping to a modest extent, are not nearly so related as to blur into a single category” (Organ, 1988a, p. 37). The rest of the dimensions appearing in a number of OCB measures are: sportsmanship which emphasises restraint and forbearance including behaviours such as focusing on what is right with the firm and avoiding complaining, courtesy which includes keeping co-workers informed or taking active steps to avoid potential problems and civic virtue including behaviours such as participation in meetings and training, and

Organisational Citizenship Behaviour

making active efforts to keep informed on matters that may be of benefit to the organisation (Parks & Kidder, 1991). Brief and Motowidlo (1986) cited 13 kinds of prosocial organisational behaviours, where most of them are more specific items of the OCBs already described. These are the following: 1. assisting co-workers with job-related matters

2. assisting co-workers with personal matters 3. showing leniency in personnel decisions

4. providing services or products to consumers in organisationally consistent ways

AL LO

5. providing services or products to consumers in organisationally inconsistent ways 6. helping consumers with personal matters unrelated to organisational services or products 7. complying with organisational values, policies, and regulations.

8. suggesting procedural, administrative, or organisational improvements

9. objecting to improper directives, procedures, or policies

10. putting forth extra effort on the job

11. volunteering for additional assignments

12. staying with the organisational despite temporary hardships 13. representing the organisation favourably to outsiders

A number of these behaviours are associated with direct functional consequences for

the organisation, others, if used inappropriately, may prove dysfunctional for the

organisation (such as showing extreme leniency in hiring, promoting or compensating

employees), and others are directed towards colleagues or firmâ&#x20AC;&#x2122;s customers or are directly related to organisational issues. However, all of them, when used in the

appropriate way, may improve the organisational effectiveness either in the long or the short term. Finally, Van Dyne et al. (1994) incorporating in the study of OCB the modern

political theory and classical philosophy, studied mediating effects on three forms of citizenship: organisational obedience which mirrors behaviours related to the importance of following and accepting organisational norms and rules, organisational loyalty involving actions which the employees take in order to show their devotion to their organisation and its leaders and last, organisational participation which is showing

Organisational Citizenship Behaviour

interest in various issues related to the firm such as keeping informed and participating in non-required meetings.

5.2.

Antecedents and consequences of OCB

As itâ&#x20AC;&#x2122;s already been said, OCB first emerged in an effort to explain the anticipated but not clearly supported relationship between job satisfaction and work performance.

Indeed as Organ (1994b) suggested, most of the studies which examined the relationship between OCB and job satisfaction, came up with moderate correlations

between the latter and the two main forms of OCB; altruism and conscientiousness. This

AL LO

relationship is usually stronger for altruism. 5.2.1 Job satisfaction and OCB

Bateman & Organ (1983), carried out the first published study examining the relationship between OCB and various forms of job satisfaction (satisfaction with work,

pay, promotions, co-workers and supervision) as measured by the Job Descriptive Index

(Smith et al., 1969). They came up with a very encouraging effect size of .41 (p< .01)

between job satisfaction and an overall score of OCB. However, a path analysis failed to identify a causal direction between citizenship and satisfaction.

Taking this study even further, Smith and her colleagues (1983) attempted to examine the determinants of OCB using causal modelling. As far as its relationship with

job satisfaction is concerned, they found a direct relationship with altruism only

(r = .27, p<005) but not with generalised compliance. Subsequently, they argued, it

seems that job satisfaction may lead only to altruistic behaviours.

In a similar effort, Organ and Konovsky (1989) attempted to identify whether the

elements exerting influence on OCB have a cognitive or an affective basis. As regards work satisfaction, they found that positive affect, a concept closely related to satisfaction, as discussed in the job satisfaction section, has moderate but significant correlations both with altruism and compliance but they didnâ&#x20AC;&#x2122;t find any relationship between them and negative affect. The explanation they offered, as will be discussed later, was built on the concepts of fairness and justice. They also came up with the interesting result that pay cognitions (i.e. how good is their pay comparing to specified

Organisational Citizenship Behaviour

referents) was the most significant predictor of both forms of OCB, underlying the importance of the cognitive basis of OCB. Similarly, Williams and Anderson (1991) illustrated the importance of cognitive components of job satisfaction. One significant finding was that extrinsic job satisfaction cognitions had a stronger effect size with OCB performance than the pay cognitions that Organ and Konovsky (1989) found in their study. But more important,

they found that the two aspects of the job cognitive components (i.e. extrinsic and intrinsic), evaluated by the employees, were differentially related to the two OCB

forms, assessed by their supervisors, measuring OCBs targeting an individual (OCBI)

AL LO

and OCBs targeting the organisation (OCBO). Thus the extrinsic component of job cognitions was found to predict OCBO and the intrinsic component OCBI, but unfortunately they did not offer an adequate explanation for this result but only advised replication of the study.

Van Dyne et al. (1994), as already discussed, used a political philosophy framework

in their effort to study OCB. They also examined the relationship between various forms

of OCB and job satisfaction and they came up with statistically significant correlations between them ranging from .15 up to .67 for the six forms of OCBs they were

measuring. Similarly, Morrison (1994) came up with correlations on the range of .14 up to .23 for five forms of OCBs in a study, which challenged the boundaries between in-

role and extra-role behaviours.

In vein with all major topics in the field of work psychology, Organ and Ryan (1995)

conducted a meta-analysis examining a number of attitudinal (e.g. job satisfaction,

fairness, organisational commitment, leader supportiveness) and dispositional

(agreeableness, conscientiousness, positive-negative affectivity) predictors of OCB. As regards job satisfaction, they came up with moderate uncorrected mean correlations of .24 between altruism and satisfaction and .22 between generalised compliance and satisfaction, which were lower (.23 and .20 respectively) when self-ratings of OCB were not included. However Organ and Ryan (1995) argued that these effect sizes, although moderate, are still larger than the respective correlations usually found in meta-analyses examining job satisfaction and in-role performance (e.g. Iaffaldano & Muchinsky,

Organisational Citizenship Behaviour

1985), giving some support to the hypothesis that OCB measures are more related to satisfaction comparing to in-role measures. In line with the results of this meta-analysis, Organ and Lingl (1995) examined the role of personality, and job satisfaction in OCB. They found that, as far as job satisfaction is concerned, overall job satisfaction accounted for an important percentage of the explained variance in altruism, but not for conscientiousness as well, where only

co-worker satisfaction contributed significantly.

Concluding the section examining the relationship between OCB and job

satisfaction, it seems that OCB lies between job performance and job satisfaction,

AL LO

occupying the common space between them (Organ, 1988a). Although their causal relationship still remains an unresolved topic in work psychology, citizenship behaviours may partially explain the moderate correlations usually found between them. According to Spector (1997), the small size effects found between satisfaction and

performance may partially exist due to the extensive use of supervisor ratings, who, has

to be said, in a lot of cases face major difficulty to differentiate between in-role and

extra-role performance. On the other hand, it seems that this relationship is getting stronger when improved performance leads to rewards (e.g. Jacobs & Solomon, 1977).

Subsequently, it could be argued that satisfaction might lead to improved performance guided by a cognitive process, rather as a mood state. An employee who is satisfied

from his supervisor, for example, will be keener to assist him with his duties or

participate in activities beyond his position if he is satisfied with his salary. In other

words, it’s a process of fairness that takes place within employees’ frame of minds in

order to decide whether they will participate in extra-role activities.

“In a nutshell, the most plausible interpretation of the job-satisfaction OCB

correlations is that OCB varies positively with the extent to which a person believes that fairness has obtained in his or her relationship with the organisation” (Organ, 1988a, p. 61).

5.2.2 Fairness and OCB The study of OCB as a function of fairness goes back to equity theory, an approach to work motivation, where the employee compares the ratio of his inputs and outputs

Organisational Citizenship Behaviour

with a referent’ s ratio. If inequity exists, the employee will take action in order to restore the feeling of equity, either by distorting the outcomes/inputs or changing the comparison other. There are two main forms of organisational justice that are studied with regard to OCB. These are distributive justice, which concerns with the allocation of resources and outcomes and procedural justice which has to do with the methods and procedures

followed for this allocation. Organ (1988a) argued that distributive justice is far more important with respect to people’s employment relationships than procedural justice.

Thus, as regards OCB and fairness, Organ (1988b) suggested that an employee who

AL LO

feels as being treated unfairly is highly unlikely to diminish his performance in terms of explicit job requirements, since this attitude will almost certainly lead to increased levels of injustice, because of the loss of potential rewards. Subsequently, he argues that it is easier and less painful for the employee to consciously withdraw from discretionary

activities. However, Organ does not suggest that simply by withholding citizenship

behaviours, the person will feel again equal to his referent, but rather that the

relationship between the individual and the organisation is redefined. According to social exchange theory, the person will not reciprocate to his firm and the exchange

between them will be limited to the requirements directed only by the contract.

In his study with Konovsky (1989), Organ came up with the very interesting result of finding stronger relationships between job and pay cognitions and the OCB dimensions

of altruism and compliance. The effect sizes of the two types of cognitions were

stronger than positive and negative affect, whereas the results of hierarchical regression

analysis showed pay cognitions to be the best predictor of both OCB dimensions,

explaining unique variance beyond the contribution of any other predictor. According to the researchers “… OCB has a deliberate, controlled character, somewhat akin to conscious decision making rather than expressive emotional behaviour” (Organ & Konovsky, 1989, p. 157). Using social exchange theory, they attempted to offer a fairness interpretation of OCB. They suggested that as long as the individual trusts his/her firm and this feeling of fairness remains stable in a long-term basis, then the employee does not need to worry about the recompense of the OCB gestures. But if this trust is violated, and the employee perceives his relationship with his/her firm as unfair,

Organisational Citizenship Behaviour

then s/he will withdraw from discretionary activities, redefining his/her relationship with the firm in terms of a more strictly defined economic exchange, offering only the contractually enforceable duties. Schnake (1991) believes at this point, although without providing evidences, that fairness may be more strongly related to generalised compliance, while affect, i.e. job satisfaction should be more related to altruism. Examining the last hypothesis, between others, Tansky (1993) found that both

altruism and conscientiousness were positively correlated to a measure of overall

fairness, but a regression analysis showed, in contrary with Schnake’ s expectations,

that only altruism, and not conscientiousness, was a statistically significant predictor of

AL LO

perceived fairness. Nevertheless, in agreement with Schnake (1991), she found a positive statistically significant relationship with job satisfaction, only for altruism but not for conscientiousness.

Tansky’ s study (1993) appeared in a special issue of Employee responsibilities and

rights journal examining the relationship between OCB and Organisational Justice. In

the same issue, Moorman et al., (1993) found more supportive evidences on the positive

relationship between procedural justice and OCB. However, the most significant finding of their study was the fact that organisational commitment and work satisfaction were

not related to OCB once the relationship between organisational justice and OCB was controlled. This result means that, at least in their sample, any effects found between

organisational commitment, work satisfaction and OCB were due to the effects of

fairness on OCB, emphasising this way the importance of the cognitive process of

fairness on citizenship behaviours. Another interesting study which appeared in the

same issue, explored the relationship between fairness and redundancy with regard to citizenship behaviours, exhibited by employees who are aware that will loose their job

soon (Bies et al., 1993). Their results provided more evidences on the emerging fairness perspective on OCB, showing that the fairness of the layoff process was the most significant factor influencing their citizenship behaviour. Summing up the findings of the studies presented in this special issue, Greenberg (1993, p. 249) denotes that “people

will behave altruistically toward the organisations in which they work if they believe they have been fairly treated by that organisation”.

Organisational Citizenship Behaviour

Konovsky and Pugh (1994) in a study examining the relationship between OCB and social exchange studied the role of distributive and procedural justice and trust in supervision with OCB. They emerged with statistically significant correlations between both types of justice and citizenship. They also confirmed their hypothesis, that trust in supervision mediates the relationship between procedural justice and citizenship behaviour. Using a social exchange framework, they maintained that the quality of the

relationship between supervisor and subordinate and the level of trust between them,

determines the participation of employees - who feel that their supervisor treats them in

a fair way - in extra-role activities.

AL LO

In the meta-analysis conducted by Organ and Ryan (1995), the findings regarding organisational justice did not show overall fairness to be a stronger predictor than satisfaction, both for altruism and generalised compliance. The 95% confidence intervals overlapped substantially and the uncorrected mean correlations were .185 and .221 between altruism and conscientiousness with overall fairness, respectively. In

general, they argued that the results of their meta-analysis failed to support the idea that

fairness is a superior predictor of OCB than job satisfaction. Finally, Konovsky and Organ (1996) in a study examining dispositional and

contextual determinants of OCB using a composite scale of perceived fairness and supervisor satisfaction, found statistically significant correlations between them and the

five common dimensions of OCB ranging from .11 up to .22. Their results showed that

work attitudes like supervisor satisfaction, perceived fairness and equity sensitivity were

better predictors than the dispositional measures used, i.e. agreeableness and

conscientiousness, with the exception of the latter and the generalised compliance form

of OCB, as will be discussed later in more detail. To sum up, it seems that there is a disagreement whether OCB has a cognitive or an

affective basis. Although the research in OCB originated from the belief that it does have an affective basis, relating it mainly with job satisfaction, it was shown that the role of cognitions might be equally, if more, important. Subsequently, if the antecedents of OCB are not mainly a function of affect, a largely dispositional variable, but rather a function of fairness, then the employers should not rely on selection as a method of

Organisational Citizenship Behaviour

increasing OCB but rather on improving management within their firm and subsequently employees’ feelings of fair treatment (Organ & Konovsky, 1989). 5.2.3 Personality and OCB Apart from the cognitive and affective basis of OCB, its dispositional determinants have also been studied with more successful and explicit results. Especially, two of the major dimensions of OCB – altruism and generalised compliance – which conceptually

are very close to two major dimensions of personality research, agreeableness and

conscientiousness have shown quite satisfactory relationships with OCB.

Smith et al. (1983) were the first to speculate on the relationship between personality

AL LO

characteristics and citizenship behaviour. More specifically, they predicted that personality would have both a direct and an indirect – through satisfaction - effect on citizenship. Nevertheless, in this first study the role of extroversion and neuroticism was studied rather than agreeableness and conscientiousness, based on past research in social

psychology about altruism. Using path analysis, they found only indirect effects of

neuroticism on altruism through its effect on satisfaction, and no effects of extroversion,

whereas regarding generalised compliance, neither direct nor indirect effects were found for any of the personality dimensions. However, Eysenck’ s lie scale had a significant

direct effect on generalised compliance. They justified this result by saying that lie

scale, in a degree, represents social approval, and it seems that people who seek social approval try to achieve that by engaging themselves in citizenship activities captured by

the generalised compliance scale. Schnake (1991) agreed with them suggesting that, at

least in self-report measures of OCB, a measure of social desirability should also be

included, which should be administered separately from the rest of the measures, and

controlled for its effects in the data analyses. Following the same approach with Smith et al. (1983) she also suggested, but without bringing any evidences, that extroversion should have a direct, positive effect on altruistic behaviour, and neuroticism an indirect, negative effect through job satisfaction. Brief and Motowidlo (1986) examining prosocial organisational behaviour, answered a tentative “yes” when they were wondering if there is a prosocial personality, i.e. whether the tendency to participate in prosocial activities is due to specific personality characteristics. Citing empirical support from earlier studies, they argued that these

Organisational Citizenship Behaviour

types of individuals have internalised higher standards of justice, responsibility, and moral reasoning and subsequently are more empathetic to other people’s feelings and therefore better able to see the world from their emotional and motivational perspective (Rushton, 1981; cited in Brief & Motowidlo, 1986). The emergence of Big Five in the personality field in the early 1990s brought a new opportunity for research in the field of OCB. Organ (1994a) in an article discussing

specifically the relationship between personality and OCB, maintained that “some

dimension or dimensions of personality underlie both satisfaction and OCB and thus

account for the consistent relationship between satisfaction and OCB” (p. 466).

AL LO

Furthermore, he argued that two of the Big Five factors, agreeableness and conscientiousness, are closely related to the personal and the impersonal dimension of OCB, i.e. altruism and generalised compliance, respectively. However, Organ (1994a) proposed that using the dimensions of the Big Five unaltered would not be the best strategy. Instead “some aspects-facets of conscientiousness might relate more to OCB,

others might not; the same could be true for other dimensions of Big Five, as well, such as agreeableness” (Organ, 1994a, p. 474). Finally, he suggested that the

conscientiousness dimension of Big Five looks to be the most promising, with regard to

OCB, “properly qualified to include aspects of achievement-striving, activity level, and affiliation” (Organ, 1994a, p. 476).

In another study studying the relationship between personality, satisfaction and OCB,

Organ and Lingl (1995) examined the hypothesis that agreeableness and

conscientiousness accounted for commonly shared variance between job satisfaction

and citizenship behaviours. Their results showed that although agreeableness and conscientiousness were significant predictors of work satisfaction – positively and

negatively, respectively – it was only conscientiousness that showed a reliable connection to OCB and only in respect to the impersonal dimension, generalised compliance. “Furthermore, the predictive power of conscientiousness became statistically reliable only when co-worker satisfaction was also in the regression equation with a similarly reliable standardised beta coefficient” (Organ & Lingl, 1995, p. 348).

Organisational Citizenship Behaviour

In the meta-analysis that was published in the same year, Organ and Ryan (1995) did not find encouraging results for agreeableness and conscientiousness, the two personality dimensions they included in their analyses, which could be a result of the small number of studies included. For agreeableness they found a sample-weighted mean estimate of .127 for altruism and .107 for generalised compliance, but in both cases, zero was included in the 95% confidence intervals around the mean. For

conscientiousness the results were more supportive, with a sample-weighted mean estimate of .217 for altruism and .302 for generalised compliance. Unfortunately, the

latter estimates were much weaker when only other-ratings of citizenship behaviours

AL LO

were included in the analyses, dropping to .043 and .228 respectively.

In a more recent study, which was included in the previous meta-analysis though, Konovsky and Organ (1996) examined once more the effects of agreeableness and conscientiousness on OCB. They predicted that agreeableness would relate particularly with altruism, courtesy and sportsmanship, whereas conscientiousness with generalised

compliance. The statistically significant correlations they came up with were quite weak; .12 between agreeableness and courtesy and .15 between conscientiousness and

generalised compliance; the rest were not statistically significant. The results of

usefulness and regression analyses they carried out next, showed that, with the exception of generalised compliance, the dispositional variables could not predict

significant variance in OCB beyond that predicted by work attitudes. However, in the

case of generalised compliance, the personality dimension of conscientiousness was the

strongest predictor of all, accounting for unique variance on the impersonal dimension of OCB. These results supported the idea that conscientiousness is the most valuable

personality dimension to study, with regard to OCB.

However, regardless of the weak relationships usually found between personality

characteristics and citizenship behaviours, it is yet inappropriate to exclude the study of personality determinants on OCB. Interestingly, causal relationships between them have not been studied extensively, whereas it is still premature to come to conclusions based on meta-analytic findings of less than a dozen studies.

Organisational Citizenship Behaviour

5.3.

Measuring OCB

A variety of measures have been used for the study of citizenship behaviour. All of the multidimensional measures share the common feature that they attempt to measure at least a personal and an impersonal form of OCB, i.e. altruism and generalised compliance-conscientiousness. Usually, they are factor analytic questionnaires, with

items developed either conceptually or after interviewing supervisors in order to identify extra-role behaviours.

Bateman and Organ (1983) used a conceptually derived measure consisted of 31 items, where the supervisor had to rate his subordinate in a 7-point scale ranging from

AL LO

“agree completely” to “disagree completely”. The items tapped a variety of behaviours such as compliance, altruism, dependability, co-operation, criticism, etc. Factor analysis using orthogonal rotation identified one dominating first factor, where all the items loaded substantially. Subsequently, the researchers decided to sum up item responses

for a single overall score of OCB. An unpublished re-analysis of the data, using oblique

rotation, did not show encouraging results as well (Organ, 1988a). Thus, it seems that

Bateman and Organ’s measure was not appropriately developed and the encouraging results they found for the relationship between OCB and satisfaction are subject to

dispute.

Smith et al. (1983) in the study which followed the same year, conducted a series of

semi-structured interviews with supervisors, asking them to describe subordinates’

behaviours that they appreciate but could not be demanded, enforced, or directly rewarded (Organ, 1988a). They originally came up with a 20-items measure, but after a

pilot study, four items were dropped because they did not load to any of the two factors

that emerged from the factor analysis. This 16-items version of the OCB measure was also used in the present study, with the following items1:

Table 15. Smith et al.’s (1983) OCB measure

1. 2. 3. 4.

Helps other employees with their work when they have been absent Exhibits punctuality in arriving at work on time in the morning and after lunch breaks Volunteers to do things not formally required by the job Takes undeserved work breaks (r)

(r) denotes reversed scoring

Organisational Citizenship Behaviour

5. Takes the initiative to orient new employees to the department even though it is not part of his/her job description 6. Exhibits attendance at work beyond the norm, for example, takes less days off than most individuals or less than allowed 7. Helps others when their work load increases (assists others until they get over their hurdles) 8. Coasts towards the end of the day (r) 9. Gives advance notice if unable to come to work 10. Spends a great deal of time in personal telephone conversations (r) 11. Does not take unnecessary time off work 12. Assists me with my duties 13. Makes innovative suggestions to improve the overall quality of the department 14. Does not take extra breaks 15. Willingly attends functions not required by the organisation, but helps in its overall image 16. Does not spend a great deal of time in idle conversation

AL LO

The subsequent factor analysis using varimax rotation showed two factors with quite straightforward interpretation. Items shown in italics loaded in the first factor, which was clearly the altruism dimension and the rest of the items, with the exception of items 8 and 15, loaded on a the clearly identified factor of generalised compliance, in

agreement with the pilot study providing a solution very close to the hypothesised.

Items 8 and 15 showed moderate loadings to any of the two factors and they were not

included in the scoring of any of them (Organ, 1988a). These two factors, although with adequate internal consistency, were highly inter-correlated showing identical results

both for oblique rotation of the factors and with orthogonal rotation. However, Dalton

and Cosier (1988) in an independent study examining the factor structure of this measure could not replicate the same factors; they came up with four factors.

Nevertheless, the subjects of their study were students that were asked to rate a

hypothetical employee in an organisation, and this probably affected the results

substantially.

The measure reported above was also used in Organ and Konovskyâ&#x20AC;&#x2122; s research

(1989) yielding three factors though. One of them was the expected altruism factor, but the generalised compliance factor was divided into two separate factors, although quite similar. The third factor consisted of three items, which although are to favour to the ratee, are negatively worded, and this is probably the reason why the formed a separate factor. However, Organ and Konovsky decided to exclude these three items from subsequent analyses.

Organisational Citizenship Behaviour

More supportive results on this measure were found in a study carried out by Becker and Randall (1994) who examined the validity of the OCB measure against a measure of objective behavioural criterion. Using a shortened 10-items version of the measure completed by the immediate supervisors of the employees, they came up with the same two factors of altruism and conscientiousness, as expected, and also with supportive results regarding the validity of the measure, although the objective measure of

behaviour used was rather unusual; whether or not the employees participated to the research.

Lowery and Krilowicz (1996) also conducted a study examining the validity of the

AL LO

instrument developed by Smith et al. (1983), in a sample consisting of production managers. They completed the OCB measure with regard to their subordinates along with a sportsmanship measure and assessed their job performance as well. Factor analysis yielded separate factors for in-role performance and citizenship dimensions, bringing this way supportive evidences on the construct validity of the OCB measure.

Moreover, the concurrent validity of the measure was also supported since altruism and conscientiousness but surprisingly not sportsmanship, were valid predictors of the

performance measure, as the results of regression analysis with the performance

measure as dependent variable, showed.

Another common approach appearing in the assessment of OCB, as it is already

discussed, is the measurement of five dimensions rather than three, i.e. altruism,

generalised compliance, sportsmanship, courtesy, and civic virtue. Various instruments

have been used, usually with supportive results, but we will not discuss their

psychometric properties here in detail, since they were not used in the present study.

In the present study it was decided to use the instrument developed by Smith et al.

(1983). The main reason for that was because it is a short instrument, with adequate and generally accepted psychometric properties. Taking under consideration the nature of the study where each supervisor should assess a number of subordinates, not only in the OCB measure but also in the competencies and the job performance measures, the length of the questionnaire was an important factor.

Organisational Citizenship Behaviour

5.4.

Conclusions

Summing up, it’ s now quite clear why citizenship behaviours were included in the present study. In the examination of a model exploring moderating and mediating effects in the personality-performance relationship, OCB may account for a significant percent of the variance, because of its relationship first with the affective state of work satisfaction, second with dispositional characteristics, and third because of its close

relation with in-role performance.

Finally, Organ (1997) has recently moved his study even further, arguing about the

importance of “morale” at work settings, in a sense equivalent to the importance of

AL LO

general intelligence. He suggested that just as general intelligence has proved to be the best predictor of in-role job performance across occupations, in the same way, morale may prove to be a general factor underlying contextual performance and citizenship behaviours. He also claimed that the moderate relationships usually found between this type of performance and job satisfaction, organisational commitment, trust, fairness,

etc., might be due to an underlying general factor, which may explain people’s thoughts and emotions described as the morale factor. If that’s really the case then it’s certainly a

topic worth studying.

Work competencies

Chapter 6. Work Competencies The expansion of personnel management and the emergence of Human Resource Management in the mid-1980s resulted in a radical development of the field followed by the appearance of a number of related approaches such as Total Quality Management (Kanji-Gopal & Asher, 1996), Quality Circles (Tang et al., 1996), empowerment

(Conger & Kanungo, 1988), business process re-engineering (Hammer & Champy, 1993), strategic / human resource planning (Lundy & Cowling, 1996), competency-

based management (Torrington & Hall, 1995), etc.

Boam and Sparrow (1992) describing the major changes occurring in HRM during

AL LO

the last decade, argued that organisations are now looking for an approach that will enable them to bring changes "by describing global issues in a way that is sensitive to the local context.....drawing upon the language of line managers and facilitating longterm changes in individuals' behaviour" (p. 5). They maintained that the time has come

for competency-based approach to address issues that large-scale organisational change

programmes have not, because (Boam & Sparrow, 1992; Sparrow & Bognanno, 1993):

1. Competency-based approaches are expressed in terms of behaviour-what people can or actually do- not what they say they do or can do;

2. They are sensitive to pick up what line managers have to do;

3. They look at the underlying nature of effective performance. As Sparrow and Bognanno (1993) argued, competencies seemed to fit the bill well.

They claimed that their increased use during 1990s was reinforced by strategic change

in organisations. The realisation that success in the new competitive environment

depends largely on the ability to learn faster than competitors, and subsequently reconstruct and adapt the organisation, has turned the focus of attention in internal

resources and capabilities of the organisation, such as employeesâ&#x20AC;&#x2122; competencies, and the integration of HR policies and practices with business strategies. The view of organisations and employees from a competence perspective, was not an exclusive use of Human Resources Management. On the contrary, competence theory has been used as a "bridge" between HR and other business fields, mainly strategic management (Foss & Knusden, 1996). Expressions, such as dynamic core competencies (Lei et al., 1996) and meta-competencies (Brown, 1993), appear frequently in strategic

Work competencies

management literature, offering a common language between researchers and practitioners in business and management. Another factor behind the development of the competency-based approach may have been the early 1980s recession, followed by the globalisation of business in the 1990s. Traditional personnel practices proved not to be effective any longer, following the waves of change in West-European and North-American economies. The established

job descriptions tailored along the lines of employees and managers, who would retain

their positions until retirement, ceased to exist, probably forever. Expressions, such as

leadership potential, innovation, creativity, strategic vision, etc. started appearing in the

AL LO

requirements sections of job adverts, seeking for a new type of employee.

The word competencies appeared in an attempt of Human Resource specialists, such as Boyatzis (1982), to capture, define and describe those new terms. Apparently, when they started their research, they were not aware of the fact that in the years to come, the

single most difficult issue in the competencies application and research, would be the

6.1.

definition of what competency is.

Defining competencies

In a literature review of the competencies field, someone would be extremely surprised, initially by the breadth of definitions of competencies, and then by the

inconsistencies, and contradictory standpoints of these researchers. Sometimes, it looks

like each writer in the field has a different view of competencies. Further, the

interchangeable use of terms such as "competence", "competences", "competency" and

"competencies" has brought even more confusion in a field that is already amid enough trouble explaining its scope and mission in business and management.

The term competence first appeared in the literature with the publication in 1982 of a

very influential book for management education and development by Richard Boyatzis, called "The competent manager". His purpose, as he defined it in the introduction of his book, was to determine the managerial characteristics related to effective performance in a variety of jobs and organisations. Boyatzisâ&#x20AC;&#x2122; (1982) attempt to describe these characteristics was built upon a model, which consists of three basic elements; jobâ&#x20AC;&#x2122;s demands and functions, the organisational environment and individualâ&#x20AC;&#x2122;s competencies.

Work competencies

These elements interact with each other, as shown in the following figure, leading to effective job performance. Boyatzis’ model of effective job performance

The individual’s competencies

The job’s demands Effective specific actions or behaviour

AL LO

Figure 3.

The organisational environment

Boyatzis (1982) defined effective job performance as: “the attainment of specific

results (i.e. outcomes) required by the job through specific actions while maintaining or

being consistent with policies, procedures, and conditions of the organisational environment” (p. 8). He argued that effective job performance would occur when all

three of the critical components of the model are consistent or “fit”. If any of these

components are inconsistent or do not correspond with each other, then it is expected that ineffective behaviour or inaction will result. He also suggests that if any two of the

components are consistent, or congruent, then there still is an increased likelihood that

effective performance will occur.

Someone would then question the role of competencies, and their relationship to

effective performance. Boyatzis (1982) explained that individual’s competencies are the certain characteristics or abilities of the person that enable him/her to demonstrate the appropriate specific actions, leading to effective work performance. Thus, in an individual level, competencies are the capabilities brought by the employee in the job context. They are underlying characteristics in that they may be motives, traits, skills, aspects, self-images, social roles, or body of knowledge used by the person. Two significant features of competencies, according to Boyatzis (1982) are first, that competencies may be known or unknown to the person, i.e. may operate on an

Work competencies

unconscious level, and second that because they are underlying characteristics, it can be said they are generic, i.e. they may be apparent in many forms of behaviour, or a wide variety of different actions. Boyatzis’ definition of competencies is very broad in the sense that it includes almost all direct and indirect personal influences on work performance, including for instance, psychological characteristics such as personality traits, motives and self-images, thus

making it problematic for research, difficult to sustain and easily misunderstood. In

addition, Boyatzis (1982) distinguished between competencies which are causally

related to effective/superior job performance and threshold competencies, which is the

AL LO

type of competency essential to performing a job but is not causally related to superior job performance. In other words, a person with a threshold competency will be able to perform only adequately at work, without reaching superior performance. Boyatzis’ research on management competencies has been essential, and will be discussed later in

more detail.

In the late 1980s the increased interest in National Vocational Qualifications and the

efforts of Management Charter Initiative (MCI) to set managerial standards across the UK, revived the competencies issue in English business literature. Weightman (1994)

claimed that other Western European countries also influenced this increase, mainly the

German training schemes which inspired European Union’s training movement.

Competence is the ability to perform the activities within an occupational area to

•

MCI’s guidelines regarding the definition of competencies included (Brown, 1993):

the level of performance expected in employment, including the ability to

transfer skills and knowledge to new situations, and include personal

effectiveness.

•

Competence is an action, behaviour or outcome which can be demonstrated, observed and validly, reliably and objectively assessed, or the demonstrable possession of underpinning knowledge or understanding.

•

Standards, which form the prime focus of training and the basis of vocational qualifications, are based on: the needs of employment, the concept of competence, the skills, knowledge and levels of performance relevant to the work

Work competencies

activity. They must be assessable and endorsed by the relevant employment sector or profession. Contrary to Boyatzis’ definition, MCI’s attempt to define competencies has been accused of being very narrow, without representing the full magnitude of management, not sufficiently sensitive, or sophisticated enough to describe and with accuracy assess

managerial performance (Brow, 1993; Holmes & Joyce, 1993). Hornby and Thomas (1989) proposing a new certificate in management, which would dovetail with the already existing Certificate and Diploma in Management

Studies, suggested that in order for the MCI to award the new certificate in

AL LO

management, the trainee must be able to demonstrate competence in the context of general management functions, knowledge, understanding, and competence in key transferable skills of management (Hornby & Thomas, 1989). They also talked about competence rather than competencies, referring to the ability to perform effectively the

functions associated with management in a work situation; they identified nine areas of

knowledge and skill, where the trainee has to be competent of.

This distinction, between competence and competency is often confused. Woodruffe (1993) suggested that competence refers to the job at which the person is competent,

whereas competencies are the aspects of the person that enable him/her to be competent.

Competencies deal with the behaviours, skills, capacities, the employee brings into a job in order to perform effectively, i.e. with competence. It is self-evident that the meaning

of those two concepts is quite different, in the sense that employees may have the

generic competencies appropriate for effective performance in a particular job, but for

reasons beyond themselves may not be in a position to perform effectively.

Nevertheless, the opposite is not also true. In order to be amongst the high performers of an organisation, the employee has to demonstrate that he/she possesses a series of workrelated competencies leading to effective performance. Collin (1989), citing Short (1984), argued that a semantic shift may have taken place in the use of the term competency. Whereas competence is normally used to imply a “state of being” whereby a person is deemed competent, competency has assumed two elements – being both a category against which one’s performance can be judged and an attribute or quality one

Work competencies

possesses. Thus, they argue that this slight shift in meaning has allowed “competency” to become synonymous with “performance”. During 1989 Personnel Management devoted a series of articles in the application of competencies approach to British Human Resource Management practices. Jacobs (1989), introducing these articles, presented the concept of competencies in the UK context and discussed in more detail MCI’s work on the field. He argued that

“competence-based assessment” was born in the USA out of dissatisfaction with the lack of precision characterising earlier management theories. Nevertheless, he

maintained that the emphasis of this approach to “scientific” methods has created a

AL LO

rather narrow view of managerial behaviour, since the definition of competencies as the necessary observable skills or abilities to complete a managerial task successfully, which are organisation-specific benchmarks of performance, excludes “soft” managerial qualities, that are not directly observable and measurable, such as innovation, creativity

and empathy, but are very important in contemporary managerial settings.

Jacobs (1989) described in detail the issues related to management competencies as

applied to assessment centres, which is the main method of assessing competencies (Smith & Robertson, 1992). The subsequent articles in the Personnel Management

series described in detail the use of assessment centres mainly, but not exclusively, as a route in identifying and measuring employees’ competencies. Glaze (1989) described

the approach followed in Cadbury’s, Greatrex and Philips (1989) BP’s system of

measuring competencies, and Cockerill (1989) explained the use of competencies in

NatWest. The competencies identified by these companies have both common and

different characteristics, as we will discuss later, describing the various competencies

models.

Sparrow (1997), in a review of the use of organisational competencies in personnel

selection and assessment, defined competencies as those behavioural repertoires (sets of behavioural patters) which some people can carry out more effectively than others, including all those behaviours that employees bring into the organisation in order to perform well. Following Boyatzis’ definition of competencies he maintained that they consist of the following elements (see table 16).

Work competencies

Table 16. Elements of competencies (adapted by Sparrow, 1997, p. 348)

AL LO

1. Bodies of knowledge: what the employee needs to know in order to achieve the goals the job specifies. Information in specific content areas. 2. Skills: what the employee has to possess in order to do the job. The ability to perform a physical or mental task and demonstrate a sequence of behaviour that is functionally related to a performance goal and can be applied to a range of situations. 3. Attitudes and values: what the employee needs to display in connection with achieving the tasks. Attitudes that predict behaviour in the short or long term. 4. Traits: characteristic or quality of a person that is associated with effectiveness. Physical characteristics and consistent responses to situations or information. 5. Motives: drive or thought that is related to a particular goal. The things a person consistently thinks about or wants that cause a desired action or goal. 6. Self-image: the understanding an individual has of him or herself in the context of values held by others. 7. Social role: the perception of social norms and behaviours that are acceptable and the behaviours that a person needs to adopt in order to fit in.

Hay (1990) argued, contrary to previous definitions, that core competencies are not

skills, or knowledge. They are “orientations”, the part of personality that may be

inferred through observation of behaviour, describing the manager’s approach to a

situation rather than the skill level exhibited. This alternative approach in management

competencies, the adaptation of a specific style, as a result of the possession of particular competencies is followed in the work by Schroder and Cockerill (1990), as

well. These ideas are developed based on Schroder’s eleven “High Performance

Managerial Competencies”, which according to the author are important determinants of the organisational performance as the rate of change in the external environment

increases. Schroder and Coskerill’s (1990) research examined the relationship between

managerial competencies and Kirton’s adaptation-innovation theory in order to propose

an integrated theory of management competence and cognitive style. They concluded that not all competencies are associated with innovative job and that managers who are mismatched with their cognitive style will be presented with fewer opportunities to use their competencies effectively. Finally, Weightman (1994) in one of the most recent books regarding competencies, examined in detail the British scene describing all the attempts initiated by professional bodies, such as the National Vocational Council, and the applications of the competency based approach in organisations, an issue discussed later in detail. The definition he

Work competencies

adopted was based on Boyatzis’ work, relating competencies to performance, as the demonstration of behaviours that lead to particular (effective) outcomes-performance. Summing up, defining competencies is not the easiest task for a researcher in the field. The situation gets even worse when someone attempts to identify, classify and describe work competencies. To summarise, the common characteristics of these

1. Competencies are related to job performance. 2. They are inner capabilities-skills possessed by individuals.

definitions may be extracted:

3. They are observable, and should be measured with reliable and valid measures.

AL LO

There are however a number of inconsistencies regarding those definitions, more specifically, the elements constituting competencies, the level of performance they should be related (acceptable-superior), the distinction between competencies and competence, and the best way to classify different types of competencies; an issue that

6.2.

will be discussed in the following section.

Competencies classification

Numerous lists and types of competencies have been described in the literature. A

number of issues related to their identification and classification are following:

6.2.1 Generic or organisation-specific competencies? The relevance of this issue is self-evident in competencies research. Is there a need

for competencies designed specifically for particular organisations or there could be a

list of generic-universally accepted- list of competencies (Woodruffe, 1993)? It will

become apparent that most of the lists described later in the section identify common

competencies across various organisations and level of positions. The problem with generic lists as Woodruffe (1993) indicates, is that they assume a uniformity in job titles, which is not usually the case, since the same job title across firms does not necessarily mean equivalent work requirements or duties. In addition, generic models of competencies, such as MCI’s or NVQs, have been heavily criticised by managers, according to Sparrow (1997), because they are “overly-bureaucratic, unspecific and irrelevant to the way we do things around here” (p. 347).

However, it is generally

Work competencies

accepted that generic, off-the-self models may provide very useful information for the development of more specific models (Iles, 1993). 6.2.2 Competencies as inputs or outputs? The distinction of competencies as either inputs or outputs is directly related to the assessment of employeesâ&#x20AC;&#x2122; competencies. Should they be identified and assessed as the characteristics and attributes people bring into a role influencing his/her ability to

perform (input), which may assessed in a selection setting for instance, or they should

be directly related to performance at work (output), in other words the characteristics

assessed in a performance appraisal scheme? Parrry (1996) claims that those two

AL LO

approaches have also different origins. The view of competences as outputs is primarily British where in the USA, competencies are usually seen as inputs. 6.2.3 General or specific competencies?

This question is largely a psychometric issue. If the statements describing the

competencies are too large and general, they are non-discriminating; if they are too

small and detailed they may become irritating, and prescriptive. Subsequently, an

appropriate balance should be achieved so that all the appropriate information is

acquired without being too specific (Weightman, 1994). 6.2.4 Individual or team / organisational competencies?

Most of the competencies schemes initiated either by professional bodies or specific

organisations are individually based. There might be cases however, when this approach

is not appropriate and a focus on the right balance of competencies within a team may

be preferred, such as within a management team where a broad range of competencies have to appear in order for the team to work effectively without conflicts (Weightman,

1994). In addition, individually assessed competencies do not help the organisation to judge individual contributions to the outcome of a group activity (Antonacopoulou & FitzGerald, 1996). 6.2.5 Hard or soft (e.g. transformational) competencies? This distinction has recently attracted the attention of competencies research, since it has been widely recognised that soft competencies such as innovation, creativity, assertiveness, and empathy, should be at the forefront of managerial positions in the

Work competencies

current organisational climate of risk, uncertainty and change (King & Anderson, 1995). Nevertheless, contemporary methods of assessing competencies, such as assessment centres, face major difficulties in assessing these type of competencies as discussed earlier (Jacobs, 1989), calling for alternative assessment methods. 6.2.6 Core or peripheral competencies? Woodruffe (1993) argues that organisations should not distinguish between core and

peripheral competencies, because there is the danger that they will concentrate their attention only to those competencies labelled as more important disregarding the

peripheral competencies. He claims that competencies list should be taken as a whole.

AL LO

Core competencies have also attracted the attention of strategic management research. Weightman (1994) maintains that the use of core competencies in strategic management acknowledges that organisational success is more than just allocating money in the short term, but also depends on employeesâ&#x20AC;&#x2122; skills / competencies, although

Sparrow (1997) suggests that strategists view management competencies in a more

technical and market directed way than psychologists.

6.2.7 What is the appropriate level of analysis?

This feature was partially mentioned when discussing Boyatzisâ&#x20AC;&#x2122; (1982) distinction between threshold competencies and competencies leading to superior performance. A

different approach to this issue has been adopted by some organisations (Weightman,

1994). They identify a series of general competencies required from their employees,

but different level of mastery is requested for different places in the hierarchy. For

instance, communication and interpersonal competencies may be required for everyone working in an organisation, but the levels of expertise for these competencies are

different for a secretary compared to the finance or personnel manager. Nevertheless, Woodruffe (1993) argues that, as was the case earlier for core competencies, there is the danger that for people in higher levels of the hierarchy, some competencies may be regarded as non-significant anymore. In the following section the main competencies models described in the literature will be outlined.

Work competencies

6.3.

Boyatzis’ model of effective performance

Boyatzis and his colleagues at McBer company developed their competencies model with information gathered by twelve public and private organisations and more than two thousand people from forty one management jobs within those organisations. Subsequently, their model concerns generic managerial competencies. The following

Figure 4.

McBer’s model of identifying managerial competencies

figure describes their steps in identifying and describing managerial competencies:

1. Determination of the appropriate measure of job performance and ways of assessing it.

AL LO

2. Job element analysis in order to identify characteristics that managers perceive as important in distinguishing superior from average performers.

3. Critical incident interviews identifying important incidents of the employees’ work behaviour.

4. Test administration measuring competencies assessed in previous stage. Validation to job performance criteria.

5. Results’ integration

Boyatzis and his colleagues (1982) initially made the hypothesis that 21

competencies were related to managerial effectiveness. Nineteen of them were

classified into five categories and two additional competencies were standing alone.

These competencies along with brief definitions are presented in the following table:

Work competencies

Table 17. Boyatzis’ managerial competencies (1982)

AL LO

Goal & Action Management Cluster It represents a concern with doing something better. At the motive 1. Efficiency level is the n-achievement occurring consciously or unconsciously. orientation It represents a disposition toward taking action to accomplish something. At the trait level is people’s sense of efficacy, i.e. the 2. Proactivity disposition to see oneself as the originator of actions in one’s life. A way of thinking in which the person identifies or recognises patterns from an assortment of information, by bringing a concept 3. Diagnostic use of to the situation and attempting to interpret events through that concepts concept. It represents a concern with symbols of power to have impact on 4. Concern with others. At the motive level it is called the n-power. impact Leadership Cluster It is also called decisiveness or presence. People with this 5. Self-confidence competence have a positive self-esteem. A competency where people make effective verbal presentations. 6. Use of oral These people see themselves as able to communicate effectively, presentations adopting roles as communicators. It represents a thought process where the person places events in causal sequence based on the perception of a series of cause-and7. Logical thought effect events. These people see themselves as orderly and systematic. It is a thought process in which the person identifies or recognises patterns in an assortment of information; that is the individual 8. Conceptualisation develops a concept that describes a pattern or structure perceived in a set of facts. Human Resource Management Cluster It is the competency where the person uses forms of influence to 9. Use of socialised build alliances, networks, coalitions, or teams. They perceive power themselves to be a part of particular groups. It is a competency in which people believe in others; they have a 10. Positive regard positive belief that people are good, adopting the role of optimist. People who have this competency can stimulate others to work 11. Managing group together effectively in group settings. These people adopt the role process of collaborator or integrator. It is a competency in which people have a realistic or grounded view of themselves. These people see their strengths and 12. Accurate selflimitations and demonstrate self-assessment skills and reality assessment testing skills. Directing Subordinates Cluster It is a competency with which managers specifically help someone to do his/her job. They see helping others as an essential 13. Developing others characteristic of the managers’ job. People with this competency use forms of influence to obtain 14. Use of unilateral compliance stimulating their subordinates to go along with their power directions, wishes, commands, policies, or procedures. It is a competency with which people can express themselves freely or easily. People with this competency may find themselves 15. Spontaneity in the role of a provoker or a jester. (table continued)

Work competencies

Focus on Others Cluster 16. Self-control

17. Perceptual objectivity

18. Stamina and Adaptability

Specialised knowledge

AL LO

19. Concern with close relationships

Memory

It is a competency with which people inhibit needs or desires in service of organisational needs. People with this trait consistently weigh the costs and benefits to themselves and to the organisation, or other system of which they are part, before expressing their personal needs or desires. People with this competency can be relatively objective and not limited in view by excessive subjectivity or personal biases. They view an event from multiple perspectives simultaneously. People with this competency have the energy to sustain long hours of work and have the flexibility and orientation to adapt to changes in life and the organisational environment. It is a competency that people care about and build close relationships with individuals best described as friendships. At the motive level is termed the n Affiliation. These people see themselves as likeable and caring. Boyatzis considers specialised knowledge to be a threshold competency, i.e. a person’s generic knowledge, motive, trait, selfimage, social role or skill which is essential to performing a job but is not causally related to superior job performance. Memory is not relevant to itself but it becomes relevant only in context with other competencies. It does not appear as a separate competency but is so basic to performance as a manager that it is a precursor to demonstration of any other competencies.

Extensive research by Boyatzis (1982) and his colleagues concluded that not all of

these competencies are related with superior performance. Some of those were found to be threshold competencies, whereas two others dropped completely from their initial

model. Those two were “concern with close relationships” and “memory”. The

following table presents those results:

Cluster

Table 18. Summary of McBer’s results (1982)

Goal and action management

Leadership Human Resource Management Directing subordinates

Competency • Efficiency orientation • Proactivity • Diagnostic use of concepts • Concern with impact • Self-confidence • Use of oral presentations • Conceptualisation • Use of socialised power • Managing group process

Threshold competency

•

Logical thought

• • • • •

Positive regard Accurate self-assessment Developing others Use of unilateral power Spontaneity (table continued)

Work competencies

• • •

Focus on others

Self-control Perceptual objectivity Stamina and Adaptability

Specialised knowledge

•

Specialised knowledge

Summing up these results, Boyatzis (1982) argued that the six clusters of management competencies may be appropriately explained in terms of the five basic

managerial functions: 1) planning, 2) organising, 3) controlling, 4) motivating, and 5)

co-ordinating. Some of the competencies’ clusters are more related to some of the five

functions rather with others, although the first three clusters, goal-action, leadership and

AL LO

human resource are very significant across all five functions. Nevertheless, Boyatzis emphasised the limitations and the exploratory nature of this study and also the necessity for further research in order to establish clear patterns of causality between generic managerial competencies and work performance. Despite its limitations, the

study was a very significant step forward in business and management research and

application, since it offered significant input and insight into two rather subtle fields,

which are very much interrelated and dependent one to the other, such as management,

6.4.

in terms of the role and functions of it, and generic competencies.

Schroder’s work

Similarly to Boyatzis, Scroder’s (1989) research on competencies focused on the

relationship between cognitive competencies and performance in group and individual

situations (Schroder, 1992). In a series of studies, he identified eleven High

Performance Managerial Competencies (HPMC), which are required for success in the new

changing

organisational

environment

emerging

response

global

transformations. The competencies he identified, and their definitions are presented in the following table:

Work competencies

Table 19. Schroderâ&#x20AC;&#x2122;s High-Performance Managerial Competencies (HPMC) (Schroder & Cockerill, 1990)

AL LO

Cognitive competencies Gathers many different kinds of information and uses a wide variety 1. Information of sources to build a rich informational environment in preparation for search organisational decision making. Builds frameworks or models or forms concepts, hypotheses or ideas 2. Concept on the basis of information; becomes aware of patterns, trends and formation cause/effect relations by linking disparate information. Identifies feasible alternatives or multiple options in planning and 3. Conceptual decision making; holds different options in focus simultaneously and flexibility evaluates their pros and cons. Motivational competencies Uses open and probing questions, summaries, etc. to understand the 4. Interpersonal ideas, concepts and feelings of another; can comprehend events, search issues, problems, opportunities from the view point of another person. Involves others and is able to build co-operative teams in which group 5. Managing members feel valued and empowered and have shared goals. interaction Creates a positive climate in which individuals increase the accuracy of their awareness of their own strengths and limitations and provides 6. Developmental coaching, training and developmental resources to improve orientation performance. Directing competencies States own stand or position on issues; takes decisions without hesitation when required and commits self and others accordingly; 7. Self-confidence express confidence in the future success of the actions to be taken. Presents ideas clearly, with ease and interest so that the other person (or audience) understands what is being communicated using 8. Presentation appropriate aids effectively. Uses a variety of methods (persuasive arguments, modelling behaviour, forming alliances, etc.) to gain support for ideas, strategies 9. Impact and values. Achieving competencies Structures the task for the team; implements plans and ideas, taking 10. Proactive responsibility for all aspects of the situation even beyond ordinary orientation boundaries-and for the success and failure of the group. Possesses high internal work standards and sets ambitions, risky and 11. Achievement yet attainable goals; wants to do things better, to improve, to be more orientation effective and efficient; measures progress against targets.

Schroder has focused his attention, as we mentioned earlier in the chapter, on the

relationship between management competencies and managerial style (Schroder & Cockerill, 1990; Schroder, 1992) using Kirtonâ&#x20AC;&#x2122;s adaptation-innovation (A-I) theory as a framework. According to Kirton, (Schroder, 1992) adaptors and innovators are equally

creative, but in different ways. Adaptors depend on generally agreed and already tested successfully solutions, where innovators are more likely to reconstruct the problem and work outside the boundaries of accepted solutions, initiating change based on different

Work competencies

ways of doing things. The two theories were integrated by Schroder and Cockerill (1990) as shown in the following figure. Two environmental dimensions are shown as being orthogonal (the rate of change and the type of change). In each quadrant, the combination of HPMC and KAI which is optimally related to organisational performance according to the integrated theory is shown. Integrated theory of HPMC and KAI (Schroder & Cockerill, 1990)

AL LO

Dynamic environment

Figure 5.

High HPMC Adaption

High HPMC Innovation

Discontinuous change

Low HPMC Innovation

Low HPMC Adaption

Continuous change

Stable environment

Their results offered partial support of their hypotheses. Seven of the eleven HPMC

competencies Schroder identified, were positively correlated with KAI scores, i.e. it was

more likely that an innovator would demonstrate those competencies, a result partly

attributed by the authors to assessment bias in favour of innovators. They also argued that this finding could be explained as a result of the relationship between job content

and the attributes of those involved, in other words because the majority of the sample were executives, involved primarily in the management of discontinuous rather continuous change, which provides innovators with far more opportunities to use HPMC competencies than adaptors (Schroder & Cockerill, 1990). The authors concluded that the findings implied that managers who are mismatched with the cognitive bias of their job will be presented with less opportunities to use their competencies than managers with equal competence who are well matched, with direct

Work competencies

consequences to organisational performance, employees’ job satisfaction and staff retention. Schroder’ s model of management competencies was applied and tested successfully in Natwest’ s developmental assessment centres (Cockerill, 1989), using a combination of personality tests (Myers-Briggs type indicator), structured interviews (using the “career anchors” approach), group discussions, and information for the HPMC gathered from various sources, including assessment centre exercises,

The model of the Management Charter Initiative (MCI)

6.5.

supervisors, peers, subordinates and self-ratings.

A few things about MCI’s attempt to develop managerial standards have already

AL LO

been discussed in earlier sections. In this section, MCI’s areas of competence will be identified and the criticisms of the model described in the literature will be highlighted. MCI’s contribution to British management and education, although controversial, has been significant since it was the first attempt carried out in this country to establish

occupational standards / qualifications for managers, aiming mainly in identifying a

series of generic competencies applicable across levels and types of managerial jobs,

which along with organisation/job specific competencies will best describe particular

managerial jobs. MCI, following extensive research in industry, but also following a severely criticised methodology, called functional analysis (Holmes & Joyse, 1993),

identified “occupational standards for managers“ at levels I and II, targeted at junior and

middle level managers respectively. Four key roles were identified for junior managers.

table:

These roles with their associated units of competence are presented in the following

Table 20. Key roles and their associated units of competence for level I in management (adapted by BTEC & CNAA, 1990) Key role

Manage operations Manage finance

Units of competence Maintain and improve service and product operations Contribute to the implementation of change in services, products and systems Recommend, monitor and control the use of resources (table continued)

Work competencies

Contribute to the recruitment and selection of personnel Develop teams, individuals and self to enhance performance Manage people

Plan, allocate and evaluate work carried out by teams, individuals and self Create, maintain and enhance effective working relationships Seek, evaluate and organise information for action

Manage information

Exchange information to solve problems and make decisions

As far as middle management is concerned, MCI identified five areas of competence,

AL LO

presented in the following table:

Table 21. Key roles and their associated units of competence for level II in management (adapted by Torrington & Hall, 1995) Key role

Identify training and development needs

Design training and development strategies and plans Provide learning opportunities, resources and support

Units of competence Identify organisational requirements for training and development Identify the learning needs of individuals and groups Design organisational training and development strategies and plans Design strategies to assist individuals and groups to achieve their objectives Obtain and allocate resources to deliver training and development plans Provide learning opportunities and support to enable individuals and groups to achieve their objectives Evaluate the effectiveness of training and development Evaluate individual and group achievement against objectives Assess achievement for public certification Contribute to advances in training and development

Evaluate the effectiveness of training and development

Provide services to support training and development practice

Support training and development advances and practice

Each of the units of competence is further divided into associated elements of

competence, performance criteria for each element, and range indicators relevant to the performance criteria. Each range indicator describes an aspect of behaviour. It is a statement of what a manager actually does to demonstrate a particular dimension of effective behaviour. In addition, a model of personal competence, spanning at all levels,

Work competencies

was also identified. The dimensions of the personal competence model are described the following table:

Clusters Planning to optimise the achievement of results

Managing oneself to optimise results

Using intellect to optimise results

AL LO

Managing others to optimise results

Dimensions of personal competence Showing concern for excellence Setting and prioritising objectives Monitoring and responding to actual against planned activities Showing sensitivity to the needs of others Relating to others Obtaining the commitment of others Presenting oneself positively to others Showing self confidence and personal drive Managing personal emotions and stress Managing personal learning and development Collecting and organising information Identifying and applying concepts Taking decisions

Table 22. The personal competence model (Management Charter Initiative, 1993)

Assessment methods of managerial competencies varied, including accreditation by

prior learning, interviews and work-based assignments. Nevertheless, although MCI

was the first professional body to establish standards in British management and relate it to vocational qualifications, it was also heavily criticised. A lot of organisations,

although initially attracted by MCIâ&#x20AC;&#x2122;s approach, then developed their own lists of

competencies, criticising it, as being static, non-responsive, and bureaucratic,

accusations often directed against generic approaches to competencies. Additionally, it was criticised for the methodology followed in identifying the types of necessary

competencies, and for its lack of external validity and rigorous empirical research

(Holmes & Joyse, 1993). However, MCIâ&#x20AC;&#x2122;s approach to management competencies is an ongoing

process,

continuously

evaluating

and

redefining

their

management

competencies, in an attempt to establish satisfactory standards for management.

6.6.

Other generic competencies lists

A list of generic work competencies was developed in Henley Management College. Dulewicz (1989) concluding a series of five articles in Personnel Management about the application of competencies in five British companies, presented the results of a study

Work competencies

of managers, rated at their organisations as highly successful, conducted during an eight-week course in Henley Management College, whereas self and boss ratings in 40 basic competencies were factor analysed resulting in 12 independent performance factors called supra-competencies. These competencies were grouped by Dulewicz under four main headings. Table 23. Dulewicz supra-competencies (adapted by Dulewicz, 1989)

Intellectual

Rises above the detail to see the broader issues and implications; takes account of wide-ranging influences and situations both inside and outside the organisation before planning or acting. Seeks all relevant information; identifies problems, relates relevant data and identifies causes; assimilates numerical data accurately and makes sensible interpretations; work is precise and methodical, and relevant detail is not overlooked. Makes decisions based on logical assumptions that reflect factual information. Plans priorities, assignments and the allocation of resources; organises resources efficiently and effectively, delegating work to the appropriate staff.

1. Strategic perspective

AL LO

2. Analysis and judgement 3. Planning and organising

Interpersonal

Adopts appropriate styles for achieving group objectives; monitors and evaluates their work; shows vision and inspiration; develops the skills and competencies of staff. Influences and persuades others to give their agreement and commitment; in face of conflict, uses personal influence to communicate proposals, to reach bases for compromise and to reach an agreement. Ascendant, forceful dealing with others; can take charge; is willing to take risks and seek new experiences; is decisive, ready to take decisions, even on limited information. Shows consideration for the needs and feelings of others; listens dispassionately, is not selective, recalls key points and takes account of them; is flexible when dealing with others, will change own position when othersâ&#x20AC;&#x2122; proposals warrant it. Fluent, speaks clearly and audibly, with good diction; n formal presentations, is enthusiastic and lively, tailors content to audienceâ&#x20AC;&#x2122;s level of understanding.

5. Persuasiveness

6. Assertiveness and decisiveness

4. Managing staff

7. Interpersonal sensitivity

8. Oral communication Adaptability

9. Adaptability and resilience

Adapts behaviour o new situations; resilient, maintains effectiveness in face of adversity of unfairness. Performance remains stable when under pressure or opposition; does not become irritable and anxious, retains composure

Results-orientation 10. Energy and initiative

Makes a strong, positive impression, has authority and credibility; is a self-starter and originator, actively influences events to achieve goals; has energy and vitality, maintains high level of activity and produces a high level of output. (table continued)

Work competencies

11. Achievementmotivation 12. Business sense

Sets demanding goals for self and for other, and is dissatisfied with average performance; makes full use of own time and resources; sees a task through to completion, irrespective of obstacles and setbacks Identifies opportunities which will increase sales or profits; selects and exploits those activities which will result in the largest returns.

Dulewicz (1989) suggested that after excessive personal research covering 20 years

of managerial assessment and job analyses, he strongly believed that the competencies

lists generated by various organisations have a high degree of commonality, claiming

that only 30 per cent are organisation-specific requirements, whereas 70 per cent are common across organisations. Dulewicz (1989) compared his list of supra-competencies

AL LO

with the lists identified in a number of a organisations reported in a series of articles in Personnel Management (see the following table), and argued for the similarities between them, in favour of supra-competencies.

Table 24. Examples of competencies as part of assessment centres

Glaze (1989) described the approach followed in Cadburyâ&#x20AC;&#x2122;s where Belbinâ&#x20AC;&#x2122;s team roles, psychometric tests and competence data have been used. The first two were mainly used in selection and placement decisions where competence data were used as part of individual development. The dimensions of management competencies used were strategy, drive, relationships, persuasion, leadership, followership, analysis, implementation, personal factors. Greatrex and Philips (1989) working with BP have tried not to rely on a general list of competencies but to be situational specific reflecting the culture of their organisation. Using job analysis techniques such as critical incident and repertory grid they identified four clusters of competencies, which according to the authors gave a very powerful picture of the corporate culture of the organisation. These clusters were achievement orientation (personal, organisational drive, impact and communication), people orientation (awareness of others, team management, persuasiveness), judgement (analytical power, strategic thinking, commercial judgement), situational flexibility (adaptive orientation). Cockerill (1989) argued that most assessment centres are not designed to be relevant to flexible organisations in dynamic environments. Using Schroderâ&#x20AC;&#x2122;s (1989) list of managerial competencies he argued that those can be suitable for superior performance in changing environments. Another list of generic work competencies was developed by Woodruffe (1993), who related competencies to personality traits and motives, without, however, establishing this claim adequately. Woodruffe (1993) was very much in favour of generic competencies, which may be used in combination with organisation specific

Work competencies

competencies, or as a check against which organisational lists may be compared. The list of generic competencies he described, with their definitions are presented in the following table: Table 25. Woodruffe’ s list of generic competencies (adapted by Wooduffe, 1993)

AL LO

Breadth of awareness to be well informed

Develops and maintains networks and formal channels of communication, within the organisation and with the outside world; uses information technology to gain information; maintains an awareness of what should be happening and what progress is being made; keeps abreast of relevant local, national and international political and economic developments; monitors competitor activity. Gets a clear overview of an issue; grasps information accurately; relates pieces of information; identifies causal relationships; gets to the heart of a problem; identifies the mot productive lines of enquiry; appreciates all the variables affecting an issue; identifies limitations to information; adapt thinking in light of new information; tolerates and handles conflicting/ambiguous information and ideas. Generates options; evaluates options by examining the positive and negative aspects if they were put into effect; anticipates effects of options on others; foresees others’ reactions; demonstrates common-sense and initiative. Identifies properties; thinks back from deadline; identifies elements of tasks; schedules elements; anticipates resource needs; allocates resources to tasks; sets objectives for staff; manages own and others’ time. Prepared to compromise to achieve a result; installs solution within time frame; innovates or adapts existing procedures to ensure a result; takes on problems; suffers personal inconvenience to ensure problems are solved; comes forward with ideas; sets challenging targets; sets out to win new business; sets own objectives; recognises areas for selfdevelopment; acquires new skills and capabilities; accepts new challenges. Express and conveys a belief in own ability; prepared to take and support decisions; stands up to seniors; willing to take calculated risks; admits to areas of inexpertise. Listens to others’ viewpoints; adapts to other person; takes account of others; needs; shows empathy in oral and written communications; aware of others’ expectations. Involves others in own area and ideas; keeps others informed; makes use of available support services; utilises skills of team members; open to others’ ideas and suggestions. Sticks to a plan; does not get side-tracked; sacrifices the present for the future; bides time when conditions are not favourable.

Incisiveness to have clear understanding

Reasoning to find ways forward

Organisation to work productively

Drive to achieve results

Self confidence to lead the way Sensitivity to identify others’ viewpoints Co-operativeness to work with other people Goal-orientation to win in the long term

Summing up, generic competency models may be an attractive, timesaving solution, for competence-based management. Nevertheless, they should be used with caution, and

Work competencies

preferably in conjunction with organisation, job-specific competencies, identified through job analyses.

6.7.

Assessing competencies

According to Smith and Robertson (1992), the next step once the competencies have been identified, is to assess the extent they are held by existing personnel, if in a

performance appraisal scheme, or by job applicants, if in a personnel selection setting. They describe four methods of assessing competencies: analogous approaches, where

the key aspects of the job are copied; analytical approaches, where “abstract” activities, quite different from those performed on the job are used; reputational approaches,

AL LO

involving other people’s judgements and miscellaneous approaches. In the following table, the different types of exercises, for each approach are identified, along with the competencies that can be measured in each exercise.

Category

Examples of competencies measured

Approach

Table 26. Methods of measuring competencies (Smith & Robertson, 1992)

In-tray exercises Role-playing Analogous

Presentations

Written reports

Psychomotor tests

Analytical

Reputational

Miscellaneous

Social skills, planning, organising Analytical skills, problem solving, planning, organisational awareness, flexibility, initiative, decisiveness Problem solving, oral communication, emotional stability, interpersonal skills Oral communication, planing, organising, emotional stability Written communication, analysis and problem solving, organising and planning Eye-hand co-ordination, keyboard ability, error correction Potential to develop job-specific competencies Visual acuity, finger dexterity, hand-eye coordination. Personality dimensions Motives (e.g. need for power, achievement), general/specific interests

Group exercises

Trainability tests Ability tests Temperament tests Motivation-interest tests References, supervisors’/peers’/ subordinates’ assessments Biodata, accomplishment record, future autobiography

Job-related competencies

Work competencies

6.8.

Applications of the competency approach

The competencies approach may be a very influential part of the Human Resources strategy of any organisation. Since the appropriate competencies have been identified, they can be applied in a number of organisational activities, including, personnel selection and assessment, training, career development, performance management, etc. Sparrow and Bognanno (1993) claimed that the creation of a core set of effective encouraging mutual behavioural reinforcement across HR policy areas.

behaviours, such as competencies, may be used in a large set of assessment settings

Feltham (1992) claimed that the competency-based approach contributed to the

AL LO

effectiveness of selection and assessment methods in three ways; the process, the implementation and the evaluation of the approach.

“The process of the competency-based approach helps the organisation to identify what it needs for its human resources, and to specify the part that selection and assessment can play. Second, the implementation of competencybased recruitment and selection systems results in a number of direct practical benefits. And third, where systems are linked to competencies, aspects of fairness, effectiveness and validity become amenable to evaluation. (Feltham, 1992, p. 96). The main benefits in adopting a competency approach in selection and assessment,

according to Feltham (1992), are the creation of shared understanding between various levels of the hierarchy of the kind personnel needed for new systems, more informed

human resource options, agreed standards and more systematic and scientific

recruitment processes, realistic job previews and identification of the most appropriate

assessment methods.

Another application of the competency approach is in future career development

(Craig, 1992). He argued that competencies in career development are important in two, stages; prior to promotion and towards the middle of the grade when both organisation and the individual should be preparing for movement towards the next grade. He described the following flow: •

Competencies set standards for progression

•

Competencies provide the referent for assessment

•

Competencies, expressed in terms of strengths and development needs, set the referent for development

Work competencies

•

Competencies set the standards for progression to the next level.

The use of competencies in training came as a result of the need to establish costbenefit links between training and improvement in job performance (Antonacopoulou & FitzGerald, 1996). Applying competencies in training may be very appropriate in deciding what to train (Weightman, 1994), in identifying specific competencies that require further improvement, either in an individual or departmental / organisational

level.

Another area, where competencies have been widely used, especially here in the UK,

is in competence-related pay and performance-related pay schemes (Torrington &

AL LO

Blandamer, 1992). The suitability of competencies in those schemes is justified, since they can offer clarity in identifying objectives and goals that have to be reached, both for the appraiser and the appraisee, facilitating the appraisal process. In addition, it offers a common language between them, and between personnel management and trade unions. Armstrong & Brown (1998) claim that the use of competencies can help to

address traditional failings of performance-related pay, being particularly appropriate in sectors, where employees’ skills and behaviours are considered as the key to

Conclusions

6.9.

competitive success.

The competency approach in human resource management, although not as

fashionable as it used to be a few years ago, can still play an important role in

organisations, as shown in this section, offering an alternative to traditional methods applied in Human Resources. Incorporating employees’ competencies in a study

investigating moderating-mediating effects in the personality-job performance

relationship accepts that a combination of particular competencies with specific personality traits or with indicators of contextual performance may account for significant variance in employees’ effectiveness at work.

Personality dispositions and personality processes

Chapter 7. Personality dispositions and personality processes The findings regarding the criterion-related validity of the five-factor model support the argument for more extensive use of big-five measures in personnel selection and assessment. Nevertheless, these encouraging results do not provide an adequate theoretical explanation of how these relationships actually work in everyday work

settings, and why they explain some of the variance in job performance. Further, the question of how specific personality dimensions interact with work-related variables

such as job satisfaction and generic work competencies, in predicting job performance still remains unresolved. Thus, in order to investigate the moderating and mediating

AL LO

effects of specific work-related variables in the validity of the FFM, which is the main focus of the present research, the adoption of an appropriate theoretical framework was essential. The focus of attention was shifted to a strong and heavily studied psychological theory, social-cognitive theory. This theory is a potentially useful tool in

the examination of the criterion-related validity of the five-factor model. The work

originated by Mischel and Shoda (1995, 1998; Shoda & Mischel, 1996) who, during the

last few years, have attempted to reconcile personality dispositions with the principles

7.1.

of social-cognitive theory was used as a guide.

The cognitive-affective personality system

In the late 1960s a number of psychologists questioned the usefulness of

psychoanalytic, trait and behavioural perspectives in explaining human nature

sufficiently. Walter Mischel describing a reconceptualised, unifying view of personality

(Mischel, 1973) proposed a set of variables called cognitive social person variables in an effort to identify an alternative to trait and psychoanalytic approaches to personality

research and practice. According to Mischel (1993), these variables interact with each other as the individuals interpret the social world around them and their actions in this world. These personal variables were the individuals' encodings, expectancies, values, self-regulatory systems and competencies. Each person perceives, construes, and encodes the self, other people, situations and events differently. The ways we do that influence our behaviour and the actions that follow. Research into selective attention, interpretation and categorisation has shown

Personality dispositions and personality processes

that people, even when they are exposed to the same events, will most probably encode them in a number of ways. Individuals' encodings are not enough though to explain behaviour. It is mainly their expectancies and beliefs that are directly related to performance and the actions they take. People generate behaviour in line with their expectations. Thus, if a person strongly believes that taking a particular action enhances the possibility of positive outcomes, then he or she will be keener to take and

successfully complete this action. Mischel (1993) identified three types of expectancies: behaviour-outcome relations, where in any given situation people produce the pattern of

behaviour that will lead to their preferred outcomes; stimulus-outcome relations, where specific "signs" or cues predict events that are likely to follow and self-efficacy

AL LO

expectations related to individuals' personal beliefs of his/her capabilities to accomplish a task.

The next set of personal variables concerns the individualsâ&#x20AC;&#x2122; subjective values. Different people rate the same behavioural outcomes differently, depending on the

values they assign to these behaviours. Thus, one person may perform a specific action because the outcomes are valued very highly; another may not perform it because s/he

does not value its outcome. For example, an employee who appreciates co-workersâ&#x20AC;&#x2122;

approval is more likely to perform activities that are accepted by them. People also determine their behaviour by self-regulatory systems and plans, i.e. by internal rules and

goals the individuals set for them. Self-regulation may influence our environment

substantially, as Mischel argues (1973, 1993), since people can actively select the

situations they are exposed to in line with the rules and standards they have set up for themselves. These personal variables largely depend on individualâ&#x20AC;&#x2122; s ability to generate

particular cognitions, thoughts and behaviours. An ability, which, according to social

cognitive theory, people acquire through direct and observational learning, and is mainly reflected to the capacity to transform and accordingly use social information and knowledge. Quite recently Mischel and Shoda (1995, 1998; Shoda & Mischel, 1996) presented a new framework based on these five social cognitive personal variables. Their theory called the Cognitive Affective Personality System (CAPS) "was proposed to reconcile paradoxical findings of the invariance of personality and the variability of behaviour across situations" (Mischel and Shoda, 1995, p. 256). They suggested five types of

Personality dispositions and personality processes

cognitive affective units in the personality mediating system, which resembled Mischelâ&#x20AC;&#x2122; s original cognitive personal variables. These units along with their definitions are given in the following table: Table 27. The five cognitive-affective units in the personality mediating system (adapted by Mischel and Shoda, 1995, p. 253)

AL LO

1. Encodings: categories (constructs) for the self, people, events and situations (external and internal). 2. Expectancies and beliefs: about the social world, about outcomes for behaviour in particular situations, about self-efficacy. 3. Affects: feelings, emotions, and affective responses (including physiological reactions). 4. Goals and values: desirable outcomes and affective states: aversive outcomes and affective states: goals, values, and life projects. 5. Competencies and self-regulatory plans: potential behaviours and scripts that one can, and plans and strategies for organising action and for affecting outcomes and oneâ&#x20AC;&#x2122;s own behaviour and internal states.

A major addition is the inclusion of affective states and emotions as a new set of

personal variables, which are now considered significant determinants of behaviour.

However, the most important contribution of this theory is the identification of these

five cognitive-affective units as psychological mediating processes underlying individual differences. These units are not seen as isolated, static components, but rather

as mental events which become activated within an individual in relation to certain

figures of the situation or of the self (Mischel and Shoda, 1998).

CAPS is seen by its pioneers as a bridge between personality processes, as described by social cognitive theory, and personality dispositions (Mischel & Shoda, 1995, 1998).

Recognising the significance of enduring individual differences in explaining behaviour,

Mischel and Shoda emphasised the importance of studying them along with processoriented approaches in personality research, specifically in relation to the distinctive organisation of relationships between the five cognitive-affective units. This interaction, they argue, will explain adequately intra-individual variation across situations and the different behavioural outcomes appearing in if...then relations. The main feature of this system is its dynamic nature, which takes into account both the dynamic change and stable structure of personality (Shoda and Mischel, 1996). Shoda and Mischel, (1996) also claim that, because of its dynamic nature, the theory will encompass how the

Personality dispositions and personality processes

individual's experiences and behaviours (i.e. then) - are meaningfully related to the psychological conditions in which they occur (i.e. if). The most important asset of this theory is the reconceptualisation of an individual's personality as not a simple aggregation of particular units, but rather turning the focus of attention to the organisation that underlies these units, into a network of interrelations that operate as a whole. This sentence describes the central idea behind the present performance via, and/or with other work-related variables.

attempt to use the CAPS model in explaining relationships between personality and job

It is not only the social cognitive theorists who recognise the need for reconciliation

AL LO

between dynamic personality processes and stable individual dispositions. A number of advocates of the dispositional approach have recently called for the demand to incorporate situational characteristics and processing-dynamic concepts into their models (e.g. Revelle, 1995). Two of the most influential supporters of the dispositional

approach and pioneers in the study of FFM, Paul Costa and Robert McCrae (1996) have

urged the development of a metatheoretical framework of personality theories where

dispositions, as expressed by the FFM, are only a part of a larger set of personality variables where the dynamic processes between them have the most important role, as

shown in the following figure and also described in table 28. Categories of personality variables as a metatheoretical framework for personality theories (adapted by McCrae & Costa, 1996, p. 73)

Figure 6.

Objective Biography

Basic tendencies

External influences

Characteristic adaptations

Self-concept

Note: All arrows represent dynamic processes

Personality dispositions and personality processes

Table 28. Examples of specific content in five categories of personality variables (adapted by McCrae & Costa, 1996)

Characteristic adaptations • Acquired competencies • Attitudes, beliefs, and goals • Learned behaviours • Interpersonal adaptations

Self concept • Implicit, explicit views of self • Self esteem • Identity • Life story, personal myth

Objective biography • Overt behaviour • Stream of consciousness • Life course

AL LO

Basic tendencies • Genetics • Physical characteristics • Cognitive capacities • Physiological drives • Focal vulnerabilities • Personality traits (FFM)

External influences • Developmental influences • Macroenvironment • Microenvironment

Big-Five theorists such as Costa and McCrae (1996) argue that sophisticated

personality psychologists have never claimed that traits determine behaviour

independently of situational context. However, this is probably the first time where the important role of dynamic processes between situational and personal variables is so

explicitly acknowledged by two of the strongest supporters of the FFM.

It is apparent then that the time has come for personality psychology to bring

together two of the most important streams and theoretical approaches of the field, social cognitive theory and trait approach into a unified framework. The view of the

person as a dynamic, goal directed individual, actively creating the situation he or she is

engaged with, who is loosely guided by a stable personality structure but is also able of controlling the influence of traits in relation to situational or other kind of demands, provides an exciting prospect in personality psychology, and most certainly in the explanation of the influence of personality at work settings.

Personality dispositions and personality processes

7.2. Using CAPS to explain links between personality and work related variables The five cognitive-affective units of the personality system may provide the theoretical link in explaining relationships between stable personality structure as described by the FFM and work-related variables and/or job performance, as illustrated in the following figure:

AL LO

Cognitive-affective units • Encodings • Expectancies and beliefs • Affects • Goals and values • Competencies and selfregulatory plans

The social-cognitive affective units and aspects of the personality-work related behaviour

Figure 7.

Job performance

Big-Five

Work-related variables • Job satisfaction • Organisational citizenship behaviour • Work competencies

The meta-analytic studies discussed earlier showed that some of the five personality dimensions (e.g. conscientiousness, emotional stability), may have a direct effect on

predicting job performance. Further, at least two of the five dimensions are valid

predictors of job satisfaction and organisational citizenship behaviour (neuroticism, extroversion and agreeableness, conscientiousness respectively), whereas it is also expected that some personality traits may be related to specific work competencies. In addition, it is suggested that the five cognitive-affective units of the CAPS are related to individuals’ stable personality structure and are also related to the work-related variables, both by having an effect on them and being influenced by them, as will be explained in the following paragraphs. Bringing together this model, in an attempt to

Personality dispositions and personality processes

improve the criterion-related validity of personality traits requires that all these units taken under consideration. In line with the Cognitive Affective Personality System, the present approach accepts that employees are not “hostages” of their traits, but active personalities. People may display stable personality characteristics, as expressed by the FFM or any other traitfactor analytic framework, but these characteristics do not fully explain behaviour.

Emergent behaviour is a complex function of stable personality dispositions, cognitive affective factors and situational variables.

AL LO

relationship between personality and work-related variables.

The following figure illustrates how the cognitive-affective units may explain the

Figure 8.

Personality, cognitive-affective units and work-related variables

Cognitive-affective units

Personality dimensions

Work-related variables

Social cognitive theorists, like Mischel and Shoda, suggest that individuals’ stable

personality structure interacts with other variables in the sphere of personality, such as

the cognitive-affective units (Shoda & Mischel, 1996). For instance, affective units such

as feelings or strong emotions could be expected to have a temporary or sometimes more permanent effect on individuals’ personality dispositions, e.g. as a result of the

loss of a beloved person. Further, an important life-career goal or beliefs about selfefficacy may exert a direct effect on levels of some personality dispositions, e.g. making an employee to exhibit more increased levels of conscientiousness or agreeableness than normal. However, the effect of stable personality dispositions will still be considerable, e.g. some dimensions of the FFM were found to predict self-efficacy in self-managed groups (Thoms et al., 1996), where a recent meta-analysis indicated a correlation of .38 between self-efficacy and performance at work (Stajkovic & Luthans, 1998).

Personality dispositions and personality processes

Moreover, an interaction is also expected between the cognitive-affective units and the work-related variables. For instance, employee’s encodings, i.e. his/her perception and interpretation of his/her working environment, including tasks, peers, supervisors, and subordinates, are expected to exert a direct influence on job satisfaction, but are also subject to change as a result of employees’ levels of satisfaction. Similarly, employees’ generic work competencies are also expected to interact with the cognitive-affective

units. Awareness of good leadership or communication skills could have a direct effect on the goals, values, expectancies and beliefs the person sets for him/herself.

AL LO

which is a rather narrower and situationally specific concept.

Nevertheless, as we will discuss later, this is not the case for citizenship behaviour,

The present attempt will start with job satisfaction, the single most examined variable in the organisational behaviour literature. Job satisfaction is, to an extent, an affective, cognitive or attitudinal response to work. It is related to a number of outcomes, apart from job performance, such as organisational and career commitment,

absenteeism, turnover intentions, etc. (Spector, 1997). As discussed earlier, a number of personality dispositions, such as extroversion and neuroticism, are expected to relate to

job satisfaction (Spector, 1997). Thus, a sociable, bold, assertive employee

(extroversion) or a calm, relaxed (emotional stability) employee is expected to display higher levels of job satisfaction compared to colleagues with personality characteristics

towards the opposite poles of the respective scales. Naturally, the influence of various

job characteristics, along with other situational variables also has a major effect on job

satisfaction and should be taken into serious consideration.

With respect to the Cognitive-Affective Personality System, four of the five cognitive-affective units (encodings, expectancies-beliefs, affects, goals-values) may

help to explain the personality-satisfaction relationship. As far as encodings are concerned, an event which occurs at work, or the interaction with another employee or supervisor at work may not have the same effect on levels of job satisfaction for everybody within the organisation. Some people may exhibit higher levels of satisfaction than others, simply because they may enjoy working with a particular colleague or under the management of a certain supervisor. Moreover, because of the dynamic relationship between encodings and satisfaction, a satisfied employee may perceive the same event or person at work differently from a colleague, who feels less

Personality dispositions and personality processes

satisfied. As regards the cognitive-affective unit of affect, a particular task may be favourable for someone but not for somebody else, influencing his or her satisfaction levels. In addition, building on the relationship between life and job satisfaction, each employee brings his or her emotions and feelings to work and these will have a major influence on levels of job satisfaction. Thus an employee who is not satisfied with his or her personal life may also feel dissatisfied at work. Similarly work satisfaction will

almost certainly have an impact on levels of general life satisfaction. As far as expectancies-beliefs are concerned, an employee may feel satisfied because

expectations are fulfilled within the work organisation, or because efforts are

acknowledged, or even because of increased levels of self-efficacy. As a result of the

AL LO

dynamic relationship between expectancies-beliefs and job satisfaction, the person may gradually alter those cognitive-affective units, if job satisfaction levels drop to a level not allowing him to function properly within the organisation. Finally, goals and values, are also related to job satisfaction. As was the case for expectancies, if employeesâ&#x20AC;&#x2122;

personal values are congruent with those of the organisation where they work, or if they

believe that at the current positions they can satisfy the goals they have set for

themselves, then it would not be surprising to find increased levels of job satisfaction.

As far as organisational citizenship behaviour is concerned, research evidence discussed earlier examining links with personality dispositions has shown that two

dimensions of the FFM, i.e. agreeableness and conscientiousness are related to the

personal and the impersonal dimensions of OCB, respectively. With respect to the

cognitive-affective units of personality, participation in extra-role activities, may be

regarded as both an affective and a cognitive response to work conditions. More specifically, it could be that employees engage in extra-role activities because

they have positive feelings about their colleagues or supervisor or their work overall (affective unit), or because they expect that the appreciation of these actions by their supervisors or colleagues will result in secondary benefits (e.g. change for promotion or rewards) capturing the expectancy-beliefs and goals-values cognitive units. OCB is different from job satisfaction and competencies, mainly because the relations between the cognitive-affective units and OCB are not two-dimensional, i.e. the engagement or not in extra-role activities would not be expected to induce any changes in the personal

100

Personality dispositions and personality processes

cognitive-affective units, since OCB is a narrower and very much situationallydependent concept, compared to job satisfaction, for example. As earlier noted, citizenship behaviour was initially studied in an effort to explain the low correlations usually found between satisfaction and performance. Its relationship with fairness and social exchange has also been discussed (Organ, 1988). Thus, an employee who feels satisfied at work will be keener to participate in extra-role

activities, compared to a less satisfied colleague, or if s/he is treated fairly s/he will also

be keener to engage in contextual performance. The same outcome will occur if they

believe that by engaging in these actions, they also increase the possibility of personal

AL LO

gains (e.g. promotion or pay raise). The cognitive affective units of affects, beliefs, and goals-values may assist in explaining citizenship behaviours.

Finally, work competencies is the last work-related variable examined here. According to the CAPS, competencies and self-regulatory plans refer to people's ability

to perform potential behaviours, plans and strategies, which lead to desirable outcomes.

Social cognitive theory suggests that these capacities are acquired through direct and

observational learning (Mischel, 1993). In work settings, competencies refer to employees' skills, which are necessary to successfully perform every-day tasks. The

acquisition of increased levels of these competencies most certainly will lead to improved work performance. Thus, an employee who is aware, for example, of his or

her ability to communicate or interact well with other people may use these skills in

order to increase performance, or may decide not to use them, if (in combination with

the expectancies-beliefs cognitive-affective unit) he or she believes that it is not worth

utilising them at work.

The role of work competencies is quite distinct from the work-related variables

discussed earlier (i.e. job satisfaction and organisational citizenship behaviour). They are directly related to effective work performance, but may also be subject to the

influence of the cognitive-affective units. Thus, an employee may posses a particular set of work competencies (e.g. leadership, or communication skills) but may hesitate in using them at work, either because of a belief that they can not help in the successful completion of tasks (encodings), or because they are not valued by his or her supervisor (values), thus not resulting in superior performance (expectancies) or because he or she

101

Personality dispositions and personality processes

does not know how to make appropriate use of them (self-regulatory plans, competencies). These behaviours emphasise once more the active role played by employees in selecting the situations they expose themselves, and the actions they decide or avoid to take at work. Concluding the section examining how the cognitive-affective units of personality may explain the relationships between personality dispositions and work-related

variables, it is necessary to emphasise that the cognitive-affective units are influenced by individual differences, as the one-dimensional arrows in figure 8 starting from the

personality dispositions clearly indicate. Employees set different goals, have different

AL LO

values, construe situations and people in various ways, and also show different feelings, as a result, to an extent, of their stable, internal, personality structure. For example, an agreeable person who generally has a positive attitude towards other people will usually show positive feelings towards them. A conscientious employee is also expected to set higher and more ambitious goals or have greater expectations than less conscientious

colleagues.

In the last section we will discuss how the cognitive-affective units of personality may explain moderating and mediating relationships of the work-related variables

between personality dispositions and job performance.

Moderating-mediating relationships and CAPS

7.3.

As already noted, research has identified links between specific personality

dispositions and performance at work. These relationships may be more meaningful if

they are discussed in the framework proposed by Mischel and Shoda (1998), and in relation to the moderating-mediating role of some work-related variables. Here the

discussion will concentrate on the variables already discussed (i.e. job satisfaction, citizenship behaviour, work competencies). However, several potential situational moderator variables could also be studied as Schneider and Hough (1995) proposed (e.g. task / goal / physical environment / work role / organisational characteristics). Returning to figure 8, first the case of job satisfaction will be discussed. It is suggested that the cognitive-affective units would be useful in explaining the hypothesis that job satisfaction moderates the relationship between two of the big-five personality dimensions (conscientiousness, and neuroticism) and job performance, whereas it

102

Personality dispositions and personality processes

mediates the relationship between agreeableness and job performance. As far as the former is concerned, it is expected that these two personality dimensions will be related to job proficiency across occupations in the present study as well, in line with the results of previous research. Nonetheless, it is claimed that this relationship will be stronger for increased levels of job satisfaction, i.e. the relationship between conscientiousness or emotional stability and overall job performance will be stronger for more satisfied

employees compared to less satisfied colleagues. Conscientious or emotionally stable

employees who feel satisfied at work will be more productive than less satisfied

colleagues, a relationship that could be attributed to the effects of the cognitive-affective units. For instance, a conscientious employee who may feel satisfied at work because

AL LO

s/he expects to receive recognition for his/her efforts, or because his/her personal goals and values are fulfilled in the position s/he is in, is expected to perform better than a similarly conscientious but less satisfied colleague. Thus, as will be discussed in the research questions section which follows, it is expected that job satisfaction will

moderate the relationship between conscientiousness or emotional stability and job

performance. On the other hand, agreeableness is not expected to have a direct effect on

job performance. The influence of agreeableness will be transmitted to overall job

performance via job satisfaction, or specific citizenship behaviours or work competencies, not directly, and only for occupations involving interpersonal interaction

(Mount et al., 1998). Job satisfaction, citizenship or competencies will mediate thus the

occupations.

relationship between agreeableness and overall job performance, for this type of

As far as organisational citizenship behaviour is concerned, it may act either as a

moderator or a mediator in the personality-job performance relationship. The

impersonal dimension of OCB is expected to moderate the relationship between conscientiousness and work performance, whereas the personal dimension of OCB to mediate the relationship between agreeableness and job performance. A conscientious employee who is rated by his/her supervisor as participating in extra-role activities captured by the impersonal dimension of OCB, is expected to receive higher ratings of overall job performance compared to a similarly conscientious colleague who prefers not to engage in extra-role activities described by the OCB-conscientiousness dimension. As earlier discussed, employees may participate in extra-role activities both

103

Personality dispositions and personality processes

as a result of their affective responses towards other people at work, or because they expect that these actions will initiate secondary benefits, which they approve. Subsequently, engaging in these activities should improve the validity of supervisory ratings for this group of employees. Thus, participation in extra-role activities captured by

the

OCB-conscientiousness

dimension

expected

moderate

the

conscientiousness-performance relationship. As far as agreeableness is concerned, it is

suggested that, similarly to job satisfaction, the personal dimension of OCB will

mediate the validity of this personality dimension for occupations involving

interpersonal interactions. It is expected that this occurs due to the relationship between agreeableness and OCB-altruism first and the link of the latter with job performance.

AL LO

Agreeable employees, partially because of the effect of the cognitive-affective units, such as affects, and expectancies-beliefs, would be keener to participate in extra-role activities receiving improved ratings of overall job performance, as a result of their participation. Further, irrespective of employees' personality characteristics, a satisfied

employee will be keener to participate in extra-role activities. A satisfied employee who

also believes that engaging in contextual performance will lead to desirable outcomes,

such as promotion or increased salary, will perform better at work. These are conscious

steps the employee is taking in order to improve conditions at work. He or she is actually in a position to actively change the situation s/he is into by taking such actions.

As discussed earlier, competencies are work-related abilities the employees may

actively use in order to improve their performance. It is more likely that the

combination of specific personality dispositions with competencies will lead to the adoption of different strategies and problem solving approaches, rather than actual

direct changes in job performance levels (Robertson & Fairweather, 1998; Robertson &

Callinan, 1998). For example, an extrovert employee who has the self-confidence to lead colleagues may successfully complete tasks, following different plans and strategies than a less extroverted colleague, who has similar skills. He or she may try harder to involve other members of the team in the group processes and discussions, or may dominate the group with his or her actions and suggestions more than the less extrovert colleague.

104

Personality dispositions and personality processes

7.4.

Conclusions

The present chapter serves two purposes. First, to briefly describe a theoretical approach on personality dispositions developed recently under the guidance of a very influential theory in mainstream psychology such as the social-cognitive theory, and second to serve as an introduction and a yardstick to the research questions described in the following chapter.

The introduction of the Cognitive Affective Personality System in the exploration of relationships in occupational settings could be very useful for a number of reasons. As

far as the criterion-related validity of personality traits is concerned, the CAPS takes

AL LO

into account both stable personality dimensions, as expressed by the FFM, and the moderating/mediating role of the cognitive-affective units of personality, which may explain the effect of a number of work-related variables identified at workplace, such as job satisfaction, citizenship behaviour and work competencies. Here it was suggested that the interactions occurring between these variables may account for aspects of work

behaviour, building on the notion that we can not discount personality traits but also that they are definitely not enough in explaining employees' behaviour, beyond a limited

percentage of variance. Employees, just like every other person, are not passive

witnesses to either situations or personality dispositions. They take actions to modify their position in their environment, whether this is the work environment or the social

environment. The five cognitive-affective units of personality, as described by Mischel

and Shoda, can prove a very useful taxonomy in explaining relationships between

personality dimensions and traditional work-related outcomes such as job performance and satisfaction. This type of approach may prove very useful both for research in the

field of personality and for practice in personnel selection and assessment, and an

attempt is made to be incorporated in the present study, both in the examination of the research hypotheses described in the following section, and also in the discussion of the study results.

105

Research questions

Chapter 8. Research questions The potential adoption of the Cognitive-Affective Personality System in the examination of the criterion-related validity of trait-factor-analytic personality tests may represent a “no-miss” opportunity for personality psychology. The use of the cognitiveaffective units as a guide in explaining moderating-mediating effects of individual

differences to behavioural outcomes, such as job proficiency, may prove very useful. For example, the influence of an affective reaction such as job satisfaction, and the

impact it may have on job proficiency either independently or interacting with

personality dispositions is examined here. Additionally, the value of competencies is

AL LO

also studied in the form of generic work competencies possessed by employees and assessed by their supervisors. But further than those two cognitive-affective units, which are specifically assessed in the present research, the remaining personal variables may also be useful as a theoretical guide and framework in explaining the role of the

five-factor model in work settings.

This approach, which is followed in the present study, accepts that people are not

“hostages” of their traits, but active personalities, that certainly have dispositional,

stable personality characteristics as expressed by the FFM, but also have the ability either consciously or unconsciously to change their behaviour, simply because they

want to, like to, or have to. Especially, in contemporary competitive work settings

where employees are continuously both under assessment from their employers and

under the effect of their peers, change is usually inevitable. People have to change their attitudes, their approach at work and sometimes even their personal characteristics, in

order to adapt to the changing circumstances of today’s occupations.

8.1.

Demographic characteristics

The data examination will begin by exploring any gender or age differences with

respect to the variables under examination. In line with the gender differences found in the normative sample of the personality questionnaire (Tsaousis, 1996), it is similarly expected that females will score significantly higher than males in four of the five personality

dimensions,

i.e.

neuroticism,

openness,

agreeableness

and

conscientiousness. As far as job satisfaction is concerned, no gender differences are

106

Research questions

expected since, as Spector (1997) claimed, inconsistencies across studies regarding this issue suggest that males and females show similar levels of job satisfaction. Further, no gender or age differences are expected regarding the performance measures, in line with viewpoints reported in recent reviews on the performance appraisal issue (Arvey & Murphy, 1998). Finally, although no age differences are expected in the personality dimensions, age is expected to correlate positively with job satisfaction, although

Personality dimensions predicting work-related outcomes

8.2.

researchers disagree whether this relationship is linear or curvilinear (Spector, 1997)

In line with meta-analyses, which found similar results (Salgado, 1997; Barrick &

AL LO

Mount, 1991) two of the five personality dimensions, namely conscientiousness and neuroticism, are expected to be valid predictors of job performance across all types of occupations. Following the results of the recent meta-analysis by Mount et al. (1998) it is expected that extraversion and agreeableness will be valid predictors of job

proficiency only in occupations involving interaction with other people.

Nevertheless we suggest that agreeableness will have an effect on job proficiency

through the mediating effects of job satisfaction, citizenship behaviour, and work

competencies.

According to socio-analytic theory, conscientiousness is a part of identity choice, an

interpersonal tactic for dealing with the members of a group (Hogan and Ones, 1997).

An employee is likely to repeat activities that bring approval from the authority.

Following that, in a work setting, a supervisor who gives credit to a conscientious employee, or an employee who watches his conscientious colleagues taking bonuses

because of their competence, order, or other similar characteristics, will engage herself

in similar activities. Thus the existence of a disposition like conscientiousness, which incorporates characteristics such as competence, order, dutifulness, achievement striving, self-discipline, and deliberation, is not simply an advantage for the employee but a requirement for any job, in any work setting. Conscientious employees are people that have self-control, and take action to plan, organise and carry out their every day tasks. Nevertheless, a moderate relationship is expected, since people scoring very high in conscientiousness are not expected to receive very favourable ratings from their supervisors. The reason is that a very conscientious employee may spend more time

107

Research questions

than necessary in carrying out every day tasks with extreme care or putting everything in order rather than getting on with the actual job. Subsequently, it is expected that a curvilinear relationship exists between conscientiousness and job performance, i.e. that employees with very high or even extreme scores in conscientiousness will receive less favourable ratings from their supervisors. Neuroticism represents the proneness of the individual to experience unpleasant and

disturbing emotions and to have corresponding disturbances in thoughts and actions

(Vestre, 1984, cited in McCrae and Costa, 1990). People scoring high in neuroticism

have feelings of insecurity, anxiety, and stress more frequently than other people. Such

AL LO

feelings usually prevent employees from operating at the desired level within their work environment. Subsequently, their performance drops the more stressful they feel, or the higher score they have in neuroticism. An emotionally stable employee will not be distracted at work; will keep his/her mind together, thus leading to effective performance. Therefore it is expected that neuroticism will correlate negatively with job

proficiency.

Extroversion, in line with the meta-analytic results of Barrick and Mount (1991), is expected to be a valid predictor of job proficiency for occupations with interpersonal

requirements, such as managers, salesmen, teachers, or personnel officers (Salgado, 1997). People who score high on this scale are sociable, enjoy being with and talking to

other people, and are regarded as hearty and enjoyable people. Thus these

characteristics will be very useful in occupations involving interaction with other

people, although it will be less useful or even destructive in occupations where the

employee has to work on his/her own.

Finally, agreeableness according to Costa & McCrae (1985) assesses the quality of

oneâ&#x20AC;&#x2122;s inter-personal orientation along a continuum from compassion to antagonism in thoughts, feelings and emotions. Thus, high scorers are described by adjectives such as sympathetic, kind, appreciative, co-operative, trusting, helpful, where low scorers are described as cynical, cruel, rude, cold, unfriendly, etc. Nevertheless, if a relationship exists between agreeableness and job proficiency it will be due to the mediating effects of either one or some combination of job satisfaction, citizenship behaviour or generic work competencies. An agreeable employee may receive favourable ratings, but this

108

Research questions

will not be due to the direct effect of the personality trait, but more likely due to the higher scores in those variables. Similarly to extroversion, agreeableness will have this kind of effect only in occupations involving some kind of interpersonal interaction, such as teachers and managers. As far as job satisfaction is concerned, in line with the literature discussed earlier about the relationship between personality and job satisfaction, it is suggested that

extroversion and agreeableness will correlate positively and neuroticism negatively with job satisfaction, because of their relationship with positive and negative

affectivity respectively. It is also foreseen that the relationship of conscientiousness

AL LO

and neuroticism with job proficiency will be moderated by job satisfaction. The definition of a moderator variable in the introduction suggested that a variable z is a moderator of the relationship between a predictor x and a criterion y when the nature or magnitude of this relationship varies across the different levels of the moderator (Aguinis & Stone-Romero, 1997), or when interacting with the predictor explains

additional variance in scores of the dependent variable (Stone-Romero & Anderson, 1994). In other words, the relationship between conscientiousness and neuroticism with

job proficiency will be different for different levels of job satisfaction. It is expected

that for more satisfied employees the magnitude of the relationship between conscientiousness or emotional stability with overall job performance will be stronger

compared to less satisfied employees. As discussed in the last chapter, conscientious or

emotionally stable employees who feel satisfied as a result of the cognitive-affective

units, are expected to perform better than less satisfied colleagues, who have similar scores on conscientiousness or emotional stability. In other words, job satisfaction is

expected to play a significant moderating role in the relationship between

conscientiousness or emotional stability and overall job performance. Finally, as earlier suggested, a hypothesis is made that job satisfaction mediates the relationship between agreeableness and job performance. It is suggested that agreeable

employees will feel more satisfied from their work and consequently because of that they will be more productive and not because they are agreeable. With respect to organisational citizenship behaviour, it is expected that the two dimensions of the OCB measure used in this study, altruism and conscientiousness or generalised compliance, will be predicted by the personality dimensions of

109

Research questions

agreeableness and conscientiousness respectively, in line with the results of previous studies (Organ, 1994; Organ & Ryan, 1995). It is also anticipated that the conscientiousness dimension of OCB will moderate the relationship between the conscientiousness personality dimension and supervisory ratings of job proficiency. For employees who receive favourable ratings from their supervisors on the OCB-conscientiousness dimension, the relationship between conscientiousness and

job proficiency is expected to be stronger compared to the employees who are rated as

not participating in extra-role activities. Due to their participation in extra-role

activities, conscientious employees will receive more favourable ratings compared to

colleagues with similar levels of conscientiousness who prefer not to engage in

AL LO

citizenship behaviours. Similarly, as noted earlier, it is expected that the altruism dimension of OCB will mediate the link between agreeableness and job proficiency. Since it is expected that altruism would relate to job proficiency and would also be predicted by agreeableness, these two relationships will provide agreeable

employees with higher ratings, and not the effect of agreeableness itself, as was the case

earlier for the mediating effect of job satisfaction to the agreeableness-job performance

relationship. Finally, following Organ and Linglâ&#x20AC;&#x2122;s (1995) suggestion, it will be

examined whether job satisfaction contribute to OCB once the effects of conscientiousness and agreeableness have been accounted, anticipating that job

satisfaction will contribute even further to the prediction of OCB.

Some of the generic work competencies are also expected to correlate with some of

the personality dimensions. More specifically, it is anticipated that action-leadership competencies will be predicted by a combination of the personality dimensions of

conscientiousness with extroversion. Thus a competent, striving to achieve employee

(the conscientiousness part), who is also assertive, dominant and active (extroversion components), will be expected to demonstrate competencies captured by the actionleadership dimension of the Generic Work Competencies Questionnaire, used in the present study. As discussed later in the development of the Generic Work Competencies Questionnaire, employees with this type of competencies are leading figures in their work environment inspiring, motivating and leading their colleagues. Further the personality dimensions of conscientiousness and extroversion are, according to Robertson and Callinan (1998), two of the five-factor dimensions usually related to

110

Research questions

leadership effectiveness. Another type of competencies, orientation towards other people, is expected to correlate to the agreeableness dimension of personality since by definition they share common variance. Employees, who are positively orientated towards colleagues, rating themselves high in agreeableness, should also possess people orientation competencies. A similar relationship is not expected between extroversion and people orientation competencies, since, although extroversion assesses interpersonal

interaction, it is more related to the quantity and intensity of this interaction rather than the quality of it, as is the case for agreeableness. Thus, it is not necessarily true that a

person scoring high in extroversion will also demonstrate positive feelings or show respect and consideration towards other people. Also, people orientation competencies

AL LO

are expected to mediate the relationship between agreeableness and job proficiency. This type of competencies brings the ability factor within the agreeableness-performance relationship. If employees are keen to demonstrate these competencies at work, then they should expect to receive favourable ratings from their

supervisors in occupations involving interpersonal interactions. If they do not have or

do not want to demonstrate these competencies, no relationship should be expected

between agreeableness and job performance. Finally, neuroticism is expected to

correlate negatively with the third type of competencies measured by the questionnaire, communication skills. An employee who is usually nervous, tense and

anxious will have difficulty to use any communication skills s/he might have in an

efficient and effective way. Similarly, to the people orientation competencies, a

relationship between extroversion and communication skills is not expected, since they mainly capture appropriate language skills and effective communication proficiency,

rather than how talkative or energetic a person is, which are characteristics assessed by

extroversion.

8.3.

Relationships between the work-related variables

Using as a starting point job satisfaction, and following the general trend of the

relevant literature (e.g. Spector, 1997; Iaffaldano & Muchinsky, 1985), it is expected that a positive correlation will exist between job satisfaction and supervisory ratings of job proficiency. Without reaching into an agreement concerning the causal links of this relationship, most studies examining this question find a positive, although

111

Research questions

often quite weak, relationship. Further, according to most studies in the OCB literature, job satisfaction is a stronger predictor of extra-role activities than in-role performance, thus a positive relationship is expected between both dimensions of OCB and job satisfaction, stronger than the one between job satisfaction and job performance. In addition, it is suggested that OCB will increase the percentage of variance explained in job proficiency, above and beyond the effect of job satisfaction. Citizenship

behaviour will have an incremental effect on ratings of overall job performance, beyond the effect of job satisfaction. Employees who participate in extra-role activities are

doing so because they are satisfied with their work, in addition to any other reasons (e.g. personality), resulting in more favourable overall performance ratings from their

AL LO

supervisors. The additive effect of job satisfaction to citizenship behaviour will lead to improved performance ratings.

The last set of hypotheses concerns the performance measures of the study. Although it is expected that the various dimensions assessed will be correlated to a

statistically significant level, it is also anticipated that contextual performance and the three dimensions of the competencies measure will be valid predictors of job

performance. If thatâ&#x20AC;&#x2122;s true, the viewpoint that supervisors can take into account both

the skills and citizenship behaviours of their subordinates during their performance appraisals, being in a position to differentiate between those different types of work-

Summary of hypotheses:

8.4.

related outcomes, will be justified.

1. Conscientiousness and neuroticism will be valid predictors of job performance across occupations.

2. Extroversion and agreeableness will be valid predictors of job proficiency for in occupations involving interaction with other people. 3. A curvilinear relationship exists between conscientiousness and job performance. 4. Extroversion and agreeableness will correlate positively with job satisfaction. 5. Neuroticism will correlate negatively with job satisfaction. 6. The relationship of conscientiousness with job proficiency will be moderated by job satisfaction. 7. The relationship of neuroticism with job proficiency will be moderated by job satisfaction.

112

Research questions

8. Job satisfaction mediates the relationship between agreeableness and job performance. 9. The OCB dimensions of altruism and conscientiousness will be predicted by the personality dimensions of agreeableness and conscientiousness respectively. 10. The conscientiousness dimension of OCB will moderate the relationship between the conscientiousness personality dimension and supervisory ratings of job proficiency.

11. The altruism dimension of OCB will mediate the link between agreeableness and job proficiency.

12. Job satisfaction contributes to OCB once the effects of conscientiousness and agreeableness have been accounted.

AL LO

13. Action-leadership competencies will be predicted by a combination of the personality dimensions of conscientiousness and extroversion. 14. People orientation competencies are expected to correlate to the agreeableness dimension of personality but not to extroversion. 15. People orientation competencies agreeableness and job proficiency.

will

mediate

the

relationship

between

16. Neuroticism is expected to correlate negatively with communication skills.

17. A positive correlation will exist between job satisfaction and supervisory ratings of job proficiency.

18. A positive relationship is expected between both dimensions of OCB and job satisfaction.

19. OCB will increase the percentage of job proficiency variance explained above and beyond the effect of job satisfaction.

20. It is expected that all the performance dimensions assessed by the supervisors will be highly intercorrelated.

21. Organisational citizenship behaviour and the three dimensions of the competencies measure will be valid predictors of job performance.

113

Method and pilot study

Chapter 9. Method and pilot study The research design adopted in the present study was that of a concurrent criterionrelated validity. The data file consisted of the responses of employees and their immediate supervisors. The employees completed two measures; a personality inventory and a job satisfaction measure and their supervisors rated their performance in

three independent measures; one questionnaire assessing organisational citizenship behaviour, a short job proficiency measure and a questionnaire measuring generic work

competencies, specifically developed for the purposes of the current study. The following sections will describe the procedure followed in the data collection and the

9.1.

AL LO

measures used.

Procedure

The present attempt started at the end of 1996 with an invitation to Greek

organisations to participate in this research project. Using Greek and European business

guides, all the major Greek and multinational companies operating in Greece (with

workforce of over 1000 employees), along with a few dozen smaller organisations (with more than 100 employees), were conducted through a letter sent to their Human

Resource managers describing the purpose of the study and inviting them to participate. A couple dozen of those organisations showed interest and a number of meetings were

held with their HR managers describing the details of the research and how their

organisations could profit from a partnership of this kind, a quite difficult task, since

organisations in Greece are not used to participate in these kind of research studies.

After these meetings, the number of firms which agreed to participate consisted the database for the first and major part of the data collection carried out between April and

August 1997. Unfortunately, the total sample emerged from the first phase was not adequate for the purposes of the present study, and a similar procedure was followed at the end of 1997 resulting in the second phase of the data collection carried out in February 1998 providing the remaining necessary data of the research. In most organisations the distribution and collection of the material was arranged centrally through the HR department following a structured, confidential procedure, although in a few small firms I had to distribute and collect the questionnaires myself.

114

Method and pilot study

The questionnaires were distributed to the participants through the following procedure. In the case of employees, envelopes were handed out containing a cover letter explaining the purposes of the study and requesting them to complete a few personal details, such as their name, age, position in the organisation, and tenure in current position. The procedure was not anonymous since it was essential to know the name of the employee in order to identify his/her immediate supervisor who would assess them

and also to provide them with the feedback of their responses. The envelope also

contained the personality questionnaire, an answer sheet and the job satisfaction

measure. The Greek versions of the job satisfaction measure along with the cover letter accompanying the personality and satisfaction measures are presented in appendix 6. As

AL LO

soon as employeesâ&#x20AC;&#x2122; questionnaires were returned, their immediate supervisorsâ&#x20AC;&#x2122; were identified and handed out an envelope containing a cover letter explaining what they had to do, the names of their subordinates they had to rate and the three questionnaires: the organisational citizenship questionnaire, the job proficiency measure and the generic

work competencies questionnaire (Appendix 7). In both cases, the questionnaires were

either returned directly to me or, as it was usually the case, they were returned in a

sealed envelope provided to a person responsible in the HR department whom I was in

contact with.

Both employees and the supervisors received feedback of their responses. Each

participant employee received a one-page summary report indicating his/her individual

scores on the five personality factors assessed by the personality inventory and their

levels of satisfaction (intrinsic, extrinsic, and total), compared to the whole sample of employees participating in the study. The supervisors received a one-page summary

report for each employee they rated, indicating the scores of each of their employees in

the dimensions assessed by the questionnaires they completed, i.e. the two dimensions of the organisational citizenship questionnaire, the job proficiency index and the scores on the three types of competencies captured by the generic work competencies questionnaire, compared to the whole sample of employees. In addition they received one page showing the group satisfaction levels of their subordinates. Similar feedback was returned to the HR departments for all supervisory ratings. Neither supervisors nor HR departments received any feedback regarding employeesâ&#x20AC;&#x2122; responses to the personality questionnaires.

115

Method and pilot study

9.2.

Measures

As noted already the employees participating in the study completed a personality inventory and a job satisfaction measure, whereas their supervisors rated their performance in three measures; an organisational citizenship questionnaire, a job proficiency index and a measure of generic work competencies. In order to receive feedback and also for purposes of identifying their respective supervisor, employees

were asked to write down their name, and position in the organisation. They were asked to record their age in one of five-year age bands from 20-30 to over 61, and also how

long they have been in their current position with their organisation (tenure).

AL LO

9.2.1 The personality questionnaire

The personality questionnaire used in this study is a normative measure called TPQue (Traits Personality Questionnaire) (Tsaousis, 1999; 1996). It was developed based on Costa and McCrae’s definitions of the most acceptable factors in the five-

factor theory, taking into account, simultaneously, the specific ethnic and cultural

characteristics of the Greek population. It consists of 206 statements and the

respondents have to indicate their reply in a 5-point from “strongly agree” to “strongly disagree”. 180 of the items measure the five dimensions and six facets per dimension,

i.e. 30 subscales, and 26 items measure social desirability, a lie scale and a validity check. The choice of this instrument was straightforward since it is the only personality

inventory used in Greece originally developed in Greek language, without being a

simple translation of American or British instruments. The psychometric properties of

the instrument were very satisfactory with very good internal consistency, factorial

structure, and construct validity. The test-retest reliabilities of the five dimensions ranged from .89 to .95 and the internal consistency (Cronbach’ s α) from .78 to .89,

supporting the existence of the Big-Five in Greek language and justifying the use of the TPQue as a valid measure of the model. Sample items for one facet for each of the five dimensions are presented in appendix 1. The scoring procedure of the instrument was based on instructions given by the author (Tsaousis, 1996). Initially, a validity check was carried out in order to examine any nay-saying or acquiescence effects. None of the respondents had more than 4 inconsistent pairs of answers which is the cut-off point for validity check (Tsaousis,

116

Method and pilot study

1996). Subsequently, the scores on lie and social desirability scales were calculated for each participant. Although the author (Tsaousis, 1996) has indicated cut-off scores for those two scales as well, these were not taken into account since they were based on the student-normative sample of the TPQue, and, as shown in the following table, the mean scores of the normative sample for those two scales are lower than the mean scores of the present sample, indicating the existence of a significant impression management

effect.

Present study (N = 227) Mean SD 1.12 1.11 6.28 2.59 5.07 2.65

AL LO

Validity check Lie scale Social desirability

Normative sample (N = 1054) Mean SD 1.54 1.24 4.67 2.45 3.55 2.26

Table 29. Comparisons of the validity check, lie scale and social desirability scores between the normative sample of TPQue and the present study.

Writers on the use of personality tests at work have extensively discussed the effects

of social desirability (e.g. Barrick & Mount, 1996; Ones et al., 1996; Ones &

Viswesvaran, 1998; Hough, 1998) concluding that although its consequences on hiring

decisions may be significant (Rosse et al., 1998), it does not have a significant effect on

the scores of the personality scales (Barrick & Mount, 1996), which were the main focus of the present study.

9.2.2 The job satisfaction measure

Since the personality measure was quite long, taking thirty to forty minutes to

complete, the satisfaction measure had to be a short straightforward instrument, with

acceptable psychometric properties however. The instrument developed by Warr and his colleagues (Warr et al., 1979) was considered appropriate since it was short with

acceptable internal consistency and strong empirical support (e.g. Clegg & Wall, 1981). It consists of fifteen items where the respondents have to indicate their responses to a seven-point scale from “extremely dissatisfied” to “extremely satisfied”, measuring intrinsic and extrinsic job satisfaction; additionally by averaging the scores of the items, a total job satisfaction score is obtained, which was used in the present study. The job satisfaction scale is shown in appendix 2. The procedure followed in order to translate the instrument will be described later, in the section presenting the development of the

117

Method and pilot study

generic work competencies questionnaire. The same procedure was followed for the performance measures (OCB and job proficiency index) as well. 9.2.3 The OCB measure Smith et al. (1983) first developed the measure of organisational citizenship behaviour used in this study. It consists of 16 items, where employee’s immediate supervisor has to rate him/her in a five-point scale indicating the number that best

describes how characteristic this behaviour is of the employee. The five-point scale

ranged from “very characteristic” to “not at all characteristic”. Subsequent factor

analysis with both varimax and oblimin rotation identified two dimensions the authors

AL LO

called altruism and generalised compliance or conscientiousness, which had adequate internal consistency. Organ and Konovsky (1989) and Becker and Randall (1994) found similar results. The items of the OCB measure are presented in appendix 3. 9.2.4 The job proficiency index

The second measure the supervisors had to assess their subordinates on was the

measure used as an overall job performance score. As it was the case for all

performance measures, it had to be a short measure with acceptable psychometric

properties, since in most cases each supervisor had to rate quite a few subordinates. The present measure was successfully used in a number of studies (e.g. Robertson &

Gibbons, 1996; Robertson et al., 1997) as an overall job performance score eliciting

internal consistency reliability of .86. It consists of six items where the supervisor has to

indicate whether she/he agrees or disagrees with the behaviour described in a five-point scale. An overall job performance score was calculated by averaging the raw scores of

the six items. The items of this questionnaire are presented in appendix 4.

9.2.5 The Generic Work Competencies Questionnaire (GWCQ) A thorough examination of the literature concerning work competencies could not

identify an instrument measuring generic work competencies across occupations and job level. All of the researchers that had worked on the competencies field used organisation-specific

instruments

sometimes

developed

after

job-analysis.

Unfortunately, it was impossible for the purposes of the pilot study to conduct proper job analyses because of both the variations of businesses and jobs involved, and time

118

Method and pilot study

constraints. Subsequently, it was decided to design the development and construction of the GWCQ having in mind the following two basic rules: 1. It should be suitable for use across occupations in a white-collar level. 2. The items should refer to the ratee’ s behaviour and not to his/her traits, feelings or thoughts (Kline, 1979) who also argued that when one wants to know how items have to be made to refer to behaviour as far as possible.

The various stages in the development of the GWCQ are following:

subjects are placed on a dimension, rather than how they think they are placed,

AL LO

1st stage

In chapter 6 the work competencies literature was reviewed and the results of most studies related to this subject were reported. Various instruments and approaches were discussed, capturing different dimensions of work competencies but as already said,

almost all of them were organisation-specific. Thus, a detailed review of all that

dimensions between them.

literature was conducted in order to help us identify, at first, a number of common

Initially, three clusters of competencies were easily identified across most of these studies. These were the leadership orientation, the human-relations (people) orientation,

and the communication skills orientation. Boyatzis (1982) described the leadership and

human resource management clusters, where Dulewicz (1989) identified, as already

mentioned, 12 “supra-competencies”, such as planning and organising, strategic perspective, managing staff, interpersonal sensitivity and oral communication between

others. Similarly, SHL’ s Inventory of Management Competencies measures the

following competencies along with others: leadership, planning and organising, interpersonal sensitivity, resilience and oral-written communication. Thus it is apparent that at least three of the competencies clusters are repeated across instruments and studies. Those are the leadership, people and communication skills clusters. Further, it was decided to include two more clusters of competencies. These were the goal-organisation orientation and the action orientation. It was considered that their inclusion would be useful since they both incorporate competencies that someone would expect to associate with superior job performance and at the same time to apply across

119

Method and pilot study

occupations and organisations in a generic level. In addition, a few of the researchers that have published on the competencies field identified these clusters between their competencies list; e.g. Woodruffe (1993) pointed out the goal-orientation competency, or Dulewicz (1989) the results-orientation competencies, and also IMCâ&#x20AC;&#x2122;s action orientation competencies cluster.

and are presented in the following table along with their definitions:

Table 30. The five competencies clusters initially identified

These five cluster-orientations consisted the initial pool of dimensions of the GWCQ

AL LO

Goal - organisation orientation In the goal-organisation orientation cluster, the employee attempts to make progress towards a goal or according to a plan. S/he is aware of the objectives and works towards them in a methodical and systematic way, making decisions based on thorough examination of the features of a situation. At the same time, s/he remains focused and is aware of what progress is being made. Action orientation In the action orientation cluster the employee is energetic, dominating his/her surrounding with his/her activities. S/he is looking for stimulating tasks being active most of the time. S/he is a proactive employee who takes initiative at work. Leadership orientation In the leadership orientation cluster, the employee influences other people and takes the role of the leader in group activities. S/he motivates and inspires his/her colleagues, showing decisiveness and persuasiveness. In addition, s/he demonstrates adaptability and selfconfidence when circumstances change. People orientation In the people orientation cluster, the employee is sensitively and positively orientated towards his/her colleagues, taking into account their opinions and suggestions. S/he respects her/his colleagues, showing consideration for their problems. Communication skills orientation In the communication skills cluster the employee has good oral and written communication abilities. In the case of oral presentations s/he communicates in a simple and comprehensible way, clearly and fluently. S/he writes in clear and concise way, using correct grammar, style and language.

2nd stage

Subsequently, a dozen items for each one of these categories was developed. Thus 60 initial pool of items were developed. During the development of the items a number of facts were taken into account, mainly that these items could be applied into different levels within an organisation, and that they should be task and not behaviour-orientated.

120

Method and pilot study

3rd stage At this stage a review by experts was considered necessary in order to evaluate the items. Eight experts - 4 occupational psychologists and 4 managers - who were familiar with the work competencies literature and the applications of competence movement at work were asked to assign each one of the items to the category they thought it belonged to. A methodology widely used in qualitative research was employed; itâ&#x20AC;&#x2122;s

processes and heeded information from behaviour (Green, 1995).

called verbal protocol analysis and is mainly used as a way of inferring thought

More specifically, the five categories of the work competencies along with their

AL LO

definitions were shown to each expert independently and then they had to assign each one of the items, presented to them written on a card, to the category they thought it belonged to. They were encouraged to think aloud when they were doing this task. Their verbalisations were taken under consideration in the following stages for the

incorporation of the items in the questionnaire.

From the following analysis, only the items where all the raters unanimously agreed

that they belonged to the specified orientation of competencies, and those where only

one rater had disagreed about its cluster were kept for inclusion in the final version of the questionnaire. Thus the final version of the GWCQ included 51 items at this stage,

which are presented in the following table:

Table 31. Clusters and items of the competencies measure used in the UK pilot

1. Goal - organisation orientation

This employee: 1. maintains awareness of the degree of the task accomplishment 2. remains focused on his/her tasks 3. examines an issue thoroughly before reaching a decision 4. follows the rules and principles of the organisation 5. works with proficiency 6. follows tasks through to completion 7. attempts to reach the required standards of his/her position 8. accomplishes his/her duties in a systematic way 9. organises in advance his/her plans to achieve goals (table continued)

121

Method and pilot study

AL LO

2. Action orientation

This employee: 1. shows energy during the accomplishment of his/her tasks 2. responds with spontaneity to new conditions 3. behaves dynamically at work 4. is proactive (i.e. spontaneously takes action in order to achieve something at work) 5. looks for stimulation at work 6. is active most of the time 7. is willing to commit himself/herself to new tasks 8. looks for new activities within his/her work environment 9. remains lively throughout his/her time at work This employee: 1. shows decisiveness in decision-making situations 2. takes the role of the leader in group activities 3. is capable of persuading his/her colleagues 4. remains calm in difficult and stressful situations 5. influences his/her colleagues 6. motivates his/her colleagues at work 7. leads his/her colleagues effectively 8. inspires his/her colleagues 9. "shows the way" to his/her colleagues 10. directs his/her colleagues positively This employee: 1. shows positive feelings towards his/her colleagues 2. interacts with other people in a sensitive way 3. has good relations with most of his/her colleagues 4. s/he tries to help her/his colleagues with their problems 5. takes into account the suggestions of his/her colleagues 6. respects his /her colleagues 7. cares about the problems of his/her colleagues 8. will do some extra-work in order to help his/her colleagues accomplish their tasks 9. displays kindness towards his/her colleagues 10. shows consideration for his/her colleagues 11. shows friendly behaviour within the organisation This employee: 1. speaks clearly 2. delivers effective oral presentations 3. uses correct grammar in writing 4. speaks fluently 5. uses an appropriate style in writing 6. uses suitable language both in writing and speaking 7. adapts a different style of language depending on the target audience 8. covers the main points of an issue (e.g. during an oral presentation) 9. does not include irrelevant material both in writing and speaking 10. uses correct spelling in writing 11. presents information in a well-organised and logical way 12. demonstrates clarity of thought both in writing and speaking

4. People orientation

3. Leadership orientation

5. Communication skills orientation

122

Method and pilot study

4th stage At this stage an appropriate rating scale was necessary to be developed. There was much concern and speculation about the side effects of performance appraisals, such as leniency-severity effects, halo effect, and central tendency errors. Subsequently, our effort was to develop a rating scale which would combine various methods already used

in performance appraisals, such as BARS, BOS, and employee comparison methods. Thus, the rating scale developed was an 8-point scale from 7 to 0 where each point of

the scale was showing the position of the employee in comparison with a percentage of

his/her colleagues in the same role in the organisation. The exact percentage points

AL LO

assigned in each point of the scale were based on the normal distribution cut-points. The actual rating scale is following in the table; it was initially shown with the instructions in the cover page of the questionnaire and was also presented as a graphical

representation next to the list of the items.

Table 32. The rating scale of the competencies measure

More than 98% of his/her colleagues in the same role in the organisation

More than 84% of his/her colleagues in the same role in the organisation

More than 67% of his/her colleagues in the same role in the organisation

More than 50% of his/her colleagues in the same role in the organisation

More than 33% of his/her colleagues in the same role in the organisation

More than 16% of his/her colleagues in the same role in the organisation

More than 2% of his/her colleagues in the same role in the organisation

Less than 2% of his/her colleagues in the same role in the organisation

After the rating scale was developed, the last step before trying out the questionnaire

was to make the final format of it. The 51 items were shuffled and two â&#x20AC;&#x153;easyâ&#x20AC;? items from the communication skills orientation were put at the beginning. Five columns next to each item were also added so that each rater would be able to rate up to 5 employees, reading the item just once. The final format of the questionnaire is presented in appendix 5. Similar format of the competencies questionnaire was adapted both for the next phase of the pilot in Greece and for the main study, as well. Along with the 51-items competencies questionnaire, the 6-items job proficiency index was also included, which had to be completed from the supervisors with regard to

123

Method and pilot study

the employees they were appraising. This would be used as a criterion of the validity of the competencies questionnaire providing an overall job proficiency score. 5th stage At the following stage the field trial of the questionnaire was necessary. Although a pilot study in Greece was scheduled, it was decided to do a small-range pilot study in

UK, since it could give a useful first indication of the validity of the items and the reliability of the scales.

Two large organisations in Greater Manchester were approached and agreed to

AL LO

participate. The HR departments distributed the questionnaires to a number of managers within their organisations and asked each of them to rate up to 5 of their subordinates. Unfortunately, initially only 10 managers responded obtaining useful data for 46 employees. In the following two months, additional questionnaires for 13 employees were received, increasing thus the sample of the pilot into 59 people, but since all the

analyses have already been made using the initial pool of 46 employees, and the main

part of the data collection in Greece had already started, these additional data were not

used. Anyway, further analyses using the whole pool of the data showed the same factor

structure obtained from the initial data and small differences only in the item level. All the respondents received feedback, where the rateesâ&#x20AC;&#x2122; scores in each cluster were

reported along with the mean score of the whole sample.

The following table shows the descriptive results of the 5 scales of the competencies

clusters and the job proficiency index along with their intercorrelations. The scale score

is the average of the individual items of each scale, in order for the scales to be easier

comparable:

124

Method and pilot study

Table 33. Means, standard deviations and inter-correlations of the competencies scale (UK pilot) 1

.89** .89** .85** .85** .34*

.96** .83** .81** .48**

.87** .86** .39**

.80** .30*

.19

Competencies scales N Mean SD Goal organisation (1) 46 4.90 1.53 Action orientation (2) 46 4.84 1.60 Leadership orientation (3) 46 4.49 1.73 People orientation (4) 46 4.83 1.45 Communication skills (5) 46 4.69 1.49 Job proficiency index (6) 46 3.89 .84 ** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed)

It’s obvious that the 5 scales of the GWCQ are highly intercorrelated; the Pearson’ s

AL LO

r correlation coefficients are all very high and statistically significant at the .01 level. The same is the case for the relationship between the job proficiency index and the 4 scales (at least at the .05 level) with the exception of the communication skills cluster. Apparently, the raters of the UK sample did not relate good communication skills to job

performance.

The reliability of the GWCQ items was very satisfactory with Cronbach’ s α of .992

and Guttman’ s split-half of .978 (.982 when corrected for length). All the individual scales had also very satisfactory Cronbach’ s α ranging from .9679 up to .9822. Similar

results came up for the job proficiency index (α: .9454). The reliability coefficients of the competencies scales were very high, a fact which could be attributed to the

homogeneity of the behavioural domains sampled, since all five scales described

behaviours, which were both very easily identifiable in the workplace and also, within

scales, captured very similar behaviours. According to Anastasi (1990), the more

homogeneous the domain under investigation, the higher the internal consistency of the scale.

The next stage of the analysis included the exploratory factor analysis of the 51 items

in order to examine the factor structure of the questionnaire. Although the sample size was very small in order to conduct a proper factor analysis, we proceeded with it because it was the best way to examine the appropriateness of the items of the GWCQ for inclusion in the questionnaire which would be trialed in Greece. The principal components factor analysis yielded 5 factors; their eigenvalues and the variance explained are presented in the following table.

125

Method and pilot study

Table 34. First factor analysis of the UK pilot Factor 1 2 3 4 5

Eigenvalue 37.11522 2.48888 2.16036 1.53219 1.03907

Pct of Variance 72.8 4.9 4.2 3.0 2.0

Cumulative Pct 72.8 77.7 81.9 84.9 86.9

The rotation of the factors using direct oblimin produced the results shown in the following table. Oblimin rotation was employed because the 5 scales apparently were

not independent, as their statistically significant intercorrelations presented earlier have

shown. Similar results, with minor changes in the item level, were obtained independent

AL LO

of the extraction method adopted (i.e. principal components or principal axis factoring).

Factor 3 .067 .036 .000 -.046 .067 .130 .100 .017 .179 .066 .193 .238 .383 .153 .314 .368 .295 .231 .058 -.064 .056

2 .056 -.076 .114 -.064 .078 -.026 .095 .044 .048 .228 .179 .020 .088 .440 .129 .193 .164 .282 .261 .147 .359

LEADER2 ACTION8 ACTION4 ACTION5 ACTION2 ACTION7 ACTION9 LEADER8 LEADER9 LEADER1 ACTION3 LEADER6 LEADER3 COMMUN2 LEADER7 LEADER4 LEADER5 LEADER10 GOAL1 GOAL9 GOAL2

1 .928 .910 .851 .841 .812 .721 .696 .631 .620 .612 .583 .564 .539 .535 .528 .494 .493 .478 .467 .454 .411

Table 35. Pattern matrix of the UK pilot

4 -.057 .149 .004 .138 .057 .213 .116 .105 .298 .037 .058 .252 -.006 -.248 .121 -.064 .021 .179 .289 .404 .128

5 .100 .114 -.016 -.218 .016 -.065 -.026 -.336 .088 -.249 -.181 -.031 -.171 -.125 -.086 -.084 -.245 .064 -.164 -.254 -.256 (table continued)

126

Method and pilot study

-.121 .155 -.251 .056 -.047 -.032 -.056 .136 -.011 .127 -.396 .261 .090 -.182 -.097 -.229 -.235 .018 -.304 -.402 -.263 -.302 -.352 .020 -.020 .028 -.026 -.303 -.408 -.382

.003 .257 -.038 .080 .055 .122 -.208 .076 -.050 .271 -.007 .027 .007 -.035 .251 -.112 .171 -.053 .089 -.059 .141 .366 .282 .784 .658 .576 .554 .519 .247 .369

AL LO

COMMUN3 -.084 .043 .911 COMMUN5 -.010 .054 .800 COMMUN10 -.125 -.007 .798 COMMUN6 .178 .148 .710 COMMUN9 .190 .103 .694 COMMUN11 .316 -.030 .663 COMMUN4 .320 .246 .580 COMMUN12 .441 .022 .577 COMMUN1 .298 .266 .570 COMMUN7 .094 .238 .505 COMMUN8 .329 .007 .450 PEOPLE9 .027 .037 .973 PEOPLE10 -.019 .131 .922 PEOPLE4 .071 -.008 .800 PEOPLE7 -.009 -.051 .800 PEOPLE2 .190 .018 .709 PEOPLE6 .062 .083 .662 PEOPLE11 .295 .217 .609 PEOPLE3 .090 .194 .557 PEOPLE1 .172 .146 .496 PEOPLE5 .228 .104 .413 GOAL4 -.261 .364 .406 GOAL3 -.111 .282 .399 GOAL6 .136 .095 .084 GOAL8 .216 .309 -.126 ACTION6 .371 .033 .174 PEOPLE8 .222 -.270 .494 GOAL7 .171 .093 .167 ACTION1 .384 .085 .146 GOAL5 .191 .243 .122 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. a Rotation converged in 24 iterations. The highest factor loadings are presented in bold.

Table 36. Inter-correlation matrix of the 5 factors of the UK pilot

Factor 1 Factor 2 Factor 3 Factor 4 Factor 5

Factor 1 1.00 .576 .615 .506 -.340

Factor 2

Factor 3

Factor 4

1.00 .549 .352 -.418

1.00 .462 -.376

1.00 -.252

The results of the factor analysis showed that quite a few amendments were necessary.

127

Method and pilot study

First as regards the fifth factor, where 2 items had the highest loadings; action1 (shows energy during the accomplishment of his/her tasks) and goal5 (works with proficiency) with factor loadings of -.40798 and -.38233 respectively. This factor had an eigenvalue of 1.039 explaining 2% percent of the total variance. Subsequently it was decided to drop out the items of this factor since they both had low loadings both with their respected factor and with the others, and in addition one of them, goal5, was

argued during the earlier discussed experts procedure, that it could be applied in all of the clusters, not necessarily to one.

The next factor, factor 4, consisted of five items, with an eigenvalue of 1.532

AL LO

explaining 3% of the total variance. It was decided to keep this factor but a number of alterations to the items looked necessary. First the three goal items that loaded to the factor and identify it quite clearly were included. These items were goal6 (follows tasks through completion), goal8 (accomplishes his/her duties in a systematic way), and goal7 (attempts to reach the required standards of his/her position) with loadings of .784,

.658 and .519 respectively. On the other hand, action6 (is active most of the time) and people8 (will do some extra-work in order to help his/her colleagues accomplish their

tasks) also loaded to this factor with .576 and .554 respectively. It was recommended

that this factor incorporates a concept of task accomplishment orientation rather than a goal orientation. Subsequently, action6 and people8 could be reworded in a more task-

orientated way. Thus, action6 became “is actively engaged to the accomplishment of

his/her everyday tasks” and people8 “will work overtime in order to achieve his

everyday tasks”. Moreover, at least one more item although didn’t load to the present factor as one would expect, had a quite high loading with this factor. It’s goal9

(organises in advance his/her plans to achieve goals) which has a loading of .45428

with the first major factor and a .40448 with the fourth factor. Subsequently, it’s suggested that this item should be reworded in the following way: “organises in advance the next day’s tasks at work” and considered in the future as part of the present orientation. In conclusion, the new task orientation cluster included the following items: 1. follows tasks through completion 2. accomplishes his/her duties in a systematic way 3. attempts to reach the required standards of his/her position

128

Method and pilot study

4. is actively engaged to the accomplishment of his/her everyday tasks 5. will work overtime in order to achieve his/her everyday tasks 6. organises in advance the next dayâ&#x20AC;&#x2122;s tasks at work with the following definition: In the task orientation cluster, the employee is actively engaged and orientated His/her major concern is the successful execution of his/her duties.

towards the accomplishment of his/her duties/tasks, working systematically towards it.

The next two factors were quite straightforward. In the following third factor all but

AL LO

one of the people orientated items had the highest loadings with the perspective factor, identifying it as a clear-cut people factor, with a minimum factor loading of .413. Thus it was decided that all these items should be included, but drop items goal4 and goal3 which unexpectedly also loaded in the present factor. Subsequently, the items kept were

the following:

This employee:

displays kindness towards his/her colleagues shows consideration for his/her colleagues s/he tries to help her/his colleagues with their problems cares about the problems of his/her colleagues interacts with other people in a sensitive way respects his /her colleagues shows friendly behaviour within the organisation has good relations with most of his/her colleagues shows positive feelings towards his/her colleagues takes into account the suggestions of his/her colleagues

The same applies for the following second factor where again all but one the

communication orientation items were loaded with a minimum loading of .450, suggesting that these items should be kept for the following pilot study in Greece, since they were identifying very clearly a communication factor:

129

Method and pilot study

AL LO

uses correct grammar in writing uses an appropriate style in writing uses correct spelling in writing uses suitable language both in writing and speaking does not include irrelevant material both in writing and speaking presents information in a well-organised and logical way speaks fluently demonstrates clarity of thought both in writing and speaking speaks clearly adapts a different style of language depending on the target audience covers the main points of an issue (e.g. during an oral presentation)

This employee:

Finally, the first factor emerging was the most significant, explaining the majority of the variance. It had an eigenvalue of 37.11522 explaining a total variance of 72.8%. In this factor all the leadership items were loaded, seven from the nine action items, three goal and one communication item. It was recommended that the subjects perceived

these items as leadership characteristics. They attributed them as part of the leadership

process only, without considering that a lot of these items may be part of non-

managerial or non-leadership positions as well. It could be argued that the participants

perceived leader as the “action” man. Furthermore, this shouldn’t be considered totally inappropriate since people that have “action” competencies usually dominate their

environment, pushing it forward by their “actions”. Subsequently, people perceive these

employees as “leaders” attributing them characteristics similar to leadership. On the

other hand, it has to be emphasised that this factor is somewhere in between leadership and action, since “traditional” leadership items, such as leader7 (leads his/her

colleagues effectively), leader5 (influences his/her colleagues), and leader10 (directs

his/her colleagues positively) had quite low loadings than expected in a leadership factor. Subsequently, it was decided to combine these two factors - the action and leadership orientations - keeping the items that identify it better, the first 13 items which are the following:

130

Method and pilot study

This employee:

AL LO

takes the role of the leader in group activities looks for new activities within his/her work environment is proactive (i.e. spontaneously takes action in order to achieve something at work) looks for stimulation at work responds with spontaneity to new conditions is willing to commit himself/herself to new tasks remains lively throughout his/her time at work inspires his/her colleagues "shows the way" to his/her colleagues shows decisiveness in decision-making situations behaves dynamically at work motivates his/her colleagues at work is capable of persuading his/her colleagues

with the following definition:

In the Action-Leadership orientation the employee is a leading figure within his/her work environment, working actively and dynamically, inspiring and motivating

his/her colleagues.

A review of these results suggests that the people and the communication skills orientations were considered independently from the other two clusters of

competencies, with one exception, commun2 (delivers effective oral presentations) which seems to be considered as a leadership competency, and was dropped from

subsequent analyses, but the same does not apply to action orientation which is

attributed, as already discussed as part of the leadership position.

Finally, 40 items were included in the 2nd version of the questionnaire piloted in

Greece, with 4 clusters instead of 5. The following table presents these 4 clusters along

with their definitions and the items assigned to each one:

131

Method and pilot study

Task orientation In the task-orientation cluster, the employee is actively engaged and orientated towards the accomplishment of his/her duties/tasks, working systematically towards it. His/her major concern is the successful execution of his/her duties. This employee: 1. follows tasks through to completion 2. attempts to reach the required standards of his/her position 3. accomplishes his/her duties in a systematic way 4. organises in advance his/her plans to achieve goals 5. will work overtime in order to achieve his/her everyday tasks 6. organises the next dayâ&#x20AC;&#x2122;s tasks at work

Table 37. Definitions of clusters and items of the competencies measure used in the Greek pilot

AL LO

Action-leadership orientation Here the employee is a leading figure within his/her work environment, working actively and dynamically, inspiring and motivating his/her colleagues. This employee: 1. responds with spontaneity to new conditions 2. behaves dynamically at work 3. is proactive (i.e. spontaneously takes action in order to achieve something at work) 4. looks for stimulation at work 5. is willing to commit himself/herself to new tasks 6. looks for new activities within his/her work environment 7. remains lively throughout his/her time at work 8. shows decisiveness in decision-making situations 9. takes the role of the leader in group activities 10. is capable of persuading his/her colleagues 11. motivates his/her colleagues at work 12. inspires his/her colleagues 13. "shows the way" to his/her colleagues

People orientation In the people orientation cluster, the employee is sensitively and positively orientated towards his/her colleagues, taking into account their opinions and suggestions. S/he respects her/his colleagues, showing consideration for their problems. This employee: 1. shows positive feelings towards his/her colleagues 2. interacts with other people in a sensitive way 3. has good relations with most of his/her colleagues 4. s/he tries to help her/his colleagues with their problems 5. takes into account the suggestions of his/her colleagues 6. respects his /her colleagues 7. cares about the problems of his/her colleagues 8. displays kindness towards his/her colleagues 9. shows consideration for his/her colleagues 10. shows friendly behaviour within the organisation

132

AL LO

Communication skills orientation In the communication skills cluster the employee has good oral and written communication abilities. In the case of oral presentations s/he communicates in a simple and comprehensible way, clearly and fluently. S/he writes in clear and concise way, using correct grammar, style and language. This employee: 1. speaks clearly 2. uses correct grammar in writing 3. speaks fluently 4. uses an appropriate style in writing 5. uses suitable language both in writing and speaking 6. adapts a different style of language depending on the target audience 7. covers the main points of an issue (e.g. during an oral presentation) 8. does not include irrelevant material both in writing and speaking 9. uses correct spelling in writing 10. presents information in a well-organised and logical way 11. demonstrates clarity of thought both in writing and speaking

Method and pilot study

6th stage

At the following stage the questionnaire had to be translated in Greek.

The translation of a complex questionnaire in another language is not such an easy task. A word by word translation would not be appropriate since a lot of the items

would not make sense in Greek language. Vijner and Hambleton (1996) have developed a series of 22 guidelines for test translation and adaptation, which were taken very

seriously into account during the following procedures. A first translation of the 40 items of the GWCQ and the 6-items job proficiency

index was carried out by 4 Greek PhD students, who were studying in Manchester. They

were all English literate with deep knowledge of Greek language as well. Furthermore,

they were aware of the purposes of our research and also three of them were occupational psychologists, familiar with the work competencies literature. Subsequently, their translations were given to 4 other Greeks, who had also studied

in UK recently, were aware of our research for the back translation in English and they were also English literate. They back-translated the questionnaire back in English. Next, as a result of the translations, an adapted version of the questionnaire in Greek was produced, and along with the original English version, a panel discussion was held between me and two experts in English and Greek languages. The first of them had

133

Method and pilot study

studied English Literature at the University of Athens and had a Master’s degree in Machine Translation from UMIST and the second had studied Greek Literature at the University of Crete, in Greece. They were both working in Schools in Greece as teachers of English and Greek literature respectively. We checked again the translations, the back-translations and the adapted version I had produced, and after the necessary rewordings and corrections we came up with final version of the GWCQ used for the

pilot study in Greece.

The same procedure was followed for all the questionnaires that needed translation

and were used in this study.

AL LO

7th stage

A pilot study of the 2nd version of the GWCQ in Greece was essential for a number of reasons. First, because it was necessary to verify that the new instrument is valid and reliable and also because the measure should be shortened, since each supervisor would

usually have to rate up to 5 subordinates, in some cases even more.

A total number of 24 managers participated in the pilot study carried out in Greece providing useful data for 107 subordinates. The Kaiser-Meyer-Olkin measure of

sampling adequacy reached a value of .930, which should be considered as “marvellous” (Norussis, 1994). Moreover, another index of the sample suitability, the

Bartlett’s test of sphericity reached a statistically significant level. These are two basic

requirements in order to proceed to the factor analysis.

The following table shows the descriptive results of the 4 scales of the competencies

clusters and the job proficiency index along with their intercorrelations for the Greek sample. Again, as it was the case in the UK pilot the scale scores are the average of the

individual items of each scale in order to be easier comparable: Table 38. Means, standard deviations and inter-correlations of the competencies scale (Greek pilot) Variables Mean SD 1 Task orientation (1) 4.64 1.61 Action-Leadership orientation (2) 4.25 1.60 .919 People orientation (3) 4.86 1.45 .639 Communication skills (4) 4.81 1.38 .817 Job proficiency index (5) 3.62 .99 .762 Note: All correlations are significant at the 0.01 level (2-tailed)

.718 .784 .752

.623 .439

.600

134

Method and pilot study

It is obvious from that table that the scales are all highly intercorrelated, more than the respective UK sample. Especially, the first two clusters, the task and the actionleadership orientation have a very high correlation coefficient of .919. Also, the job proficiency index have statistically significant correlations with all 4 clusters. Apparently, according to the managers of our sample, there is a strong positive

relationship between all four types of work competencies and job proficiency, although the strength of this relationship varies between clusters of competencies.

The reliability of the GWCQ items of the Greek sample was again very satisfactory, as was the case in the UK sample. The Cronbach’ s α was .980 and the Guttman’ s

AL LO

split-half was .969 (.970 when corrected for length). All the individual scales had also very satisfactory Cronbach’ s α ranging from .9244 up to .9635. Similar results came up for the job proficiency index (α: .9532). As was the case earlier, during the first pilot carried out in the UK, the high internal consistency reliabilities were attributed to the

high homogeneity and similarity of the items.

The factor analysis that followed including all 40 items yielded the following 5

factors.

Eigenvalue 22.973 3.722 2.014 1.690 1.146

Pct of Variance 57.434 9.305 5.035 4.225 2.865

Cumulative Pct 57.434 66.739 71.774 75.999 78.865

Factor 1 2 3 4 5

Table 39. First factor analysis of the Greek pilot

The oblimin rotation produced the following pattern matrix, which was similar to the

solution produced by a principal-axis factoring extraction.

135

Method and pilot study

Table 40. Pattern matrix of the first factor analysis (Greek pilot)

IN PR

5 -.010 -.028 .035 .369 .276 .135 -.005 .044 .034 -.012 -.001 -.112 .052 .080 -.143 .025 -.055 .036 .070 .053 .380 .132 .047 .288 .097 -.150 .214 .130 -.148 .056 -.008 -.249 .120 .097 -.127 -.216 -.028 .024 .646 .529

4 -.227 -.035 -.320 -.210 -.230 .154 .113 .045 -.018 -.080 -.102 -.077 .004 .074 -.238 -.361 .037 .048 .052 -.155 .006 -.379 -.375 -.886 -.862 -.798 -.764 -.694 -.669 -.612 -.602 -.599 -.575 -.472 -.468 -.464 -.460 -.458 -.136 -.450

AL LO

Factor 1 2 3 ACTLEA5 .285 .161 .562 TASK5 .238 .371 .546 ACTLEA11 .156 .187 .542 ACTLEA6 .152 .167 .497 ACTLEA4 .102 .241 .449 PEOPLE8 -.090 -.036 1.017 PEOPLE9 .011 -.029 1.009 PEOPLE6 .040 -.041 .936 PEOPLE10 -.124 .044 .900 PEOPLE3 -.093 -.034 .898 PEOPLE2 .047 -.023 .831 PEOPLE1 -.032 -.005 .802 PEOPLE4 .124 .109 .743 PEOPLE7 .239 .177 .735 PEOPLE5 .170 .033 .580 ACTLEA12 .239 .069 .451 COM2 -.085 .052 .961 COM9 -.050 .060 .930 COM4 .135 -.063 .926 COM5 .177 -.033 .761 COM8 .159 .017 .607 COM7 .151 .091 .472 COM3 -.400 .325 .406 ACTLEA8 -.127 -.032 -.075 ACTLEA2 .051 .026 .003 TASK1 .222 -.108 .126 ACTLEA10 -.150 .191 .001 ACTLEA3 .077 .147 .072 COM10 .033 .010 .314 ACTLEA1 .122 .130 -.032 TASK3 .079 -.026 .369 TASK2 .212 .356 -.033 TASK4 .183 .070 .239 TASK6 .402 -.045 .295 COM1 -.421 .241 .388 COM11 -.133 .205 .420 ACTLEA13 .255 .447 .013 ACTLEA7 .224 .321 .087 COM6 -.127 .138 .266 ACTLEA9 .268 .099 -.094 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. a Rotation converged in 55 iterations. The highest factor loadings are presented in bold.

136

Method and pilot study

Table 41. Inter-correlation matrix of the 5 factors of the Greek pilot Factor 1 1.00 .215 .254 -.320 .172

Factor 1 Factor 2 Factor 3 Factor 4 Factor 5

Factor 2

Factor 3

Factor 4

1.00 .445 -.552 .162

1.00 -.560 .198

1.00 -.214

It is obvious from the factor pattern matrix that more modifications of the questionnaire were necessary and also that there are quite a few similarities with the UK

pilot study; both the people orientation and the communication skills clusters, i.e. the

AL LO

second and third factor of the pattern matrix, emerged again very clearly. Furthermore, the items of the task orientation and the action-leadership orientation came together in two factors. In the first one, four items from the action-leadership cluster are combined with one from the task orientation, but without very high loadings, and similarly, in the third factor, seven action-leadership items are combined with five task items, along with

three communication items.

The first point that should be made here is that apparently, the participants could distinguish the concepts of people orientation and communication skills easier than the

other two competencies clusters, i.e. the task orientation and the action-leadership orientation. Subsequently, it has been decided that these two clear cut factors (people

and communication skills) would be included in the subsequent analyses and that the

focus of attention would concentrate to the interpretation of the remaining two factors.

Numerous exploratory factor analyses were conducted, repetitively excluding items

with low loadings, trying to identify a simple, coherent and interpretable factor structure, but all the results were similar with the results of table 40. Four factors were

always coming up, two of them were the people and communication factors, whereas the other two were combinations of the task and action-leadership items. Subsequently, since the problem had essentially to do with the task and actionleadership clusters, it was decided to drop one of them out of the study. It was not an easy decision, since plenty of information would be missed, but it was apparent that the participants could not distinguish the differences between these two groups of competencies. Thus it was decided to keep the action-leadership cluster and drop out the

137

Method and pilot study

task orientation items. This was considered as more appropriate for the purposes of the study for the additional reason that it would be the only way of measuring leadership potential, which was also of more interest for the participating organisations comparing to the task orientation competencies. An hierarchical regression analysis (Cohen and Cohen, 1983) was also conducted in order to examine whether the remaining items could predict the dropped items of the

task orientation cluster. The three scales, people, communication, and action-leadership

orientation were entered as independent variables in the equation with the task scale as

dependent. The results of this regression are summarised in the following table.

Variables ACTIONLEADERSHIP

Adj. R2

.845

.844

573.372

.000

SE of B

Beta

Sig of T

.926

.039

.919

23.945

.000

4.049

.000

.919

ACTIONLEADERSHIP (constant) Variables

.176

Mult. R

Adj. R2

.932

.869

.867

345.877

.000

SE of B

Beta

Sig of T

.728

.058

.723

12.656

.000

.292 .145

.067 .207

.250

4.378 .700

.000 .486

ACTIONLEADERSHIP COMMUNICATION

.711

Step 2

Mult. R

Step 1

AL LO

Table 42. Hierarchical regression analysis of the competencies scales (Greek pilot)

ACTIONLEADERSHIP COMMUNICATION (constant)

Dependent variable: the â&#x20AC;&#x153;droppedâ&#x20AC;? task orientation scale

Summing up, the results of the regression analysis show that two of the remaining

three scales, the action-leadership orientation and the communication skills, predict the task orientation very well with a multiple R = .932 (R2 = .869, F(2,104) = 345.877,

p < .000). Dropping the task orientation items from the subsequent analyses should not

be considered a big loss, since almost 87% of its variance was explained by the remaining three competencies clusters. The first exploratory factor analysis including all but the task orientation items resulted again in a 4-factors solution, with the action-leadership items divided in two groups. But when an analysis was requested for three factors only, instead of asking for

138

Method and pilot study

all the factors with eigenvalues over 1, a very clear cut and easily identified solution came out, explaining 72.237 % of the total variance. The following table shows the rotated factor pattern matrix of this analysis, using oblimin rotation. Identical results were obtained with principal-axis factoring extraction.

AL LO

Factor 1 2 3 ACTLEA9 .052 -.176 .973 ACTLEA6 .033 -.029 .913 ACTLEA11 -.034 .027 .832 ACTLEA4 .055 .075 .818 ACTLEA5 -.160 -.006 .743 ACTLEA2 -.066 .132 .736 ACTLEA8 -.025 .093 .695 ACTLEA3 -.164 .149 .665 ACTLEA1 -.131 .037 .610 ACTLEA13 -.417 .005 .599 ACTLEA7 -.292 .091 .581 ACTLEA10 -.229 .161 .576 ACTLEA12 -.394 .042 .562 COM7 -.037 .449 .549 COM10 -.075 .427 .439 COM6 -.045 .241 .411 PEOPLE8 -.057 -.030 -.959 PEOPLE9 -.006 -.042 -.955 PEOPLE3 -.005 .036 -.896 PEOPLE10 -.023 .097 -.879 PEOPLE6 .093 -.055 -.876 PEOPLE1 -.008 .056 -.799 PEOPLE2 .160 .001 -.794 PEOPLE4 .212 .064 -.663 PEOPLE7 .313 .069 -.607 PEOPLE5 .306 .035 -.551 COM2 -.068 -.017 .956 COM9 .027 .004 .895 COM4 .217 .166 .818 COM5 .382 .117 .689 COM1 -.130 -.364 .647 COM3 -.057 -.401 .617 COM11 .110 -.282 .566 COM8 .447 .125 .471 3 factors were requested Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. a Rotation converged in 8 iterations. The highest factor loadings are presented in bold.

Table 43. Second factor analysis of the competencies scale (Greek pilot)

139

Method and pilot study

Table 44. Inter-correlation matrix of the 3 factors of the Greek pilot Factor 1 1.00 -.534 .578

Factor 1 Factor 2 Factor 3

Factor 2 1.00 -.429

From the results of table 43, it is apparent that all the action-leadership items loading

in the first factor explain the majority of the variance (56.71%). Again as it was the case in the previous analyses, the people and the communication skill factors emerged very

clearly. A similar factor analysis that followed, excluding the three communication

items that had their highest loadings with the action-leadership factor instead of the

AL LO

communication factor, showed the clearest factor structure so far, with all the items loading in the intended factors with similar loadings to table 43. Subsequently, it was decided to choose the items for inclusion in the final version of the GWCQ used in the main study, based on these results.

Since this questionnaire would be one of the three that would be completed from the

supervisors with regard to their subordinates, along with the Organisational Citizenship

Questionnaire and the job proficiency index, it would be better to be as short and as

simple as possible. Thus, six items with the higher loadings from each factor were chosen, and a similar factor analysis was repeated. The resulted factor pattern matrix is

following. Here it has to be noted that three factors were emerging anyway, without

specifically requesting for three dimensions, with eigenvalues over 1, explaining a total

variance of 79.5%.

140

Method and pilot study

AL LO

Factor 1 2 3 ACTLEA6 -.022 .006 .926 ACTLEA9 -.043 -.117 .889 ACTLEA4 .005 .117 .839 ACTLEA11 -.055 .078 .815 ACTLEA5 -.171 .025 .768 ACTLEA2 -.162 .211 .586 PEOPLE8 .022 -.050 -.942 PEOPLE9 .078 -.054 -.916 PEOPLE3 .027 .027 -.903 PEOPLE10 .040 .095 -.865 PEOPLE6 .165 -.070 -.854 PEOPLE1 .037 .071 -.768 COM2 -.006 .019 .950 COM9 .081 .016 .882 COM4 .302 .164 .780 COM5 .412 .114 .696 COM1 -.150 -.374 .667 COM3 -.068 -.429 .629 Extraction Method: Principal Component Analysis. Rotation Method: Oblimin with Kaiser Normalisation. a Rotation converged in 8 iterations. The highest factor loadings are presented in bold.

Table 45. Final factor analysis of the competencies scale (Greek sample)

Table 46. Inter-correlation matrix of the final factor analysis of the Greek pilot Factor 1 1.00 -.425 .494

Factor 2

Factor 1 Factor 2 Factor 3

1.00 -.397

The reliability of the scale was very satisfactory with Cronbachâ&#x20AC;&#x2122; s a for the whole 18-

items scale of .9512 and for the individual scales a minimum of .9292 for the

communication scale, .9574 for the people orientation scale and .9355 for the actionleadership scale. At the following stage it was decided to drop out item com3 (speaks fluently) from the communication scale, because its Greek translation was very similar to the translation of com1 (speaks clearly). Dropping out this item brought down the reliability of the scale to .9254 which was not considered as such a big loss and subsequently the

141

Method and pilot study

17th-items solution was retained for the final version of the GWCQ used in the main study in Greece; these items are presented in the following table. Table 47. Clusters and items of the competencies measure retained for the main study

AL LO

People orientation This employee: 1. shows positive feelings towards his/her colleagues 2. has good relations with most of his/her colleagues 3. respects his /her colleagues 4. displays kindness towards his/her colleagues 5. shows consideration for his/her colleagues 6. shows friendly behaviour within the organisation

Action-leadership orientation This employee: 1. behaves dynamically at work 2. looks for stimulation at work 3. is willing to commit himself/herself to new tasks 4. looks for new activities within his/her work environment 5. takes the role of the leader in group activities 6. motivates his/her colleagues at work

Communication skills orientation This employee: 1. speaks clearly 2. uses correct grammar in writing 3. uses an appropriate style in writing 4. uses suitable language both in writing and speaking 5. uses correct spelling in writing

Following that, it was considered appropriate to explore the relationship of these

three scales with job proficiency. Thus, an hierarchical regression analysis was

conducted in order to examine how well the job proficiency index is predicted by the three competencies clusters. The results of this regression analysis are summarised

below:

Table 48. Stepwise regression analysis of the three competencies scales Step 1

Variables

Mult. R

Adj. R2

ACTIONLEADERSHIP

.706

.499

.494

104.629

.000

B .389

SE of B .038

Beta .706

T 10.229

Sig of T .000

2.040

.169

12.065

.000

ACTIONLEADERSHIP (constant)

Dependent variable: job proficiency index

142

Method and pilot study

The results of this regression analysis showed that the best predictor of the jobproficiency index was the action-leadership competencies cluster and not the other two clusters. Only, the action-leadership was included in the equation with a multiple R=.706 (R2=.499, F(1,105)=104.629, p<.000). Subsequently, it can be argued that the participants of the study are not considering the possession of good communication

skills or exhibiting a friendly behaviour towards other people within the organisation as associated with higher job proficiency, at least not as much as the competencies

included in the action-leadership orientation cluster. In a similar vein, the

intercorrelations of these three scales with job proficiency presented in the following the highest correlation with job proficiency.

AL LO

table along with their descriptive statistics show that the action-leadership cluster has

Table 49. Means, standard deviations and inter-correlations of the competencies scales 2

.450 .389

.490

Variables Mean Std 1 Action-Leadership orientation (1) 4.07 1.80 People orientation (2) 5.03 1.46 .543 Communication skills (3) 4.87 1.57 .609 Job proficiency index (4) 3.62 .99 .706 Note: All correlations are significant at the 0.01 level (2-tailed)

At the last stage of the pilot study in Greece, an attempt was made to replicate the results of the exploratory factor analysis doing a confirmatory factor analysis, using

EQS for Windows, version 5.5 (Bentler & Wu, 1997).

variances,

parameters,

A first analysis of the 17 items, requesting a 3-factor solution, where all the covariances

were

freely

estimated,

unsatisfactory

Comparative Fit Index (CFI) of .805 was obtained, whereas the minimum acceptable

cut-off point is .90 (Byrne, 1994). Thus a number of amendments were necessary in order to improve the CFI. 1. Two of the variables (COM1 and COM5) were set to be related not only with their intended factor, i.e. the communication factor, but were also allowed to correlate with one more. COM1 (“speaks clearly”) with people orientation factor and COM5 (“uses correct spelling in writing”) with the action-leadership factor. Their parameters were set to be freely estimated.

143

Method and pilot study

2. The covariances of the 3 factors between them were also set to be freely estimated along with the largest covariances between 7 pairs of â&#x20AC;&#x153;errorsâ&#x20AC;?. The model developed showed a CFI of .963 which was considered very satisfactory, being well above the minimum accepted level of .90. The results of the confirmatory factor analysis provided further support on the parsimony of the instrument replicating

and extending the results of the exploratory factor analysis. Based on the previous analyses the final version of the Generic Work Competencies Questionnaire consisted of 17 items tapping three clusters of competencies: Action-

leadership Orientation, People Orientation and Communication Skills Orientation. After

AL LO

examination of its psychometric properties, it was considered that it was a reliable, valid measure of generic work competencies, adequate for the purposes of the present study. Furthermore, in a conceptual basis, leadership, people orientation and communication skills are part of almost every measure in the competencies field. Thus the questionnaire

was included as one of the three measures completed by the immediate supervisors of

the employees who completed the personality and job satisfaction questionnaires, along

with the Organisational Citizenship Behaviour Questionnaire and the job proficiency

index.

144

Results

Chapter 10. Results 10.1. Sample description The total number of employees participating in the study was 227. The majority of them were below the age of 40, predominantly females, as demonstrated in the

Participantsâ&#x20AC;&#x2122; gender

Figure 9.

following two figures.

AL LO

male 43%

female 57%

120

107

100

G TI

31-40 41-50 51-60 26

20-30

80 60

140

61+ 4

51-60

61+

20-30

31-40

41-50

Figure 10. Participantsâ&#x20AC;&#x2122; age

Figure 10 shows that 107 participants (47%) were in the age of 20-30, 82 employees (36%) in the age band of 31-40, 26 in the age of 41-50 (11%), 4 in the age of 51-60 (2%) and finally 2 employees were over 61 years of age (1%). Employees were grouped

145

Results

in the subsequent analyses into two age categories of similar size, above and below 30 years of age. The two new age bands consisted of 107 employees (48.4%) below the age of 30, and 114 employees (51.6%) above the age of 30. It was expected that the majority of participants in the first category would be newcomers in their organisations, compared to older employees who were expected to be in their post longer. This result was replicated from a comparison of the tenure means for the two age bands (young

employees mean: 20.5 months; older employees mean: 81.5 months; t (213)= -8.57, p < .000).

Fifty-seven supervisors in total (41 males) provided the performance data (a ratio of

AL LO

1 supervisor for 4 subordinates). Further demographics are not available for supervisors. The participants were drawn from 22 organisations in total, i.e. an average of 10 employees from each organisation. Half of them were working for two years or less in their current position, with a mean tenure of 4.3 years, employed under 40 different job

titles. For statistical purposes, and also in order to examine differences between groups

of occupations, the different job titles were grouped into the job families presented in

the following table. In order to improve the reliability of the procedure, the researcher and two other Greek occupational psychologists repeated the task of assigning each job

title to a job family, independently from each other and then the researcher brought together the different viewpoints producing the following job families. The average

level of agreement between the raters was 78%.

Table 50. Job families of the sample

Teachers Managers Administrative Accountants HR-Training Arts/Design Sales/Insurance Scientific Total Missing cases Total

Frequency

Percent

37 30 22 16 17 21 36 13 192 35 227

16.3 13.2 9.7 7.0 7.5 9.3 15.9 5.7 84.6 15.4 100.0

Valid percent 19.3 15.6 11.5 8.3 8.9 10.9 18.8 6.8 100.0

Cumulative percent 19.3 34.9 46.4 54.7 63.5 74.5 93.2 100.0

146

Results

The first category of “teachers” was quite straightforward including all the teachers of the sample, who were working in elementary and high schools. The job titles including the word “manager” in their title or any similar positions, such as supervisors, created the job family called “managers”. Jobs such as secretaries, and computer officers were included in the “administrative” category, whereas in the “accountants” category were included all the jobs with the word accountant in their title, and also tax

consultants. In the “HR-Training” category, human resources consultants of personnel/training officers were included, and in the “arts/design” category, graphic

designers or marketing producers were assigned. All the salesmen and insurance or

pharmaceutical consultants were included in the “sales/insurance” category. Finally,

AL LO

engineers and chemists were included in the “scientific” category. Subsequently, the job families were grouped into two types of occupations, involving interpersonal interaction and direct communication with other people or not, in order to examine the specific hypotheses of the study. Thus, it was suggested that occupations such as teachers,

managers, human resources and sales, which involve a great deal of interaction with

other people, such as students, subordinates, colleagues or customers, are quite distinct

from posts in administration, accountancy, arts/design and science. People employed in

these posts normally are not exposed in every-day interaction with people, as part of their duties, and their interpersonal skills are not usually essential in exhibiting

successful job performance. An exception to that could be secretaries, which was

assigned as an administrative post, under the family of jobs not involving interpersonal

interaction. Some secretaries, especially in personal assistant positions spend a lot of their time interacting with other people. Nevertheless, it was suggested that they still

have to do a lot of administrative work and further, most of simple secretarial posts have

only to do this type of tasks. The following table presents the statistics of the two types of occupations where interpersonal interaction is central.

147

Results

Table 51. Job families and centrality of interpersonal interaction

Jobs involving interpersonal interaction

Jobs not involving interpersonal interaction

Teachers Managers HR-Training Sales/ insurance Administrative Accountants Arts/Design Scientific Total Missing

Percent

Valid percent

120

52.9

62.5

31.7

37.5

192 35

84.6 15.4

Job family

100.0

Type of interaction

AL LO

10.2. Reliability and factor structure of the measures

Before examining in more detail the available data, it was necessary to examine the internal consistency and factor structure of the three performance measures, i.e. the OCB measure, the competencies measure and the job proficiency index. The

examination of the factor structure of the job satisfaction questionnaire was not

considered necessary, since it was decided to use it as an overall measure of job

satisfaction rather than a two-dimensional measure of intrinsic and extrinsic job satisfaction, as Warr et al. (1979) suggested, following the results of cluster analysis. A

total job satisfaction score was computed for each employee, by averaging the scores of all fifteen items, where employees were reporting their job satisfaction level. In the

following sections the factor stability of the performance measures will be discussed

concluding with the internal consistencies of the measures. 10.2.1 The OCB measure

The OCB measure originally developed by Smith et al. (1983) assesses two

dimensions, as discussed already in the relevant section; altruism and conscientiousness or generalised compliance. An exploratory factor analysis of the OCB items produced a four-factor solution, explaining 64.4% of the total variance, where the altruism dimension was coming out as the first major factor, along with one more item from the conscientiousness dimension (item 6: â&#x20AC;&#x153;Exhibits attendance at work beyond the normâ&#x20AC;?). The remaining items of the conscientiousness dimension split into three dimensions. When the analysis was repeated, excluding item 6, three factors emerged, explaining 58.4% of the variance, with a clear altruism factor and the remaining items split into

148

Results

two factors. Thus, in the following analysis, two factors were requested, excluding item 6. The results of the oblimin factor analysis, using principal components, are presented in the following table, along with the scree plot of the analysis. They were identical with the results of principal axis factoring extraction method. Table 52. Factor analysis of the OCB measure (excluding item 6) Factor 2

.83

-.08

.82 .81 .80

-.01 -.11 .10

.77

-.04

.75 .72 -.27 -.16 .14 .07 .29 -.09

-.00 .15 .75 .67 .58 .57 .51 .47

.18

.35

.24

.31

AL LO

Takes the initiative to orient new employees to the department even though it is not part of his job description Helps others when their work load increases Makes innovative suggestions to improve department's overall quality Volunteers to do things not formally required by the job Willingly attends function not required by the organisation, but help its overall image Helps other employees with their work when they have been absent Assists me with my duties Takes undeserved work breaks (r) Coasts towards the end of the day (r) Does not take extra breaks Does not take unnecessary time off work Does not spend a great deal of time in idle conversation Spends a great deal of time in personal telephone conversations (r) Exhibits punctuality in arriving at work on time in the morning and after lunch breaks Gives advance notice if unable to come to work

OCB-Conscientiousness

OCB-Altruism

Item

Kaiser-Meyer-Olkin Measure of Sampling Adequacy = .806 Eigenvalues Factor 1 = 4.90 Factor 2 = 2.27 % of variance explained Factor 1 = 32.70 Factor 2 = 15.13

Factor 2

Factor 1 .192

Intercorrelation matrix of the factor analysis of the OCB measure

149

Results

AL LO

Figure 11. Scree-plot of the OCB (two factors requested)

Although some of the factor loadings of the OCB-conscientiousness dimension were

quite low, and 10% of the total variance was lost as a result of dropping the third factor, the results of this analysis were considered satisfactory, replicating to a point the two-

factor solution found originally. They were also correlated to a statistically significant

level with the job proficiency measure, providing further evidence of the construct

validity of the measure, as will be discussed later in more detail. The two scales showed internal consistencies of .90 and .71 respectively, which were considered acceptable at

this stage, especially altruism’s internal consistency. However, in order to improve

conscientiousness’ dimension, one more item was dropped from subsequent analyses, which, when dropped, resulted in a minor increase in the scale’s α. The item dropped

was item 10 (“Spends a great deal of time in personal telephone conversations”), increasing scale’s α to .717. Thus, the personal dimension of OCB or altruism consisted of items 5, 7, 13, 3, 15, 1, 12 and the impersonal dimension or conscientiousness from items, 4, 14, 11, 8, 16, 2, 9. The two scales of OCB-altruism and OCBconscientiousness were computed for each employee, by averaging the perspective scores of the items for each scale.

150

Results

10.2.2 The competencies measure The pilot study discussed earlier identified three dimensions for the Generic Work Competencies Questionnaire developed for the purposes of the study. An oblimin factor analysis of the items replicated the results of the pilot study, providing further evidences on the appropriateness of the measure as a valid and reliable instrument for the measure of generic work competencies. The following table presents the results of the oblimin

factor analysis using principal components, which were identical with the principal axis factoring extraction method, and figure 12 shows the scree plot of the analysis.

Actionleadership skills

Comm/tion skills

People orientation

Displays kindness towards his/her colleagues Respects his/her colleagues Shows positive feelings towards his/her colleagues Shows friendly behaviour within the organisation Has god relations with most of his/her colleagues Shows consideration for his/her colleagues Uses correct spelling in writing Uses correct grammar in writing Uses suitable language both in writing and speaking Uses an appropriate style in writing Speaks clearly Takes the role of the leader in group activities Motivates his/her colleagues at work Behaves dynamically at work Looks for new activities within his/her work environment Looks for new activities within his/her work environment Is willing to commit him/herself to new tasks

Kaiser-Meyer-Olkin Measure of Sampling Adequacy = .944 Eigenvalues Factor 1 = 11.54 Factor 2 = 1.63 % of variance explained Factor 1 = 67.92 Factor 2 = 9.61

Factor 2 -.01 .08 .00 .01 .01 -.01 .96 .95 .92 .92 .68 -.02 -.08 .08 .03 .25 .03

AL LO

Item

Table 53. Factor analysis of the competencies measure

1 1.02 .97 .93 .91 .88 .81 .05 -.03 .06 -.04 .10 -.03 .05 .02 .12 -.04 .41

3 -.06 -.09 .02 .03 .05 .16 -.09 .03 -.02 .08 .14 .93 .90 .81 .79 .74 .45

Factor 3 = 1.25 Factor 3 = 7.36

Intercorrelation matrix of the factor analysis of the OCB measure Factor 1 Factor 2 Factor 3

Factor 1

Factor 2

.647 .694

.587

151

Results

AL LO

Figure 12. Scree plot of the competencies measure

The internal consistency reliabilities of the three scales were .957, .978 and .938 for the three scales respectively, which were considered very satisfactory. Similarly to the

OCB measure, the scores for each individual were computed by averaging the

perspective scores of the items for each scale.

10.2.3 The job proficiency index

Finally, as far as the job proficiency measure is concerned, a factor analysis of the 6

items generated, as expected, one factor explaining 76.75% of the variance. The factor loadings of the items ranged from .848 to .898 with Cronbachâ&#x20AC;&#x2122;s Îą of .9389 for the scale.

A total job proficiency score was computed for each employee, by averaging the scores of all six items. Summing up, the following table summarises the internal consistencies of all the

scales used in the study.

152

Results

222 222 222 222 222 200 209 200 218 218 218 217

.83 .85 .70 .67 .85 .90 .90 .71 .95 .97 .93 .93

alpha

AL LO

Extroversion Neuroticism Openness Agreeableness Conscientiousness Job satisfaction OCB-altruism OCB-conscientiousness Communication skills People orientation Action-leadership Job proficiency index

Table 54. Internal consistency (Cronbach’s α) of the study’s measures

10.3. Preliminary analyses

The next set of analyses involved the examination of the descriptive statistics of the

study’s variables, which are presented in the following table:

PR Openness

O Mean 52.00 21.40 21.00 21.46 21.47 20.71 22.04 18.09 17.63 16.48 16.05 18.47 13.68 20.54 22.94 21.62 20.79 20.00 19.28

N 215 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222

Tenure in months Warmth Gregariousness Assertiveness Activity Excitement Seeking Positive Emotions Anxiety Angry Hostility Depression Self-consciousness Impulsiveness Vulnerability Fantasy Aesthetics Feelings Actions Ideas Values

Neuroticism

Extroversion

Table 55. Descriptive statistics of the study variables

SD 61.61 2.88 3.68 4.04 3.85 3.57 3.61 4.67 4.63 4.31 4.14 3.52 4.40 4.13 4.07 3.24 3.88 3.43 4.60

Skewness 1.88 -.49 -.23 -.44 -.51 -.28 -.16 .08 .25 .27 .31 .02 .72 -.26 -.48 -.20 -.27 -.15 -.09

Kurtosis 3.41 .14 .05 -.24 .05 .00 -.26 -.66 -.40 -.06 -.03 -.36 .39 -.29 -.05 .19 -.32 -.21 -.62 (table continued)

153

.18 .12 -.36 -.07 -.07 -.14 .43 -.33 .08 -.13 -.18 .13 .09 .15 -.09 .23 .41 .35 -.68 -.21 .46 .84 -.39 .06

-.39 .08 .03 -.26 -.35 -.13 -.61 -.34 -.40 -.26 -.22 -.33 -.30 .35 -.01 -.27 -.38 -.67 -.36 -.48 -1.00 -1.28 -.49 -.63

3.17 2.85 2.70 3.78 3.71 2.88 3.32 4.50 3.28 3.55 3.35 4.03 16.12 19.54 14.91 11.91 16.83 .95 .96 .68 1.62 1.72 1.58 .77

AL LO

21.50 19.89 19.79 19.64 19.74 21.47 22.47 19.67 22.58 22.22 21.77 21.02 128.10 100.42 125.19 122.05 129.76 4.97 3.59 3.94 5.13 5.33 4.74 4.00

222 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222 219 209 200 218 218 218 218

Trust Straitforwardness Altruism Compliance Modesty Tender-mindedness Competence Order Dutifulness Achievement-striving Self-discipline Deliberation Extroversion Neuroticism Openness Agreeableness Conscientiousness Job satisfaction OCB altruism OCB-conscientiousness Communication skills People orientation Action-leadership Job proficiency index

Big-Five

Consc/sness

Agreea/ness

Results

The most notable comment regarding the descriptive statistics of the study is that the

scores of the six performance scales, especially the competencies scales and the job

proficiency index, are positively skewed and lenient, with the mean scores leaning towards the upper level of the distribution, questioning the existence of normal

distribution in these scales. The examination of the descriptive statistics of the sample

also called for examination of range restriction effects. Range restriction effects are

examined by comparing the standard deviations of the variables under research for the present study, to the standard deviations of the normative samples, where these

measures come from (Smith & Robertson, 1993). If in a particular scale the ratio of the two standard deviations is far from one, then range restriction might be present in the scale of interest. The examination of the standard deviations for range restriction effects is presented in the following table.

154

Results

Table 56. Range restriction examination of the studyâ&#x20AC;&#x2122; s scales Normative Ratio SD Warmth 2.88 3.4 0.85 Gregariousness 3.68 4.13 0.89 Assertiveness 4.04 4.16 0.97 Activity 3.85 3.72 1.04 Excitement Seeking 3.57 3.65 0.98 Positive Emotions 3.61 3.84 0.94 Anxiety 4.67 4.66 1.00 Angry Hostility 4.63 4.61 1.01 Depression 4.31 4.39 0.98 Self-consciousness 4.14 4.34 0.95 Impulsiveness 3.52 3.38 1.04 Vulnerability 4.40 4.93 0.89 Fantasy 4.13 4.15 1.00 Aesthetics 4.07 4.74 0.86 Feelings 3.24 3.57 0.91 Actions 3.88 3.89 1.00 Ideas 3.43 3.41 1.01 Values 4.60 4.5 1.02 Trust 3.17 3.37 0.94 Straitforwardness 2.85 3.34 0.86 Altruism 2.70 3.06 0.88 Compliance 3.78 4.01 0.94 Modesty 3.71 3.96 0.94 Tender-mindedness 2.88 3.17 0.91 Competence 3.32 3.64 0.91 Order 4.50 4.82 0.94 Dutifulness 3.28 3.63 0.90 Achievement-striving 3.55 3.76 0.94 Self-discipline 3.35 3.7 0.91 Deliberation 4.03 4.05 1.00 Extroversion 16.12 17.16 0.94 Neuroticism 19.54 18.96 1.03 Openness 14.91 15.69 0.95 Agreeableness 11.91 13.07 0.91 Conscientiousness 16.83 17.43 0.97 OCB altruism .96 5.6 1.07 OCB-conscientiousness .68 3.8 1.44 Note: The standard deviations for the personality scales were obtained from Tsaousis (1996), and for the citizenship scales from Organ & Konovsky (1989). No data were available for the other performance measures and the overall satisfaction scale.

Big-Five

Consc/sness

Agreea/ness

Openness

AL LO

Neuroticism

Extroversion

The findings of this table do not call for special attention as regards range restriction, since the standard deviations in the present sample are very similar to the respective deviations of the normative samples, with the exception of the OCB-conscientiousness

155

Results

scale, which is not identical in the two studies, a fact which could be attributed to the poor construct validity of the OCB-conscientiousness dimension, as the results of the earlier discussed factor analysis showed. 10.3.1 Gender-age differences It was initially worth-exploring any gender or age differences found in the sample with respect to personality, job satisfaction and the performance measures, although

specific hypotheses were not identified, because of the lack of literature concerning the behaviour of these variables in Greece. The following table presents the results of these

Table 57. Gender-age differences

Agreeableness

Conscientiousness

Job satisfaction

Mean

131.56 125.57 130.57 125.80 91.07 107.29 103.04 98.07 124.14 125.96 128.3 122.35 119.44 123.97 120.57 123.39 133.33 127.14 125.76 133.44 5.26 4.76 4.98 4.96

94 128 107 114 94 128 107 114 94 128 107 114 94 128 107 114 94 128 107 114 93 126 106 112

Neuroticism

Openness to experience

Extroversion

Gender/ Age males females 20-30 31+ males females 20-30 31+ males females 20-30 31+ males females 20-30 31+ males females 20-30 31+ males females 20-30 31+

AL LO

analyses:

Sig.

13.41 ** 17.48 15.05 * 16.88 16.80 ** 18.59 19.62 n/s 19.30 13.67 n/s 15.76 15.15 ** 14.17 11.77 ** 11.68 11.85 n/s 11.89 15.62 ** 17.26 17.01 ** 15.91 .84 ** .97 .95 n/s .95 (table continued)

156

Results

n/s n/s *

n/s n/s

1.01 .92 .93 .99 .72 .64 .68 .69 1.69 1.56 1.51 1.74 1.92 1.47 1.68 1.78 1.69 1.49 1.61 1.58 .814 .739 .76 .80

n/s

* n/s

AL LO

Males 88 3.52 females 121 3.65 OCB-altruism 20-30 97 3.62 31+ 106 3.57 males 82 3.80 females 118 4.04 OCBconscientiousness 20-30 91 4.06 31+ 103 3.85 males 94 4.91 124 5.30 Communication skills females orientation 20-30 103 5.29 31+ 109 5.03 males 94 4.87 females 124 5.67 People orientation 20-30 103 5.49 31+ 109 5.19 males 94 4.62 females 124 4.83 Action-leadership orientation 20-30 103 4.73 31+ 109 4.77 males 94 3.87 females 124 4.10 Job proficiency index 20-30 103 4.01 31+ 109 4.00 ** Difference is statistically significant at the 0.01 level * Difference is statistically significant at the 0.05 level n/s Non statistically significant differences

As far as gender differences are concerned, table 57 shows that females scored

higher than males in neuroticism and agreeableness, whereas males scored higher than

females in extroversion and conscientiousness; no differences were found with respect

to openness to experience. These results resembled partially to the gender differences found in the normative sample of the TPQue (Tsaousis, 1996), where females scored

higher than males in neuroticism, openness, agreeableness and conscientiousness. Nevertheless, it should be emphasised that the normative sample of the personality questionnaire consisted of university students, comparing subsequently directly the two samples with respect to personality differences should not be appropriate. Males also scored higher than females in the satisfaction scale, contrary to expectations which suggested that no gender differences exist in job satisfaction, a result that could be attributed to the improved job opportunities offered to men in Greece, a country where

157

Results

equal opportunities laws passed only in the early 1980s and a male-dominated organisational culture is well-established in most firms. As far as the performance measures is concerned and more specifically organisational citizenship behaviour, in line with what was expected, no differences were found as regards the personal dimension of OCB. However, this result was not replicated with regard to the impersonal dimension or conscientiousness. Employees’

immediate supervisors rated their female subordinates as scoring significantly higher

than their male colleagues, i.e. females participated more than males in activities

captured by the OCB-conscientiousness dimension. With respect to employees’

AL LO

competencies, the only statistically significant difference was found in the people orientation competencies, where supervisors rated their female subordinates as possessing in a significantly higher level this particular type of competencies, a result which conceptually is in the same line with the gender differences found in the agreeableness personality dimension, suggesting that women behave in more female-

orientated way than men. Finally, females were also rated higher than males in ratings of overall job proficiency, a result contradicting recent reviews in performance appraisal

performance measures.

(Arvey & Murphy, 1998), suggesting that no gender differences exist in overall job

With respect to age differences, Costa and McCrae (1985) using the NEO-PI found

that college students are somewhat higher than adults in neuroticism, extroversion and

openness and lower in agreeableness and conscientiousness. The results of the present

study partially replicated these findings, indicating that younger employees were more

extrovert, more open to new experiences, and less conscientious compared to “older” (i.e. above 30 years of age) colleagues. This finding, is in line with previous studies,

suggesting that older employees are more conscientious (e.g. Goldberg et al., 1998). “Younger” employees were also equally satisfied from their work compared to “older” employees, but they were rated by their supervisors as more willing to participate in contextual activities captured by the conscientiousness dimension of OCB. This latter result could be due to the fact that younger employees are with the organisation less time than “older” colleagues, and are keener to prove themselves to colleagues and supervisors leading into actions, such as arriving at work on time, or avoiding to take unnecessary time off work, described by the OCB-conscientiousness dimension. As far

158

Results

as the generic work competencies and the job proficiency index is concerned, no differences were found between the two age groups, suggesting that supervisors do not take into account their employeesâ&#x20AC;&#x2122; age when they assess their competencies and the overall job performance.

10.4. Examination of research questions

Next, the results regarding the specific hypotheses of the study will be discussed.

AL LO

The following two tables present the intercorrelation matrices of the study variables.

159

ED 4

1. Tenure in months

1.000

2. Extroversion

-.123

1.000

3. Neuroticism

.088

-.329(**)

1.000

4. Openness to experience

-.248(**)

.249(**)

-.003

1.000

5. Agreeableness

.242(**)

-.002

.056

.119

6. Conscientiousness

.148(*)

.210(**) -.526(**) -.205(**)

1.000

Pearson Correlation 5 6

LO W

Table 58. Scalesâ&#x20AC;&#x2122; inter-correlation matrix

-.003

1.000

-.048

.182(**)

1.000

.025

.086

.133

1.000

-.032

.145(*)

-.242(**)

-.156(*)

8. OCB-altruism

.161(*)

-.011

.028

.066

9. OCBconscientiousness

-.128

-.094

.110

.099

-.036

.088

-.014

.217(**)

1.000

10. Communication skills

.034

-.075

.013

.164(*)

.036

-.051

-.049

.280(**)

.163(*)

1.000

11. People orientation

.059

-.031

.068

.124

.069

-.075

-.002

.406(**)

.145(*)

.682(**)

1.000

12. Action-leadership orientation

.126

.055

-.013

.087

-.029

.024

.060

.595(**)

.193(**)

.667(**)

.761(**)

1.000

-.122

.077

.038

.101

.057

.050

.750(**)

.441(**)

.392(**)

.390(**)

.571(**)

.194(**)

13. Job proficiency index

7. Job satisfaction

R IN

** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

ED LO W

Table 59. Inter-correlation matrix between the 30 TPQue sub-scales and the performance measures Job satisfaction

Job proficiency index

Communication skills

People orientation

Action-leadership orientation

OCB-altruism

R IN

Warmth .003 -.102 -.038 .035 .017 Gregariousness .150(*) -.092 -.156(*) -.059 -.011 Assertiveness .156(*) -.015 -.005 -.016 .112 Activity .077 -.065 .021 -.014 .094 Excitement Seeking .197(**) -.149(*) -.128 -.137(*) -.053 Positive Emotions .042 -.141(*) -.041 .063 .075 Anxiety -.188(**) .119 .009 .103 .042 Angry Hostility -.277(**) -.061 -.063 -.050 -.066 Depression -.192(**) .044 .008 .109 -.016 Self-consciousness -.170(*) .116 .065 .034 -.049 Impulsiveness -.077 .059 .012 .025 .034 Vulnerability -.173(*) .080 .034 .087 .001 Fantasy -.089 .011 .092 .070 .031 Aesthetics -.135(*) .087 .171(*) .130 .081 Feelings -.222(**) .094 .088 .136(*) .080 Actions -.083 .010 .092 .074 .066 Ideas -.071 -.101 .064 .033 .042 Values -.026 .037 .109 .040 .038 Trust .008 .022 .003 -.003 .028 Straitforwardness -.138(*) .058 .039 .019 -.016 Altruism .008 .040 .024 .079 -.001 Compliance .046 .091 .022 .078 -.031 Modesty -.067 .115 .067 .049 -.050 Tender-mindedness -.053 .029 -.032 .031 -.029 Competence .276(**) .047 -.045 -.055 .062 Order .083 .077 -.019 .020 .032 Dutifulness .107 .002 -.053 -.151(*) -.065 Achievement-striving .225(**) .014 -.015 -.029 .106 Self-discipline .074 -.028 -.102 -.111 -.017 Deliberation .093 .121 -.010 -.049 -.012 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

-.049 -.010 .051 .029 -.044 -.045 .062 -.062 .010 .063 .054 .013 .040 .073 .067 .040 .014 .022 -.036 .020 .054 .045 -.009 .028 .104 .078 -.030 .121 .053 .054

OCBconscientiousness -.098 -.062 -.054 -.056 -.088 -.072 .069 .073 .097 .124 .084 .059 .075 .080 .132 .024 .055 .018 -.085 .090 -.141(*) -.008 .021 -.029 .003 .083 .116 .060 .008 .115

Results

10.4.1 Hypotheses 1, 3 The first hypothesis suggesting that conscientiousness and neuroticism would predict job performance is not confirmed with the present sample. The correlation of the job proficiency index with those two personality dimensions was too small, almost zero. More specifically, the correlation of job proficiency with conscientiousness was .057 and with neuroticism .077. Further, none of the big five personality dimensions was

related to a statistically significant level with overall job performance. Subsequently, the third hypothesis is also not confirmed, regarding the existence of curvilinear

relationship between conscientiousness and overall job performance.

AL LO

The validity of the block of the five personality dimensions was also examined, using hierarchical regression analysis (Cohen and Cohen, 1983) with the five personality dimensions forced to enter the equation in order to see the total variance in overall job performance explained by all five-factors. The results presented in the following table,

show that no more than three percent of the variance in overall job performance is

predicted.

Independent variables

Table 60. Regression analysis of the block of big-five to overall job performance Adj. R2

.222

.049

.026

2.139

Sig.

.062

B 2.087 (Constant) -.007 Extroversion .005 Neuroticism .005 Openness .006 Agreeableness .008 Conscientiousness

Sig. .057 .051 .164 .183 .206 .043

Big-Five

F change

Dependent variable: Job proficiency index

The results of this analysis replicated the bivariate correlations that no one of the five

dimensions has a statistically significant correlation with the job proficiency index,

questioning the criterion-related validity of the big-five personality constructs in the present sample. They also called for examination of the same issue on the narrower level of the facets, of which the five domains were calculated. The question of the appropriate level of analysis in the examination of the validity of personality constructs was discussed extensively in the relevant chapter. Here, a hierarchical regression analysis (Cohen and Cohen, 1983) was employed in order to

162

Results

identify the facets of the five-factor best predicting overall job performance. The results of this analysis are presented in table 61: Table 61. Hierarchical regression analysis of the 30 sub-scales of big-five to overall job performance R2 change .022

Sig.

4.677 (constant) -.032 (exc. seeking)

.000 .030

Personality Adj. R R2 subscales R2 Excitement 1 .149 .022 .017 seeking Dependent variable: Job proficiency index Step

Only one of the 30 personality sub-scales predicts overall job performance in a

statically significant level. Still, the excitement seeking sub-scale predicts less than 2%

AL LO

of the total variance in personality, suggesting that even when the low fidelity approach is adopted, the validity of the personality traits, as described by the TPQue, is very limited.

10.4.2 Hypothesis 2

As far as the second hypothesis is concerned, that extroversion and agreeableness

will be valid predictors of job performance in occupations involving some kind of interpersonal interaction, table 62 presents the inter-correlation matrix between the big-

five personality dimensions and overall job performance for this type of job families.

Table 62. Big-five and performance measures for occupations involving interpersonal interactions

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

1 2 3 4 Job proficiency index 1.000 Job satisfaction .042 1.000 Communication skills .521(**) -.151 1.000 People orientation .510(**) -.061 .747(**) 1.000 Action-leadership .681(**) .050 .696(**) .733(**) OCB-altruism .757(**) .167 .349(**) .502(**) OCB-conscientiousness .390(**) -.029 .316(**) .161(**) Conscientiousness .129 .262(**) -.097 -.085 Extroversion -.064 .150 -.217(*) -.080 Neuroticism -.003 -.230(**) .128 .102 Openness to experience .216(*) -.203(**) .253(**) .312(**) Agreeableness .218(*) .118 .066 .093 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

1.000 .684(**) 1.000 .233(**) .082 .002 .170 -.003 .089 .104 -.045 .242(**) .203(**) .058 .136

163

1.000 .080 -.132 .010 .115 -.133

Results

According to table 62 no relationship exists between extroversion and overall job performance for occupations involving interpersonal interaction. Nevertheless openness to experience and agreeableness are valid predictors of overall job performance, for this type of jobs. Especially, openness to experience shows consistent relationships to all the performance measures with the exception of OCB-altruism. This result is in partial agreement with Mount et al.â&#x20AC;&#x2122;s (1998) recent meta-analysis, who examined the effect of

personality on occupations involving interpersonal interaction, and also found that

openness to experience and agreeableness, along with conscientiousness and emotional

stability are valid predictors of overall job performance for these occupations. Agreeable employees are helpful, trusting, and friendly; attributes very useful in

AL LO

occupations where employees are in every day contact with other people and their success depends very much on how well they are getting on with other people. The effect of openness to experience was somewhat surprising, although Mount et al. (1998) also found that openness is a moderate predictor of job performance in service

jobs involving teamwork and interpersonal interaction. It is maybe the case that it is desirable for people working in positions involving interpersonal interaction to

demonstrate behaviours such as strong fantasy and creativity, and being open to new

experiences, especially in the contemporary new work environments, where the only steady characteristic is constant change. Another suggestion for the positive correlations

between openness and the performance measures for this type of occupations could be

the link usually found between this personality trait and measures of cognitive

(Ferguson & Patterson, 1998) or verbal ability (King et al., 1996). This suggestion is further supported by the positive correlations between openness and the three types of

competencies. Its positive relationship to OCB-altruism was also surprising, since it has

not been established in the past, suggesting that creative and imaginative employees are more willing to participate in extra-role activities captured by the personal dimension of OCB.

In order to examine the moderating effect of the type of job (i.e. involving or not interpersonal interactions) the respondents were divided into two groups, high and low performers, based on the mean score of the job proficiency index. The mean score of this scale was 4. One hundred and four employees (45.8%) had a mean score below four and one hundred and fourteen employees (50.2%) had a mean score above four on the

164

Results

job proficiency index. The following two figures display the moderating effect of the type of job on the relationships between openness-agreeableness and the job proficiency index.

AL LO

Figure 13. The moderating effect of job type on the openness-job performance relationship

165

Results

AL LO

Figure 14. The moderating effect of job type on the agreeableness-job performance relationship

10.4.3 Hypotheses 4, 5

With respect to the following two hypotheses, regarding relationships between

personality and job satisfaction, it was expected, building on the links between positive-

negative affectivity and big-five, that extroversion, agreeableness and neuroticism will

correlate with job satisfaction. The following table shows the inter-correlation matrix of

personality and job satisfaction for the whole sample.

Table 63. Big-five and job satisfaction 1. 2. 3. 4. 5. 6.

1 2 3 Job satisfaction 1.000 Conscientiousness .182(**) 1.000 Extroversion .145(*) .210(**) 1.000 Neuroticism -.242(**) -.526(**) -.329(**) Openness to experience -.156(*) -.205(**) .249(**) Agreeableness -.048 -.003 -.002 * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

1.000 -.003 .056

1.000 .119

166

Results

According to table 63, neuroticism had a negative correlation to job satisfaction of r = -.242 (p < .01), whereas agreeableness showed no relationship to job satisfaction, and extroversion had a statistically significant relationship of r = .145 (p < .05). These results confirmed partially the respective research questions, establishing the significance of positive and negative affect in the prediction of satisfaction. The remaining two dimensions of the big five were also correlated to job satisfaction to a

statistically significant level. Conscientiousness correlated positively to job satisfaction (r = .182, p<.05) and openness to experience negatively (r=-.156, p<.05). Conscientious

employees felt more satisfied at work, where employees who scored higher in openness

felt more dissatisfied probably because they can not express their creativity and strong

AL LO

imagination in the positions they work. A hierarchical regression analysis (Cohen and Cohen, 1983), with job satisfaction as dependent variable and the five personality dimensions entered as independent variables, showed that neuroticism and openness to experience are unique predictors of job satisfaction explaining almost 8% of the total

variance in job satisfaction, as shown in the following table.

Personality variables

Neuroticism

.242

Openness to experience

.288

Adj. R2

R2 change

.059

.054

.059

.083

.075

.025

B 6.163 (constant) -.012 (neuroticism) 7.420 (constant) -.012 (neuroticism) -.010 (openness)

Sig. .000 .000 .000 .000 .017

Step

Table 64. Hierarchical regression analysis of the big-five dimensions on job satisfaction

Dependent variable: Job satisfaction

The regression analysis (Cohen and Cohen, 1983) was repeated in order to examine

the effects of the block of big-five in job satisfaction. Tokar and Subich (1997) found

that when the five personality variables are entered as one block, they contribute significantly to the prediction of job satisfaction. The results of this analysis are presented in table 65:

167

Results

Table 65. Regression analysis of the block of big-five to job satisfaction Independent variables

Big-Five

.311

Adj. R2

.097

.076

F change

4.566

Sig.

.001

Sig.

6.465 (Constant) .007 Extroversion -.009 Neuroticism -.012 Openness -.001 Agreeableness .001 Conscientiousness

.000 .087 .017 .011 .808 .827

Dependent variable: Job satisfaction

These results replicated Tokar and Subich’ s findings (1997), regarding the

prediction of satisfaction at work from the big-five personality dimensions. Further, the

AL LO

total variance of job satisfaction explained by personality in the present study is almost twice as big as the one reported by Tokar and Subich (1997). They argued that 3-5% of satisfaction’ s variance is due to personality, where in the present study, an adjusted R2 of .076 means that almost 8% of employees’ satisfaction at workplace can be explained

by the big-five personality dimensions.

Summing up, the results regarding relationships between personality and job

satisfaction replicated the notion that neuroticism and extroversion, from all big-five personality dimensions, are valid predictors of job satisfaction. Moreover when all big

five personality dimensions are taken into account simultaneously, they can be valid predictors of job satisfaction, more than what was expected and usually found until

recently. However, the correlations found between two more of the five personality

dimensions, conscientiousness and openness to experience, call for further research on

the relationship between personality traits, as expressed by the FFM and satisfaction at

work, especially in Greece.

10.4.4 Hypotheses 6, 7 The following two hypotheses examined the moderating effect of job satisfaction to

the relationship between conscientiousness and neuroticism with overall job performance. In order to examine the moderating effects of job satisfaction, the scores of both the personality dimensions and job satisfaction have to be standardised, before creating the interaction term of personality dimension by job satisfaction. Following the guidelines of Stone (1988) and Cohen and Cohen (1983), the interaction term was entered last as

168

Results

an independent variable in a moderated multiple regression (MMR), where the other two predictors were entered separately before the interaction term. If the R2 change of the interaction term is statistically significant, then a moderating effect exists. In the following two regression analyses, the personality dimension was entered first, followed by job satisfaction and their interaction, with job performance as the dependent variable.

Adj. R2

R2 change

F change

Sig.

.057 .070 .081

.003 .005 .007

-.002 -.005 -.008

.003 .002 .002

.672 .341 .355

.413 .560 .552

1 Neuroticism .077 .006 .001 .006 2 Job satisfaction .105 .011 .001 .005 3 Interaction .134 .018 .004 .007 Dependent variable: Job proficiency index Note: The variables were standardised before entering the equation

1.239 1.051 1.474

.267 .306 .226

Model

1 2 3

AL LO

Independent variables Conscientiousness Job satisfaction Interaction

Table 66. The moderating effects of job satisfaction to the conscientiousness-neuroticism job performance relationship

significant,

job

satisfaction

Since in both cases the R2 change of the interaction terms is not statistically does

not

moderate

the

relationship

between

conscientiousness or neuroticism and job proficiency. The variance of overall job performance that is explained by the interactions of personality with job satisfaction is

very small.

10.4.5 Hypothesis 8

The next hypothesis suggested that job satisfaction would mediate the relationship

between agreeableness and job proficiency for the occupations involving some kind of

interpersonal interaction. As we saw in table 62, the correlation between agreeableness and job performance for occupations involving interpersonal interaction was .218, which was statistically significant at the .05 level. If our hypothesis is correct, this correlation will not be significant any more when the effect of job satisfaction is taken into account. In order to examine the mediating effects of agreeableness, the suggestions of James and Brett (1984) and Baron and Kenny (1986) were followed. Using multiple regression analysis, it is expected that when the effect of job satisfaction is taken into account, then

169

Results

the relationship between agreeableness and job performance will not be statistically significant anymore. In order to examine this hypothesis multiple regression procedures are employed, where the expected mediator variable is entered first and the predictor is entered second. If the R2 change is statistically significant in the case of the suspected mediator, but is not significant anymore in the case of the predictor, then the mediator variable fully mediates the relationship between the predictor and the dependent

variable. The following table presents the results of the multiple regression analysis relationship for occupations involving interpersonal interaction.

examining the mediating effect of job satisfaction in the agreeableness-job performance

Independent R R2 variables 1 Job satisfaction .042 .002 2 Agreeableness .218 .048 Dependent variable: Job proficiency index

Adj. R2

R2 change

F change

Sig.

-.007 .030

.002 .046

.194 5.261

.661 .024

Model

AL LO

Table 67. The mediating effect of job satisfaction to the agreeableness-job performance relationship for occupations involving interpersonal interactions

According to table 67, the effect of agreeableness to overall job performance in occupations involving interpersonal interaction remains intact even when the effect of

job satisfaction is taken into account. Subsequently, job satisfaction does not mediate the validity of agreeableness, in line to Mount et al.â&#x20AC;&#x2122; s (1998) recent meta-analysis, and

agreeableness is a valid predictor of overall job performance in this type of jobs,

irrespectively of the effect of job satisfaction. 10.4.6 Hypothesis 9

The following hypothesis involved the examination of the two OCB dimensions. The

literature discussed earlier identified that the personal dimension of OCB was related to agreeableness and the impersonal dimension to conscientiousness. The results of the present study question this notion, since no relationship was established between the two OCB dimensions and the respective personality dimensions, as shown in the intercorrelation matrix in table 58. Agreeableness had a correlation of .025 with OCBaltruism and conscientiousness a correlation of .088 with OCB-conscientiousness, both non-significant. However, these results are not in contrast to the meta-analytic findings of Organ and Ryan (1995) who found that when other-ratings of OCB were used, as was

170

Results

the case in the present study, zero-order correlations between personality and OCB are identified, at least as far as altruism is concerned. 10.4.7 Hypotheses 10, 11 The next two hypotheses examined the moderating and mediating effects of the two OCB dimensions in the relationship between specific personality dimensions and

overall job performance. First it was suggested that the conscientiousness dimension of OCB would moderate

the relationship between the personality dimension of conscientiousness and the job proficiency index. The results of this MMR analysis are presented in the following

AL LO

table.

Table 68. The moderating effect of OCB-conscientiousness to the conscientiousness-job performance relationship Independent R R2 Adj. R2 R2 change variables 1 Conscientiousness .057 .003 -.002 .003 OCB2 .442 .195 .187 .192 conscientiousness 3 Interaction .445 .198 .185 .002 Dependent variable: Job proficiency index Note: The variables were standardised before entering the equation

F change

Sig.

.623

.431

45.797

.000

.590

.443

Model

The MMR analysis suggested that no moderating effect exists here, since the

interaction effect is not statistically significant. The effect of the OCB-

conscientiousness dimension on job proficiency is statistically significant, as will be

discussed later, but its interaction with the personality dimension of conscientiousness is

not, a result which statistically could be due to the non-existent relationship between

conscientiousness and overall job performance in the present study. It was also expected that the OCB-altruism dimension would mediate the

relationship between agreeableness and overall job performance for occupations involving interpersonal interaction. The following table presents the results of the multiple regression analysis.

171

Results

Table 69. The mediating effect of OCB-altruism to the agreeableness-job performance relationship for occupations involving interpersonal interactions Independent R R2 variables 1 OCB-altruism .757 .572 2 Agreeableness .765 .586 Dependent variable: Job proficiency index Model

Adj. R2

R2 change

F change

Sig.

.568 .578

.572 .013

145.895 3.489

.000 .065

The inclusion of OCB-altruism prior to agreeableness to the prediction of overall job performance reduces the effect on agreeableness to statistically non-significant levels,

marginally though. Thus the hypothesis that OCB-altruism mediates the relationship

between agreeableness and job performance for occupations involving interpersonal

AL LO

interaction is confirmed. Although no relationship exists between OCB-altruism and agreeableness, one of the reasons that agreeableness has an effect on job proficiency in this type of jobs, is because it transfers its effect through the altruism dimension of OCB, which is strongly related to the measure of overall job performance in the present

study. However, this result should be interpreted with caution, since as Baron and

Kenny (1986) suggested, in the examination of mediational effects, the dependent

variable should not cause the mediator. In this case, as we will discuss later, the citizenship measures are highly inter-correlated with the measure of job proficiency

index, although a causal inference may not be made. Contrary to Baron and Kenny (1986), James and Brett (1984) claimed that it is desirable in mediation models to have

high degrees of covariation between the mediator and both the antecedents and

consequences, although again in the present study, the OCB-altruism dimension is not

related to agreeableness.

10.4.8 Hypothesis 12

The following hypothesis examined the effect of job satisfaction on the OCB

dimensions,

once

the

potential

effects

the

personality

dimensions

conscientiousness and agreeableness are controlled. Two multiple regression analyses (Cohen and Cohen, 1983) were carried out, one for each dimension of OCB. The first one regarding OCB-conscientiousness included the whole sample, where the second included only the occupations involving interpersonal interactions. The results of these regression analyses are presented in the following tables.

172

Results

Table 70. The contribution of job satisfaction to OCB-conscientiousness over conscientiousness Independent R R2 variables 1 Conscientiousness .088 .008 2 Job satisfaction .093 .009 Dependent variable: OCB-conscientiousness Model

Adj. R2

R2 change

F change

Sig.

.002 -.002

.008 .001

1.471 .175

.227 .676

Table 71. The contribution of job satisfaction to OCB-altruism over agreeableness Adj. R2

R2 change

F change

Sig.

.019 .042

.009 .024

.019 .023

2.031 2.560

.157 .113

Independent R variables 1 Agreeableness .136 2 Job satisfaction .204 Dependent variable: OCB-altruism Model

AL LO

In both cases the addition of job satisfaction as the second term in the equation did not explain any additional variance in the prediction of organisational citizenship behaviour. In other words, employeesâ&#x20AC;&#x2122; participation in extra-role activities is not a result of either their personality characteristics or their level of overall satisfaction. This result

is in contrast to the findings of Organ and Lingl (1995), who found that job satisfaction

accounts for incremental variance in OCB, after the effects of personality have been

partialled out. Subsequently, the question of what causes OCB since it is neither

personality nor job satisfaction still remains unresolved. 10.4.9 Hypotheses 13-16

The following set of hypotheses examined the relationship between personality and

generic work competencies. According to the inter-correlation matrix in table 58, none

of the three types of competencies is related to any of the personality dimensions, with

the exception of communication skills, which are positively correlated to openness to experience (r = .164, p<. 05). More specifically, it was expected that the action-

leadership competencies would be predicted by a combination of the personality dimensions of conscientiousness and extroversion. A multiple regression analysis (Cohen and Cohen, 1983) was carried out in order to examine this hypothesis, with the two personality dimensions entered as one block. The results of this analysis are presented in the following table.

173

Results

Table 72. Personality predictors of action-leadership competencies Independent variables Conscientiousness Extroversion

Adj. R2

F change

Sig.

.057

.003

-.006

.340

.712

Sig.

3.926 (constant) .005 (extroversion) .001 (consc/ness)

.000 .456 .854

Dependent variable: Action-leadership competencies

The non-significant results of this analysis indicate that the action-leadership competencies were not predicted by a combination of conscientiousness with

extroversion, as it was expected. Similar analyses including the other three personality

factors as well yielded similar results, without any of the personality dimensions

AL LO

separately or as a block to predict this type of competencies. Even when all thirty personality facets were used in an hierarchical regression analysis with actionleadership competencies as the dependent variable, no one could enter the equation, exceeding the .05 probability of F. Further, the hypotheses suggesting that people

orientation competencies would correlate positively with agreeableness and that

neuroticism would relate negatively to communication skills were also not confirmed.

No relationship was also found between this type of competencies and extroversion, as expected. These results provide support to the idea that generic work competencies are

not related to or affected by individual differences, but rather capture the ability factor

brought by employees at work.

Hypothesis 15 suggested that people orientation competencies would mediate the

relationship between agreeableness and job proficiency for occupations involving

interpersonal interaction. The results of this multiple regression analysis (Cohen and

Cohen, 1983) are presented in the following table.

Table 73. The mediating effect of people-orientation competencies to the agreeableness-job performance relationship for occupations involving interpersonal interactions Independent R R2 variables People orientation 1 .510 .260 competencies 2 Agreeableness .538 .289 Dependent variable: Job proficiency index Model

Adj. R2

R2 change

F change

Sig.

.253

.260

39.283

.000

.276

.029

4.567

.035

The analysis shows that people orientation competencies do not act as mediators in the agreeableness-job performance relationship. On the contrary, a personality

174

Results

dimension such as agreeableness adds incrementally in the prediction of overall job performance, above and beyond the effect of competencies, working as a valid predictor of overall job performance for people-orientated occupations. Identical results were produced even when all three types of competencies were entered as a block in the equation, emphasising even further the importance of agreeableness for this type of

jobs, as shown in the following table: Table 74. The effect of agreeableness on overall job performance above and beyond the effect of competencies in occupations involving interpersonal interaction Adj. R2

R2 change

F change

Sig.

.468

.453

.468

32.227

.000

.481

.032

6.916

.010

.500

AL LO

Independent R variables Action-Leadership 1 .684 People Orientation Communication skills 2 Agreeableness .707 Dependent variable: Job proficiency index Model

Agreeableness alone explains almost an additional 3% of the total variance in overall

job performance in occupations involving interpersonal interactions. This increase is

statistically significant and remains significant even when the two OCB dimensions are

taken into account by entering the equation at the second step, as shown in the following

table:

Table 75. The effect of agreeableness on overall job performance above and beyond the effect of competencies and citizenship in occupations involving interpersonal interactions Independent variables

Model

Action-Leadership .666 People Orientation Communication skills OCB-conscientiousness 2 .849 OCB-altruism 3 Agreeableness .856 Dependent variable: Job proficiency index 1

Adj. R2

R2 change

F change

Sig.

.444

.427

.444

26.097

.000

.722

.707

.277

47.843

.000

.733

.717

.012

4.192

.043

The statistical significance of agreeableness, entered in the third step, is surprising considering that earlier it was found that OCB-altruism mediates the relationship between agreeableness and overall job performance for occupations involving interpersonal interaction. Here, agreeableness explains additional variance of job performance above and beyond the effect OCB and work competencies. Maybe the

175

Results

addition of both OCB dimensions and all competencies scales earlier in the chain restored the effects of agreeableness to job performance. Nevertheless, similar results are obtained both when the order of the first two steps is reversed and also when an hierarchical approach is followed, strengthening the argument for the importance of agreeableness to occupations involving interpersonal interaction. 10.4.10

Hypothesis 17

The next hypothesis questioned the traditional issue in organisational behaviour regarding the relationship between satisfaction and performance. It was suggested,

without any claims on the causal link between them, that a positive correlation should

AL LO

be expected. According to the inter-correlation matrix in table 58, no relationship exists between satisfaction at work and overall job performance (r =.050, n/s) suggesting that, in the present sample, those two concepts are independent from each other.

10.4.11

Hypothesis 18

and their relationship to job satisfaction.

Finally, the last set of hypotheses examined the validity of the performance measures

In the research questions section it was suggested that citizenship behaviour would

correlate positively to job satisfaction. Further, it was expected that the magnitude of this correlation would be stronger than the one found between satisfaction and

performance, building on the literature of OCB that satisfied employees are more keen

to engage themselves in extra-role activities rather than in-role performance. As can be

seen in the inter-correlation matrix presented earlier (table 58), no one of the

correlations between OCB-altruism or OCB-conscientiousness with job satisfaction reached a statistically significant level, questioning the relationship between job

satisfaction and contextual performance for the present sample. 10.4.12

Hypothesis 19

This hypothesis suggested that OCB would explain additional variance on overall job performance, above and beyond the effect of job satisfaction. Although, the hypothesis expecting positive correlation between job satisfaction and job performance was not confirmed, it was expected that OCB would explain incremental variance on job performance. Two separate regression analyses (Cohen and Cohen, 1983) were carried

176

Results

out for each OCB dimension. The results of these analyses are presented in the following table. Table 76. The incremental effect of OCB to the satisfaction-performance relationship R

.050 .752

.003 .565

Adj. R2 -.002 .561

.003 .198

-.003 .190

Independent variables

1 2

Job satisfaction OCB-altruism

F change

Sig.

.502 255.966

.480 .000

.003 .196

.479 46.076

.490 .000

1 Job satisfaction .050 2 OCB-conscientiousness .445 Dependent variable: Job proficiency index

R2 change .003 .562

Model

AL LO

In both cases the OCB dimensions contributed significantly to the prediction of overall job performance, confirming the hypothesis. However, this result was not unexpected considering the high correlations of the performance measures, discussed in the following hypothesis. Hypotheses 20, 21

10.4.13

Next, relationships between the performance measures were examined. Initially, it

was suggested that all performance scales would be highly inter-correlated. This indeed is the case as can be seen from the following inter-correlation matrix, where the three

performance measures, OCB, competencies, and overall job performance are highly

inter-correlated. Especially OCB shows correlations of up to .750 for OCB-altruism with the job proficiency index, highlighting the significance of OCBs in the assessment

of job performance in the present sample. Table 77. Inter-correlation matrix of performance scales

1 2 3 Job proficiency index 1.000 Communication skills .392(**) 1.000 People orientation .390(**) .682(**) 1.000 Action-leadership orientation .571(**) .667(**) .761(**) OCB altruism .750(**) .280(**) .406(**) OCB conscientiousness .441(**) .163(*) .145(*) ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

1. 2. 3. 4. 5. 6.

1.000 .595(**) .193(**)

1.000 .217(**)

Thus, hypothesis 21 examines the validity of the citizenship and competencies scales, suggesting that OCB and the three competencies scale will be valid predictors of overall job performance. An hierarchical regression analysis (Cohen and Cohen, 1983)

177

Results

was carried out in order to investigate this hypothesis, with job proficiency index as the dependent variable. The two scales of the citizenship measure were entered in the first step and the three competencies scales in the second, in order to examine the additional effect of competencies. The results of this analysis are shown in the following table. Table 78. Hierarchical regression analysis of the citizenship and competencies measures (OCB entered first) R2

.562 .644 .669

R2 change .562 .081 .025

F change

Sig.

244.217 43.196 14.408

.000 .000 .000

AL LO

1 OCB-altruism .750 2 OCB-conscientiousness .802 3 Communication skills .818 Dependent variable: Job proficiency index

Adj. R2 .560 .640 .664

Independent variables

Model

These results suggest that the two OCB dimensions and communication skills are unique predictors of job performance. Also, communication skills explain additional variance in overall job performance, above and beyond the effect of citizenship. Similar

equation, as shown in the following table.

results are obtained when the three competencies measures are entered first in the

Table 79. Hierarchical regression analysis of the citizenship and competencies measures (competencies scales entered first) Model

Independent variables

Action-leadership .571 orientation 2 OCB-altruism .766 3 OCB-conscientiousness .813 Dependent variable: Job proficiency index

Adj. R2

R2 change

F change

Sig.

.326

.323

.326

92.006

.000

.586 .661

.582 .656

.260 .075

118.951 41.584

.000 .000

In this case, from the three competencies scales, the action-leadership orientation

enters the equation, whereas the two OCB dimensions remain unique predictors of job

performance. When all three competencies scales and the two OCB dimensions are entered in a hierarchical regression analysis, the two OCB dimensions still remain unique predictors of overall job performance, along with communication skills, as shown in the following table.

178

Results

Table 80. Hierarchical regression analysis of the citizenship and competencies measures on overall job performance Model

Independent variables

1 OCB-altruism .758 2 OCB-conscientiousness .803 3 Communication skills .812 Dependent variable: Job proficiency index

.574 .644 .659

Adj. R2 .572 .640 .654

R2 change .574 .070 .015

F change

Sig.

255.820 37.266 8.482

.000 .000 .004

Summing up, these results show that the two OCB dimensions are valid predictors of

overall job performance, irrespectively of the effect of the generic work competencies.

It seems that supervisors value the significance of citizenship behaviours at work,

AL LO

although the high correlations between them and the overall job performance measure, especially in the case of OCB-altruism, questions the ability of the supervisors, at least in the present sample, to distinguish between in-role and extra-role performance, since the two OCB dimensions predict 64% of the total variance in employeesâ&#x20AC;&#x2122; overall job performance. Apparently, Greek supervisors value very much the importance of

citizenship behaviours, suggesting that the assessment of overall job performance is very much based on contextual factors rather than work-related outcomes, or they

expect that their employees participate in extra-role behaviours, as part of their every-

day duties.

The three competencies scales were also highly correlated to the job proficiency

index. However, action-leadership competencies and communication skills were the

best predictors of job performance, with communication skills adding incremental

validity to job performance, when the effect of citizenship behaviours was taken into

account. On the other hand action-leadership competencies were considered from employeesâ&#x20AC;&#x2122; supervisors as an essential part of successful job performance, since they

predict almost 33% of the total variance (table 79) in employeesâ&#x20AC;&#x2122; overall job performance, where the other two competencies dimensions predict only an additional 1%, as the results of the following two regression analyses suggest.

179

Results

Table 81. Hierarchical regression analysis of the three competencies scales on overall job performance Model

Independent variables

1 Action-leadership skills .571 Dependent variable: Job proficiency index

.326

Adj. R2 .323

R2 change .326

F change

Sig.

104.596

.000

Table 82. Regression analysis of the block of competencies to overall job performance F change

Sig.

.324

35.595

.000

Adj. R2

AL LO

Independent variables R R2 Action-leadership .577 .333 Communication skills People orientation Dependent variable: Job proficiency index

180

Discussion

Chapter 11. Discussion The purpose of the present research was to examine the criterion-related validity of a personality measure based on the FFM, taking into consideration the effect of a number of work-related variables such as job satisfaction, organisational citizenship behaviour and generic work competencies, in Greece. Specific research questions were the criterion-related validity of a personality measure.

The discussion of the results will take the following structure:

investigated regarding the moderating-mediating effects of some of these variables in

AL LO

1. First, the criterion-related validity of personality dimensions to organisational outcomes (overall job performance, organisational citizenship behaviour and generic work competencies) will be discussed.

2. Next, the results regarding the role of job satisfaction and its relationship with

personality and organisational outcomes (overall job performance and organisational

citizenship behaviour) will be discussed.

3. Also, the findings regarding the validity and the relationships between the

performance measures will be discussed.

4. The final section will concentrate on the limitations of the study and the theoretical-

practical implications of the study.

11.1. Personality and organisational outcomes The present study investigated the criterion-related validity of personality traits in a

series of work-related outcomes such as overall job performance, organisational

citizenship behaviour and work competencies examining research findings of studies carried out in the USA and Europe, in Greece, a South-European country and member of the European Union, where none of these research questions have been explored. 11.1.1 Personality and job performance Earlier reviews on the criterion-related validity of personality traits concluded that they are very poor predictors of work performance (Schmitt et al., 1984). However, this attitude changed during 1990s with the emergence of the five-factor model, as discussed earlier. Personality traits such as conscientiousness (Barrick & Mount, 1991), and

181

Discussion

emotional stability (Salgado, 1997) were suggested to predict performance across occupations and criteria including overall job performance, where agreeableness is said to be a valid predictor of overall job performance in occupations involving interpersonal interaction and co-operation in a service environment (Hogan et al., 1984). It has to be noted that the majority of these suggestions were based on meta-analytic findings, where the results of numerous studies are accumulated controlling for various errors,

such as restriction of range, attenuation of criteria, etc. implying that it is quite likely that the results of independent, smaller studies will not necessarily be of the same

magnitude, although some studies using single samples, found results with the same

AL LO

pattern (e.g. Piedmont & Weinstein, 1994; Salgado & Rumbo, 1997).

These findings however, were not replicated in the present study. None of the bigfive dimensions was a valid predictor of the job proficiency index, used as an overall measure of job performance. No correlations were found between four of the five dimensions and the job proficiency index, with the exception of agreeableness and

openness to experience, which were valid predictors of job performance for occupations involving interpersonal interaction. Further, even when the low fidelity approach was

used, only one subscale predicted job performance; excitement-seeking, a facet of

extroversion, explaining a very small percentage of performanceâ&#x20AC;&#x2122;s overall variance. These results, at a first glance, question seriously the ability of the personality

questionnaire to predict work-related outcomes, assessed by supervisory ratings, such as

the job proficiency index and the measures of organisational citizenship behaviour and

work competencies, since non-significant correlations of very small effect sizes were detected between the five personality dimensions and the performance measures, with

the exception of openness which was a valid predictor of communication skills.

Despite the poor criterion-related validity of the personality measure in the present

study, some researchers have suggested that the validity of personality, especially the argument concerning the universal impact of conscientiousness, should be more carefully considered. Robertson and Callinan (1998), in a review of the use of personality at work settings, argued that the previously discussed meta-analytic studies treat conscientiousness in the same way as earlier studies examining the validity of general mental ability, where a positive linear association is expected between measures of mental ability and overall job performance across types of occupations. They also

182

Discussion

claimed (Robertson & Callinan, 1998) that this should not necessarily be the case for conscientiousness, suggesting that people with different personality characteristics may perform at the same level of competence, by carrying out their work tasks in various ways. They maintained that different personality characteristics may be associated with different strategies adapted by people in performing the required tasks effectively. This idea was further discussed by Robertson and Fairweather (1998), who argued that it is

more likely for personality traits to predict specific performance dimensions (competencies) rather than overall job performance. They claimed that â&#x20AC;&#x153;it is perfectly

possible to imagine similar levels of overall performance, for different jobs in the same general family, which result from substantially different capabilities in the set of

AL LO

competencies of relevance to the jobâ&#x20AC;? (p. 12), indicating that the focus of attention in future validation research in personality should be shifted into the examination of its relationship with tactics and style of behavioural competencies, rather than the overall level of performance.

These results are in line with a number of recent studies that failed to replicate relationships between Big-Five and job performance using similar predictors and

criterion measures. Nelson et al. (1999) using the same criterion measure of overall job

performance, did not find any significant relationships between the five personality dimensions and job performance, with employees working in small and medium sizes

firms (SMEs), i.e. in a sample similar to the sample used here. Similar results were also

obtained by Flint-Taylor et al. (1999), who also could not find any links between a well-

established measure of the FFM (NEO PI-R) and overall job performance. In an earlier study Hough (1992) examined the validity of her big-nine framework, as opposed to the

FFM. She found that the variants of the FFM used in her study were poorly correlated

with measures of overall job performance, in the range of .01 to .19. Schneider and Hough (1995) suggested that these results show that personality is less predictive of inrole, overall job performance, rather than contextual performance or other criteria, such as commendable behaviour and law abiding behaviour. Nevertheless, as far as the present study is concerned, personality failed to predict contextual performance as well, as measured by the OCB measure. Another reason behind the failure of personality to predict overall job performance in the present study could be due to the wide use of less valid and reliable selection

183

Discussion

methods in Greece. A study by Kantas et al. (1997) examining the selection methods in 80 privately-owned Greek SMEs, which represent the majority of the Greek private sector, showed that psychological tests (including personality questionnaires) was the least frequently used selection method, compared to more traditional and subjective methods

such

unstructured

interview,

CVs

evaluation

and

personal

recommendations. Further, Greek managers perceived psychological tests as the least

effective method, rating interview as the most effective method of all. These results are in clear contrast with the findings of numerous studies, which have established the poor

criterion-related validity of these methods, especially unstructured interviews. They also

identified the existence of a major gap between the scientific evidences for these

AL LO

selection methods and the degree of their implementation and perception of effectiveness in Greece. Similar findings were obtained from another study exploring managerial selection in Greece (Eleftheriou, 1997). Tixier (1996) in a comparative study of recruitment methods across the countries of European Union claimed that

although Greek HR managers value the importance of employees’ personality

characteristics equally to technical skills, the use of personality tests is still very limited

and treated with reservation (Tixier, 1996), an attitude that leads recruiters to elicit

inferences about candidates’ personality from employment interviews exclusively. The findings of these studies reflect a lack of knowledge from human resource

professionals in Greece about the use of appropriate recruitment methods. The adoption

of less valid and reliable methods may have direct effects on both organisational and

individual performance, partially influencing the results of the present study, as well. Further, the lack of job specifications and job descriptions in many jobs across

organisations (Papalexandris, 1992a), highlights the minor role given to job analysis in

personnel selection in Greece, which, is not considered as a required and essential first step in establishing both the characteristics of a successful applicant and the most appropriate method of selection, is also expected to flaw the present results. The most

common pattern of selection practices in Greek SMEs, which consist the 98% of firms in Greece (Papalexandris, 1992a), is the search for an “appropriate” candidate through personal contacts or colleagues in other companies, resulting into the high usage of personal recommendations (Kantas et al., 1997), followed by an informative discussion (the unstructured interview), and the final decision. This practice is not usually followed

184

Discussion

by large Greek or multinational companies based in Greece, which tend to use more objective methods (Papalexandris, 1992a). Nevertheless, the majority of the present sample was drawn from SMEs, and subsequently the results are expected to be heavily influenced from those effects. It would be quite surprising to find evidence for the criterion-related validity of specific personality dimensions, when employeesâ&#x20AC;&#x2122; during the selection and performance appraisal processes.

personality characteristics are not considered appropriate to be taken into account

Despite these criticisms, two of the big-five personality dimensions, agreeableness

and openness to experience were related to overall job performance for occupations

AL LO

involving interpersonal interactions, although this result should be interpreted with caution because of the subjective classification of the occupational titles from three raters. Especially openness showed statistically significant positive correlations with all performance scales, with the exception of the OCB-conscientiousness dimension.

The emergence of agreeableness as a valid predictor of job performance for this type

of jobs was not surprising. The moderating effect of job type in the criterion-related

validity of agreeableness was established in a number of studies (Hough, 1992; Mount et al., 1998) highlighting the importance of this personality dimension for occupations

involving interaction with other people. At these positions (e.g. teachers, sales people, managers, etc.) employees have to demonstrate the ability to get on well with either

colleagues/subordinates or customers, since successful interaction with them is usually

accompanied by successful completion of job requirements, which could be interpreted

as either better relations with colleagues/customers or even improved sales record.

The validity of openness however was not surprising considering the suggestions

made from a lot of researchers that measures of openness are positively correlated with measures of intellectual ability (Ferguson & Patterson, 1998). Mount et al. (1998) also found that openness is a moderate predictor of overall job performance for this type of jobs, but they neither provide an adequate explanation and discussion of their result nor discuss its relation to the ability domain. The relationship often found between openness to experience and cognitive/verbal ability measures (Ferguson & Patterson, 1998; King et al., 1996) which are now considered as the best predictors of overall job performance (Ree et al., 1994; Schmidt & Hunter, 1998), provide an explanation of this result

185

Discussion

although the question of why this relationship was not found for the whole sample still remains. A suggestion could be that it is desirable for employees, whose one of the main responsibilities is to successfully interact with other people, to be open to the ideas, proposals, and feelings of the people they work with. Take for example a manager, a teacher or a human resource specialist who hesitates to accept new suggestions or does not take into account the ways that colleagues or customers feel about him/her or his/her

work. This person will probably have major difficulties in establishing good personal

contacts with his/her colleagues/students/customers since it will be hard for him/her to

AL LO

him/her with, with apparent effect on his/her work behaviour.

appreciate them as individuals and accept the diversity and variety they may provide

Summing up the criterion-related validity of personality in the present study, encouraging results were found only with respect to occupations involving interpersonal interaction. No personality variable was a valid predictor of overall job performance for the whole sample, although openness and agreeableness were valid predictors of job

performance for people-orientated occupations. These results provide partial support for the use of the personality test used in the current study as a tool in selection and

assessment in Greece, advocating the necessity for further research on the criterion-

related validity of the personality instrument, as we will discuss later in more detail.

11.1.2 Personality and organisational citizenship behaviour The research findings in the field of contextual performance and organisational

citizenship behaviour were supportive for the existence of a prosocial personality i.e. a

pre-disposition in participating in extra role activities (Brief & Motowidlo, 1986). This

was especially true for two of the big-five personality dimensions, agreeableness and

conscientiousness, since they were conceptually related to organisational citizenship behaviour,

correlating

the

personal

(altruism)

and

the

impersonal

(conscientiousness or generalised compliance) aspects of OCB (Organ, 1994a; Organ, 1994b). The research evidence however is more supportive for conscientiousness than agreeableness (Organ & Lingl, 1995; Organ & Ryan, 1995). The results of the present study did not identify any links between the five personality dimensions and citizenship behaviour. Actually, no correlations existed for nine of the ten pairs, suggesting that the personality characteristics of the present sample

186

Discussion

are not predictive of OCB, as rated by employeesâ&#x20AC;&#x2122; immediate supervisors, not even for conscientiousness and agreeableness, as expected. This result may be attributed to the following reasons. First, the potential disagreement regarding the boundaries between in-role and extrarole performance, both between employees and supervisors and also between supervisors across organisations. A lot of supervisors commented on the questionnaires

that some of the citizenship behaviours described by the measure, were actually in-role performance and an integral part of the job. Those comments are further supported by

the lack of job specifications and detailed job descriptions in the majority of Greek

AL LO

companies, with the exception of multinational companies (Papalexandris, 1992a). Someone could argue that this deficiency would lead employees to extensive participation in citizenship behaviours, because of the lack of precise definitions of their duties. Morrison (1994) has provided support for these two notions arguing that there is a discrepancy between employees and managers of what is conceived to be extra-role

behaviour, and also that employees who define their job responsibilities more broadly

tend to display behaviour commonly assumed to be OCB. Managersâ&#x20AC;&#x2122; comments regarding citizenship behaviours along with the results

regarding the validity of the performance measures, which will be discussed later, suggest that Greek managers perceive contextual performance as an integral part of

employeesâ&#x20AC;&#x2122; job performance. Such a notion would be further supported from the make-

up of the sample which is largely drawn from employees working in Greek SMEs.

Employees at this type of companies are often required to carry out tasks outside their

duties because of their limited workforce at these companies. Further, managers of these companies certainly prefer to recruit people who will be keen to work overtime without

complaints or will assist their colleagues or supervisor with their duties. Lack of resources in SMEs suggests that employees have to participate in a number of activities outside their duties, which could be considered citizenship behaviours. Another reason behind the lack of association between personality and citizenship behaviour could be due to the construct validity of the OCB measure itself. It was shown earlier that although the OCB-altruism dimension had a very good factor structure with high factor loadings for all items loading on the altruism factor and

187

Discussion

acceptable internal consistency of .90, this is not the case for the OCBconscientiousness. A few of the items of this factor had quite low factor loadings and the internal consistency of the dimension was .71; just above the minimum level usually required (Nunnally, 1978). Subsequently, the poor construct validity of the OCBconscientiousness dimension may have biased the criterion-related validity of personality to this dimension of OCB. This result could be as well due to the fact that

supervisors may have a difficulty in considering behaviours captured by the OCBconscientiousness dimension as extra-role performance, although this is not the case for

OCB-altruism, which has a connotation of “helping” behaviour and is more easily

AL LO

distinguishable as extra-role behaviour.

Thus it could be argued that the lack of any relationship between the personality dimensions of conscientiousness and the impersonal dimension of OCB could be, first due to the poor construct validity of the OCB-conscientiousness dimension, compared to the OCB-altruism dimension particularly, and second because most of the behaviours

described by the impersonal dimension of OCB are very much regarded by Greek managers, as part of employees’ in-role performance. These suggestions however do not

offer an adequate explanation of the failure of agreeableness to predict the OCB-

altruism dimension. The main reason for this failure could be the fact that most of the studies examining the relationship between agreeableness and OCB-altruism usually

assess citizenship using self-ratings, and when other ratings of OCB are used, zero–

order correlations are usually found (Organ & Ryan, 1995). Subsequently, the use of

supervisory ratings in the present study may have reduced dramatically the effect of personality dispositions on citizenship behaviour, especially between agreeableness and

altruism. Nevertheless, we strongly believe that assessing OCB with self-ratings

involves the serious limitation of impression management effects, and supervisory or peer ratings of OCB should be employed, if possible. Finally, another reason for the failure to identify any links between personality and

citizenship behaviour could be approached from a social-cognitive perspective. The hypothesis that personality would validly predict citizenship was also based on the assumption that employees would engage in citizenship behaviours because of their expectations and beliefs that these would lead into secondary benefits, such as promotion, financial rewards, etc. Nevertheless, this is rarely the case in Greece where

188

Discussion

the use of performance appraisal is still very limited (Papalexandris, 1992a), especially in SMEs, and any decisions taken about promotions or pay raises favour seniority and tenure as the main criteria. Subsequently, an employee who knows that, even if s/he is willing to participate in extra role activities, s/he is not going to receive any immediate rewards at least, it is highly unlikely that s/he will spend additional time and effort for these activities, even if it is assumed that his/her personality disposition drives him/her

to behave this way.

The moderating-mediating role of citizenship behaviours in the personality-job

performance relationship was also explored, for the first time. More specifically, it was

AL LO

expected that the OCB-conscientiousness dimension would moderate the relationship between conscientiousness and overall job performance, a hypothesis which was not confirmed, and also that OCB-altruism would mediate the agreeableness-job performance relationship for occupations involving interpersonal interaction, a hypothesis confirmed from the results. The lack of a moderating effect of the OCB-

conscientiousness dimension was not surprising in the light of the finding that conscientiousness is not a valid predictor of overall job performance, as we were

expecting. Moreover, the mediating role of OCB-altruism was in line with our

expectations, since our hypothesis was that agreeableness would have an effect on job performance for this type of jobs, only through the effect of job satisfaction, OCB-

altruism and people orientation competencies, an expectation justified, at least as far as

OCB-altruism is concerned. This result suggests that agreeable employees receive

favourable ratings of job performance because they tend to participate in activities captured by the OCB-altruism dimension. Nevertheless, this result should be interpreted

with caution since no bivariate relationship exists between agreeableness and OCB-

altruism as Baron and Kenny (1986) suggested, and also the mediating effect of OCBaltruism was only marginally confirmed. 11.1.3 Personality and work competencies Few studies have investigated the criterion-related validity of personality characteristics on work competencies. The main reason being the lack of a classification of job competencies, similar to the FFM in the personality field. The development of such a taxonomy is not an easy task. Although the competency frameworks discussed

189

Discussion

earlier have a lot of common elements, which formed the basis for the development of the competencies instrument used in the present study, most competency measures also attempt to reflect the specific characteristics of the jobs concerned. Subsequently, researchers usually measure competencies via instruments or methods, such as assessment centres, that are specifically developed for use in particular positions and organisations.

A number of researchers have attempted to predict employeesâ&#x20AC;&#x2122; job competencies by their personality characteristics (Robertson & Kinder, 1993; Gibbons et al., 1995;

Nyfield et al., 1995; Robertson & Gibbons, 1996; Saville et al., 1996). Using a low

AL LO

fidelity approach, they made the hypothesis that the narrowly defined subscales of the OPQ personality questionnaire (Saville & Holdsworth, 1990), used in these studies, would predict specific work competencies better than measures of overall work performance. Overall, their results supported these notions providing support for the

fidelity approach in the examination of the validity of personality traits.

The measure of work competencies used here was specifically developed for the

current study following an extensive search of the available taxonomies and competencies measures, and rigorous tests of earlier versions of the instrument, both in

the UK and in Greece, in order to establish its reliability and content, construct and criterion-related validity. The final version of the instrument measured three types of

generic work competencies (action-leadership competencies, people orientation

competencies and communication skills). In terms of their relationship with employeesâ&#x20AC;&#x2122;

personality characteristics, it was anticipated that action-leadership competencies would

be predicted by a combination of the personality dimensions of conscientiousness with extroversion, a hypothesis not confirmed. This result is in the same vein with the earlier

discussed results failing to identify links between personality characteristics and overall job performance and citizenship behaviour. Generic work competencies, as assessed by supervisory ratings in a performance

assessment setting, were not predicted by employeesâ&#x20AC;&#x2122; personality characteristics. Contrary to expectations, conscientious and extrovert employees did not demonstrate, or from a cognitive perspective did not want to demonstrate, action-leadership competencies at work, either because they were not relevant to their position or maybe

190

Discussion

because they believed that they would not lead to any secondary benefits. The lack of job specifications and job analyses in most jobs in Greece, as discussed earlier, suggest that in many posts the existence and valuable use of the three types of competencies measured by the Generic Work Competencies Questionnaire were not related in any way to employeesâ&#x20AC;&#x2122; individual personality characteristics. Further support for the lack of a relationship between personality and work

competencies across types of jobs was provided by the failure to identify any links

between agreeableness and people orientation competencies, and also between

neuroticism and communication skills. Although it was expected that employees who

AL LO

are rated by their supervisors as possessing people orientation competencies would do so because of their personality characteristics as well, or that employees who effectively use their communication skills at work would do so mainly because of their personality characteristics, these relationships were not established.

Finally another hypothesis which also failed to confirm was about the mediating

effect of people orientation competencies to the agreeableness-job performance

relationship for occupations involving interpersonal interaction. It was earlier discussed that for employees at this type of occupations in the present sample, those who scored

higher in openness to experience and agreeableness were rated higher from their supervisors in the job proficiency index. Here, as it was earlier the case for OCB, it was

expected that people orientation competencies would mediate the criterion-related

validity of agreeableness for this type of occupations, a hypothesis not confirmed.

Agreeableness has a direct effect on measures of overall job performance, at least for

this type of jobs, irrespectively of the effect of people orientation competencies. On the contrary, agreeableness explains additional variance on job performance above and

beyond the effect of work competencies, a result further highlighting the significance of agreeableness for occupations involving interpersonal interactions. Nevertheless, when the personality-work competencies relationship is explored for

occupations involving interpersonal interactions, openness and extroversion showed a clear pattern of relationships with competencies. Openness was positively related to all three types of work competencies and extroversion negatively to communication skills. The negative correlation between extroversion and communication skills suggests that

191

Discussion

extrovert employees in people orientated positions are rated from their supervisors as lacking communication skills, maybe because as a result of their excessive activity, assertiveness and gregariousness, they induce their supervisors to attribute them as less capable to control and effectively use their communication skills. The results regarding openness provide further support to the notion that it is a distinctive and very important dimension of the FFM (Ferguson & Patterson, 1998) and is especially so for

occupations involving interaction with other people, as shown in the present study. The

question however is why this relationship was only identified for this type of jobs and

not for the entire sample, as it was earlier discussed. It is maybe the case that employees

who work at these positions and are imaginative and creative are in a more

AL LO

advantageous position to use their competencies more effectively, a fact that may also account for the positive relationship between openness and overall job performance at these occupations. This result support further the idea that openness is related to measures of general mental ability, as discussed earlier.

Summing up, the results regarding the criterion-related validity of personality on work competencies suggest that they should be considered in relation to the type of

occupations involved. The failure to establish links between personality and work

competencies for the whole sample could be attributed either to the poor recruitment methods used in Greece, as discussed earlier, or to the insufficient criterion-related

validity of the personality measure, a suggestion further supported by the poor criterion-

related validity of the personality questionnaire, when the analyses were not taking into

account the moderating effect of the job type. However, for occupations involving interpersonal interaction, openness to experience provided support on its usefulness to

predict organisational outcomes, such as generic work competencies.

11.2. Antecedents and consequences of job satisfaction At this section the criterion-related validity of personality dispositions to job

satisfaction will be first discussed followed by the discussion on the effects of job satisfaction to overall job performance and organisational citizenship behaviour. 11.2.1 Personality and job satisfaction Very few studies have explored the relationship between personality traits as expressed by the FFM and employeesâ&#x20AC;&#x2122; overall job satisfaction. Actually a search on a

192

Discussion

bibliographic database such as Psycinfo using the keywords “five-factor-personalitymodel” and “job satisfaction” identified only one study examining this issue since 1989, which was a literature review on personality and vocational behaviour (Tokar et al., 1998). This could be due to a general lack of interest in individual determinants of job satisfaction, which is considered, especially from American personality psychologists, as accompanying any employee who is lucky enough to simply have a job (Paul Costa,

personal communication, Athens, Greece, May 1997). Subsequently, we used as a guide

in this part of our research, literature and studies examining the affective (positive-

negative) determinants of job satisfaction, making inferences from positive-negative

AL LO

affectivity about the relationship between the FFM and job satisfaction.

Positive and negative affectivity were linked to extroversion (Diener, 1996; Tokar et al., 1998) and neuroticism respectively, although agreeableness is also linked positively to job satisfaction (Yang, 1997). Subsequently the initial hypotheses regarding bivariate

relationships between personality and job satisfaction were along those lines.

The statistical analyses showed that extroversion and neuroticism were indeed

related positively and negatively to work satisfaction, although this was not the case for agreeableness. Further, conscientiousness was positively related to and openness

negatively to job satisfaction, showing an unexpected pattern of relationships between these big-five dimensions and satisfaction at work. The results also showed that

neuroticism and openness to experience are unique predictors of job satisfaction and

that the block of the big-five personality dimensions predict a substantial percentage of

satisfaction’s total variance, highlighting their significance on employees’ work

satisfaction. Of course, it is impossible using a correlational research design to infer the direction of causation. It could be that extrovert, emotionally stable, conscientious, and

conventional employees tend to be happy at work or that job satisfaction may lead employees to behave in an extrovert, calm, conscientious and conservative way. The relationship between extroversion, neuroticism and job satisfaction was not

surprising. People who describe themselves as affectionate, optimistic, fun-loving and calm, relaxed and self-assured are more easily satisfied not only from their work but from their life in general (Diener, 1996). The capacity of joy and a positive and optimistic way of thinking form the core of extroversion (Costa & McCrae, 1985),

193

Discussion

explaining why these employees are more satisfied than less extrovert colleagues. As far as emotional stability is concerned, employees who score low in the neuroticism scale, are usually calm and relaxed under stressful or difficult situations, feeling secure and self-satisfied (Tsaousis, 1996), characteristics they also bring in their work environments determining to a point their levels of work satisfaction. On the other hand the positive relationship between conscientiousness and job

satisfaction was somewhat surprising. Apparently, ambitious, competent and self-

disciplined employees feel more satisfied at work because their attitude is usually

welcomed and rewarded, especially in non-managerial positions (Robertson et al., with indirect effects on levels of job satisfaction.

AL LO

1997). Subsequently, this behaviour is expected to lead to increased work opportunities,

As regards openness, the negative correlation found with job satisfaction may be due to the fact that employees who are imaginative, intellectual, and open to new

experiences, would be more difficult to get satisfied from aspects of their work A reason being that their

compared to colleagues with low scores in openness.

proneness to new experiences, or their tendency to experience feelings and emotions as an important part of their life, may be not very appropriate in a lot of posts. Further,

employees scoring high in openness are expected to be more hardly satisfied, because their imaginative and aesthetic character makes them seek for working conditions which

are “different” from what most people would expect, leading to decreased levels of job

satisfaction, when they have to cope with “normal” and steady working conditions and

job opportunities. Thus, it could be proposed as an implication for further research, that

person-job fit will mediate the relationship between openness to experience and job satisfaction. Employees who have the opportunity to constructively use their creativity

and imaginative spirit will feel more satisfied at work, compared to colleagues who do not “fit” in their position. It would be expected that this would be the case in Greece as well, where the lack of a well-established vocational guidance policy (Patiniotis & Stavroulakis, 1997) and the poor recruitment methods (Kantas et al., 1997), often lead

people to occupations incompatible to their interests and personality characteristics. An essential part of the present study was the expectation that job satisfaction would moderate the conscientiousness-job performance and neuroticism-job performance

194

Discussion

relationships. It was earlier discussed that conscientious or emotionally stable employees who feel satisfied at work, partly because of the effect of their cognitiveaffective units or due to other characteristics of the situation, would be expected to perform better than less satisfied colleagues, who have similar scores on conscientiousness or emotional stability. Thus conscientiousness (or emotional stability) would show different patterns of association with job performance for different levels of

job satisfaction. Both hypotheses were rejected. This results was not unexpected in the light of the non-significant relationships between these two personality dimensions and

job performance and also between job satisfaction and job performance. Although it is desirable in moderated multiple regression (Cohen & Cohen, 1983) the moderator

AL LO

variable to be uncorrelated with both the predictor and the criterion (Stone, 1988; Baron & Kenny, 1986), as it was the case here, in order to provide a clearly interpretable term, their cross-product (interaction) term could not explain any variance of the measure of overall job performance, suggesting the lack of any influence of job satisfaction on the

personality-job performance relationship, which was also not established in the current

study.

Job satisfaction was also examined as a mediator in the agreeableness-job

performance relationship for occupations involving interpersonal interaction. It was discussed earlier, in the research questions section, that it was expected that any effect

agreeableness could have on overall job performance for this type of occupations would

be transferred through the effect of either OCB-altruism or people orientation

competencies. The results confirmed partially the first hypothesis, regarding OCBaltruism, but rejected the case of people-orientation competencies. Similarly, job

satisfaction failed to act as a mediator in the agreeableness-job performance relationship

for people-orientated occupations. For employees at this type of jobs, agreeableness showed a pattern of clear, direct relationships to job performance, without being affected by either job satisfaction or competencies, but only slightly by OCB-altruism. This result was quite significant, highlighting the importance of agreeableness in line with earlier findings (Hough, 1992; Mount et al., 1998) suggesting that agreeableness is a valid predictor of overall job performance across positions for occupations where being considerate, respectful and getting on well with other people is considered significant.

195

Discussion

Another significant result regarding the determinants of job satisfaction was that the male participants were significantly more satisfied than their female colleagues. This result is not surprising considering the position of women in Greece, where traditionally are seen as having a family, supportive-to-husband role. This attitude has started changing recently, especially between young women and dual-career families, but the work environment is adapting very slowly. Consequently, men still have better work

opportunities and are considered as stronger candidates for recruitment and/or promotion than women, with direct effects on female employees’ level of work

AL LO

11.2.2 Consequences of job satisfaction

satisfaction.

The next set of research questions examined the effects of job satisfaction on employees’ work performance. Following the results of previous studies, it was expected that job satisfaction would be related both to overall job performance and

organisational citizenship behaviour.

Without making any inference for the direction of relationship, or causality, a

hypothesis was reported suggesting that a positive moderate correlation should be expected between job satisfaction and employees’ job performance based on earlier

meta-analysis on this issue (Iaffaldano & Muchinsky, 1985; Petty et al., 1984). This

hypothesis was not confirmed from the present results. No relationship existed between levels of job satisfaction and supervisory ratings of employees’ performance. As it will

be discussed later, supervisory ratings are heavily influenced by employees’ contextual

performance and work competencies, and not from the level of satisfaction they

demonstrate. On the other hand, employees who are (dis)satisfied do not alter their

levels of performance, according to their supervisors. This result although quite surprising, should be interpreted taking under consideration the recent socio-economic status in Greece, the weaker member of the European Union, where the primary target of all recent governments during 1990s is the financial convergence with Europe almost at any cost. This situation has direct effects on levels of employment, which although is still at a quite low level compared to other EU countries, threatens numerous positions and organisations, especially in traditional jobs and SMEs, which progressively face severe competition from foreign companies. Under such a difficult economic

196

Discussion

environment, it would not be surprising if employees were suppressing their negative feelings or thoughts regarding aspects of their work, under the suspicion that the management may examine their responses, since the questionnaires were not anonymous. This result was further supported by the generally high responses given by employees on the job satisfaction measure. It was also expected that satisfied employees would receive more favourable ratings

of citizenship behaviour in line with the suggestions made by researchers in the field of

OCB and contextual performance that job satisfaction is more likely to lead to increased

participation to citizenship behaviours, rather than in-role job performance (Organ,

AL LO

1988b). Hence, this result was not confirmed in the present study. No relationship was found between job satisfaction and ratings of OCB, although the correlation between OCB-altruism and job satisfaction failed marginally to reach a statistically significant level, suggesting that this relationship would probably be statistically significant with a larger sample. Apparently, even satisfied employees in Greek SMEs can not see any

reasons why they should participate in extra-role activities, since as discussed earlier, this behaviour does not usually guarantee any secondary benefits, such as promotion or

pay raises, with the latter usually be subject to negotiations between the confederation

of Greek workers with the Greek government and the confederation of Greek Industry in order to establish national minimum wages, which usually form the basic pay

structure across positions and organisations.

Finally, the incremental effects of job satisfaction on personality in predicting

citizenship behaviour was examined, and of OCB on job satisfaction in the prediction of

overall job performance. As far as the former is concerned, it was expected that once the effects of conscientiousness and agreeableness have been accounted, job satisfaction

would further contribute to OCB. This research question was explored following the suggestions made by Organ and Lingl (1995), who built on the individual determinants of citizenship behaviour (i.e. conscientiousness and agreeableness) and the effect of employeesâ&#x20AC;&#x2122; job satisfaction on extra-role performance. The results of the current study did not replicate their findings, an expected result considering the lack of any relationship between personality and OCB and between job satisfaction and OCB.

197

Discussion

Hence, the incremental effect of OCB on job satisfaction in predicting the job proficiency index was confirmed, a result which was due to the high inter-correlation between OCB and overall job performance, as we will discuss in the following section. Job satisfaction did not predict any variance on job performance, as expected, leaving most of the job performance variance to be explained by the remaining performance measures.

Summing up, no consequences of job satisfaction on supervisory ratings of job

performance and OCB were detected in the present study, implying its limited influence

on employees’ work-related outcomes, as assessed by their supervisors. Apparently,

AL LO

following our earlier discussion, the difficult economic situation in Greece, inhibits employees from showing their negative feelings at work, or let them affect their work behaviour. Moreover, it could be argued that managers are not aware of their employees’ satisfaction level, since an autocratic management style is quite common in Greek companies (Tixier, 1994), and the use of employees’ surveys are not a common

ground, especially in SMEs.

11.3. The validity of the performance measures

The results regarding the validity of and the relationship between the performance measures highlight the importance of contextual behaviour and work competencies in

work settings and more specifically in performance assessment and performance

management, confirming the claim that they both are valid predictors of supervisory

ratings of job performance. The hypothesis that supervisory ratings of job performance would be affected by employees’ citizenship behaviour as well as by their generic work

competencies was supported. The results suggest that both forms of OCB were positive

and unique predictors of ratings of overall job performance and also that they explain incremental variance in job performance variance above and beyond the effect of competencies. Action-leadership competencies, along with communication skills, were also unique predictors of overall job performance. As regards age and gender differences, younger employees tend to participate more in activities captured by the OCB-conscientiousness dimension whereas females received higher ratings of overall job performance, people orientation competencies and in the OCB-conscientiousness dimension.

198

Discussion

Prior research (Mackenzie et al., 1991; Lowery & Krilowicz, 1996) has provided evidence that supervisors take citizenship behaviours into consideration when evaluating subordinates’ job performance. The findings of the present study replicated these results across a number of divert organisations and positions in Greece. Lowery and Krilowicz (1996), who found similar results regarding citizenship behaviour, suggested that one of the reasons that OCB affect supervisors’ evaluations could be

because these behaviours actually contribute to organisational effectiveness, as Organ

(1988a) first suggested. They also claimed that supervisors may have a preconceived

concept of what makes an employee effective, and OCB may be a part of this concept. In the present study it seems that engaging in citizenship behaviours is considered a

AL LO

very important aspect of work performance for managers in Greek SMEs. In fact it may be more significant than employees’ work competencies in the assessment of overall work proficiency.

The expectation that a large proportion of supervisory ratings would be predicted by

the combination of citizenship behaviours and work competencies was also supported. Organisational citizenship behaviour and work competencies accounted for 66% of the

variance in overall performance ratings, highlighting the influence they exert in

supervisors’ appraisal of their subordinates. Although employees’ participation in extrarole activities that promote the organisation and assist colleagues is considered as more

important than work competencies, at least two types of work competencies such as

communication skills and action-leadership competencies are thought to be essential

parts of successful performance. The positive relationship between these two types of competencies and job performance suggests that Greek supervisors acknowledge their

importance on top of citizenship behaviours. Summing up, the finding that OCB is more

important than work competencies in predicting job performance suggests that Greek managers believe that participating in citizenship behaviours is probably more significant for organisational performance than the abilities that each individual brings

into his/her post. As far as age differences is concerned, it was found that younger employees tend to participate significantly more than older colleagues in activities captured by the impersonal dimension of OCB. This result may indicate that younger employees who have also been with their organisation less time than older colleagues are keener in

199

Discussion

exhibiting behaviours, such as avoiding taking extra breaks or having unnecessary time off work, since they are still trying to establish themselves at work, between their colleagues and supervisors. The higher ratings received by women in the OCB-conscientiousness dimension, in people orientation competencies and in overall job performance, contradicts recent suggestions that no gender differences exist in supervisory ratings of work performance

(Arvey & Murphy, 1998) implying that overall, the female employees of the study are

rated as performing better than their males counterparts, a result which was repeated

AL LO

although not all of them were statistically significant.

across all performance scales, where females received higher mean scores than males,

These results highlight the importance of the combined effects of two substantial aspects of work behaviour, organisational citizenship behaviour and generic work competencies, in the prediction of job performance. Their significance for performance

assessment was successfully established, highlighting the important role they have in

employeesâ&#x20AC;&#x2122; work behaviour across settings in a country where most companies of the

private sector are small and medium enterprises.

11.4. Limitations of the study â&#x20AC;&#x201C; Theoretical and Practical Implications

The present study has a series of limitations, which will be discussed in the following

paragraphs.

The first limitation was that the researcher was not entirely in control of the test

administration and collection. Usually this was arranged centrally through a contact person in the HR department. If the company was very small, lacking personnel

administration I usually had to do the administration and collection of the questionnaires

myself. However, although the participants were contacted during working time, they had to complete the questionnaires in their spare time, because of the length of the measures. Subsequently most employees completed their questionnaires and returned them at a later time. This may or may not have an effect on participantsâ&#x20AC;&#x2122; responses. Although the instructions usually given for test administration, especially for personality questionnaires, require that they should be completed in a controlled environment under supervision, this was not feasible in the present study. The

200

Discussion

researcher faced with numerous difficulties in getting approval from companies directors or personnel managers in permitting their employees to participate in the study in the first place, and the request to get their employees together during working hours in order to participate in a research would not be welcomed. Another limitation, originating from the same reason, was the use of the job satisfaction, and OCB measures without first carrying out a pilot study of the translated

versions in Greece, in order to establish their factor structure. Although this did not have

an effect on the job satisfaction measure, which was used as an uni-dimensional

construct with very good internal consistency, this was not the case for the OCB

AL LO

measure. The two-factor solution of the OCB measure emerged was not identical to the one proposed by Smith et al. (1983). Although this is not necessarily a problem, the generally low factor loadings of the OCB-conscientiousness dimension and the just above the accepted level internal consistency (Nunnally, 1978), questions, to a degree, the construct validity of the conscientiousness dimension at least, suggesting that a pilot

study may have been necessary. Such a pilot study could have taken the form of two self-reported measures; a personality assessment of agreeableness and conscientiousness

and the OCB measure of personal and the impersonal dimensions of OCB, in order to

establish the construct validity of the OCB measure and also get an early indication of the personality-OCB relationship as assessed by employees themselves. Ideally, the

OCB measure should have been an other-reported measure, competed by the

supervisors of the employees who completed the personality measure, along the lines of

recent research in the field of personality and OCB (Konovsky & Organ, 1996; Organ & Lingl, 1995). However, although serious thought was given to both approaches, they

were rejected in an effort to minimise the difficulties of acquiring the sample for the

main study. This was also the reason why the first pilot of the competencies measure was carried out in the UK. An alternative to that problem would be to ask the managers participating in the pilot study of the competencies measure in Greece to complete the OCB measure, as well. Again, for similar reasons this idea was dropped. The main aim of this pilot study was to establish the construct and content validity and internal consistency of the new competencies measure. The addition of items in the already quite long competencies questionnaire, where supervisors were asked to assess as many of their subordinates as they could, would decrease the likelihood of getting enough data

201

Discussion

for the development of the new measure. Summing up, our decision not to pilot the OCB measure did not provide us with major problems, since the results of the factor structure or the internal consistency of the OCB measure could have been much worse, especially considering the major difficulties we were faced with getting together the participants of the main study. A third limitation, which may have affected the results of the study, has to do with

the eponymous completion of the personality and job satisfaction measures.

Unfortunately, there was no feasible alternative to that, since the use of other

approaches, such as assigning numbers to employeesâ&#x20AC;&#x2122; questionnaires, which would be

AL LO

revealed only to the researcher, required excessive organisational input, not welcomed from organisations, which only wanted to participate with the minimum effort and human-resource expense. Nevertheless, the eponymous completion of the employeesâ&#x20AC;&#x2122; questionnaires may have had a significant effect only for the job satisfaction scores, since a number of studies have discussed extensively the limited effect of impression

management and social desirability on scores of personality questionnaires (Barrick & Mount, 1996; Ones et al., 1996). A way of exploring the effects of the eponymous

completion of the satisfaction measure would be by asking some of the respondents to

fill it in anonymously, and then compare their responses with the responses of the main sample. Unfortunately, this method would decrease the number of useful job

satisfaction questionnaires, with doubtful results, since employees would probably still

hesitate to honestly indicate their level of satisfaction, considering that the management

may still be able to find out the responsesâ&#x20AC;&#x2122; source.

Finally, another limitation could be that all the performance ratings originated from the same source (supervisors) resulting in possible contamination from common method

variance. However, the emergence of multiple factors in the factor analyses weighs against significant influence from common method variance (Begley, 1998). Further, even if it exists, there is no reason to expect that the differences in correlations between OCB, competencies and overall performance are due to the effect of common method variance, since its presence would not be expected to exert differential bias on the observed relationships.

202

Discussion

The study also has a series of theoretical and practical implications. In a theoretical level, the lack of significant correlations between personality dimensions and the performance measures justifies the view that personality measures should be used in personnel selection and assessment only after careful theoretical examination of the relevant to the jobs concerned, characteristics (Robertson & Kinder, 1993). This view was further supported from the results of the present study regarding the criterion-

related validity of agreeableness and openness for occupations involving interpersonal interaction. A confirmatory approach should always be used when exploring the

application of personality dispositions at work. The adoption of an exploratory approach in the present study was supported however, from the need to investigate in a such a

AL LO

form the application of the FFM in Greece. This view was further supported from the small but statistically significant mediating effect of OCB-altruism to the agreeablenessoverall job performance relationship for occupations involving interpersonal interactions. Finally, the relationship between four of the five personality dimensions, as

assessed by the FFM, and job satisfaction, the first result of this kind using the FFM,

suggests that individual differences play a major role and may account for a significant

variance in determining employeesâ&#x20AC;&#x2122; satisfaction.

Another serious theoretical implication of the present study was the distinct role of openness and its positive relation to most of the performance measures, replicating to a

point the well-established notion that measures of intellectual ability (as related to the

personality dimension of openness) are the best predictors of work performance

(Schmidt & Hunter, 1998). This result replicated in other studies as well (Ferguson & Patterson, 1998), provide additional evidences for the criterion-related validity of

personality traits, as expressed in the FFM, above and beyond the effect of

conscientiousness or emotional stability. Finally, we believe that the use of a strong theoretical approach, such as the

cognitive-affective personality system, in the examination of the criterion-related

validity of personality may prove very useful and should be adopted more in the future. Unfortunately, practical implications did not allow the distinctive examination of the CAPS model in parallel to the FFM and the other work-related variables examined here, in order to establish their relationships, but such a proposition is probably the most important theoretical implication of the present study, and we believe a significant

203

Discussion

suggestion for further research. A careful examination and assessment of the CAPS could explain the non-significant results found in the present study, or explain more adequately the unexpected results found and are only assumed to exist because of the effect of the cognitive-affective units of personality. As far as the practical implications of the study is concerned, the relationship between four of the five personality dimensions and job satisfaction is very significant,

since most organisations and human resource consultants concentrate almost exclusively on the situational determinants of job satisfaction (e.g. job characteristics)

ignoring the significant effect of individual differences. Another implication both in a

AL LO

theoretical and practical level is the multidimensionality and high intercorrelations of the performance measures supporting the significance of citizenship behaviours and work competencies in predicting supervisory ratings of overall job performance, a result which showed that especially in Greece the former (OCB) is a very important determinant of employeesâ&#x20AC;&#x2122; performance ratings, probably because of the nature of most

Greek companies. Nevertheless, the construct of OCB should be examined more carefully in Greece both in a theoretical and practical level, since the results of the

present study showed that extra-role performance is not considered in Greece in the

same way as in other West European countries or the USA. Finally, the most significant practical implication of this study is that it is the first of its kind carried out in Greece, a

small country where only recently people have started to become aware of the

consultancy-psychological services that may be offered from organisational

psychologists and which is still considered as a new market in terms of psychological-

personality testing.

11.5. Conclusions The main objective of the present study was to examine the role of personality traits

as expressed by the FFM at work settings in Greece. Although the results did not replicate some of the well established notions in the American-West European Big-Five literature, we believe that the findings regarding the personality-job satisfaction relationship, the significance of openness for occupations involving interpersonal interactions and the role of citizenship behaviours on supervisory ratings of job performance, along with culture-specific results, such as the lack of any relationship

204

Discussion

between job satisfaction and the performance measures, and the different position given to OCB with respect to employeesâ&#x20AC;&#x2122; job performance, may prove useful for understanding work behaviour in countries, such as Greece, where a lot of the well

AL LO

established notions in work and occupational psychology may not apply.

205

References

References Aguinis, H. & Stone-Romero, E. F. (1997). Methodological artefacts in moderated multiple regression and their effects on statistical power. Journal of Applied Psychology, 82, 192-206. Anastasi, A. (1990). Psychological testing. New York: Maxwell Macmillan.

Antonacopoulou, E. P. & FitzGerald, L. (1996). Reframing competency in management development. Human Resource Management Journal, 6, 27-47.

Armstrong, M. & Brown, D. (1998). Relating competencies to pay: the UK experience. Compensations and Benefits Review, 30, 28-39.

AL LO

Arvey, R. D. & Murphy, K. R. (1998). Performance evaluation in work settings. Annual Review of Psychology, 49, 141-168.

Arvey, R. D., Bouchard, T. J., Segal, N. L., & Abraham, L. M. (1989). Job satisfaction: environmental and genetic components. Journal of Applied Psychology, 74, 187-192.

Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Barrick, M. & Mount, M. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1-25.

Barrick, M. & Mount, M. (1993). Autonomy as a moderator of the relationships between the big five personality dimensions and job performance. Journal of Applied Psychology, 78, 111-118.

Barrick, M. R. & Mount, M. K. (1996). Effects of impression management and selfdeception on the predictive validity of personality constructs. Journal of Applied Psychology, 81, 261-272.

Barrick, M., Mount, M., & Strauss, J. (1993). Conscientiousness and performance of sales representatives: Test of the mediating effects of goal setting. Journal of Applied Psychology, 78, 715-722. Bateman, T. S. & Organ, D. W. (1983). Job satisfaction and the good soldier: the relationship between affect and employee "citizenship". Academy of Management Journal, 26, 587-595. Becker, T. E. & Randall, D. M. (1994). Validation of a measure of organisational citizenship behavior against an objective behavioral criterion. Educational and Psychological Measurement, 54, 160-167.

206

References

Begley, T. M. (1998). Coping strategies as predictors of employee distress and turnover after an organisational consolidation: a longitudinal study. Journal of Occupational and Organizational Psychology, 71, 305-329. Bentler, P. M. & Wu, E. (1997). EQS for Windows. Statistical package for Structural Equation Modelling.

Bies, R. J., Martin, C. L., & Brockner, J. (1993). Just laid off, but still a "good citizen?" Only if the process is fair. Special Issue: Organizational justice-citizenship behavior. Employee Responsibilities and Rights Journal, 6, 227-238.

Boam, R. & Sparrow, P. (1992). The rise and rationale of competency-based approaches. In R. Boam & P. Sparrow (Eds.), Designing and achieving competency, pp. 3-15. London: McGraw-Hill.

AL LO

Bouchard, T. J. (1997). Genetic influence on mental abilities, personality, vocational interests and work attitudes. In C. L. Cooper & I. T. Robertson (Eds.), International Review of Industrial and Organizational Psychology, pp. 373-396. Chichester: Wiley. Boyatzis, R. (1982). The competent manager. New York: Wiley.

Brief, A. P. & Motowidlo, S. J. (1986). Prosocial Organizational Behaviors. Academy of Management Review, 11, 710-725.

Brown, R. B. (1993). Meta-competence: a recipe for reframing the competence debate. Personnel Review, 22, 25-36.

BTEC & CNAA (1990). The assessment of management competences. Project Report.

Burgoyne, J. G. (1993). The Competence movement: Issues, stakeholders and prospects. Personnel Review, 22, 6-12.

Byrne, B. M. (1994). Structural Equation Modelling with EQS and EQS/Windows. London: SAGE.

Campbell, J. P. (1994). Alternative models of job performance and their implications for selection and classification. In M. G. Rumsey, C. B. Walker, & J. H. Harris (Eds.), Personnel selection and classification, pp. 33-51. Hillsdale, New Jersey: Lawrence Erlbaum Ass. Cattell, R., Cattell, A., & Cattell, H. (1994). Sixteen Personality Factor Questionnaire Fifth Edition (16PF5). Windsor: NFER-NELSON. Clegg, C. W. & Wall, T. D. (1981). A note on some new scales for measuring aspects of psychological well-being at work. Journal of Occupational Psychology, 54, 221-225.

207

References

Cockerill, T. (1989). The kind of competence for rapid change. Personnel Management, 21, 52-56. Cohen, J. & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences. London: Lawrence Elbraum. Collin, A. (1989). Managers' competence: rhetoric, reality and research. Personnel Review, 18, 20-26.

Conger, J. A. & Kanungo, R. N. (1988). The empowerment process: Integrating theory and practice. Academy of Management Review, 13, 471-482.

Costa, P. & McCrae, R. (1985). The NEO Personality Inventory manual. Odessa: Psychological Assessment Resources.

AL LO

Costa, P. & McCrae, R. (1989). The NEO-FFI Manual Supplement. Odessa: Psychological Assessment Resources. Costa, P. & McCrae, R. (1992). Four ways five factors are basic. Personality and Individual Differences, 13, 653-665.

Costa, P. & McCrae, R. (1993). Bullish on personality psychology. The Psychologist, 7, 302.

Craig, S. (1992). Using competencies in career development. In R. Boam & P. Sparrow (Eds.), Designing and achieving competency. A competency-based approach to developing people and organizations, pp. 111-127. London: McGraw-Hill.

Cropanzano, R., James, K., & Konovsky, M. A. (1993a). Dispositional affectivity as a predictor of work attitudes and job performance. Journal of Organizational Behavior, 14, 595-606.

Cropanzano, R., James, K., & Konovsky, M. A. (1993b). Dispositional affectivity as a predictor of work attitudes and job performance. Journal of Organizational Behavior, 14, 595-606.

Dalton, D. R. & Cosier, R. A. (1988). Psychometric properties of the organisational citizenship behavior scale. Educational and Psychological Measurement, 48, 479-482. Dekker, I., Barling, J., & Kelloway, E. K. (1996). Work force size and multifaceted job satisfaction: A cross-national study. The Journal of Social Psychology, 136, 201208.

Deluga, R. J. (1994). Supervisor trust building, leader-member exchange and organisational citizenship behaviour. Journal of Occupational and Organizational Psychology, 67, 315-326.

208

References

Diener, E. (1996). Traits Can Be Powerful, but Are Not Enough: Lessons from subjective well-being. Journal of Research in Personality, 30, 389-399. Digman, J. (1990). Personality structure: Emergence of the five factor model. Annual Review of Psychology, 41, 417-440. Dulewicz, V. (1989). Assessment centres as the route to competence. Personnel Management, 21, 56-59.

Eleftheriou, A. (1997). A survey of management selection practices in Greece. Unpublished M.Sc. Dissertation, Manchester School of Management, UMIST, Manchester, UK.

AL LO

Eysenck, H. (1991). Dimensions of personality: 16, 5 or 3?-Criteria for a taxonomic paradigm. Personality and Individual Differences, 12, 773-790. Eysenck, H. J. (1995). Trait theories of personality. In S. E. Hampson & A. M. Colman (Eds.), Individual differences and personality, pp. 40-58. London: Longman.

Eysenck, S. & Eysenck, H. (1967). Salivary response to lemon juice as a measure of introversion. Perceptual and motor skills, 24, 1047-1053.

Feltham, R. (1992). Using competencies in selection and recruitment. In R. Boam & P. Sparrow (Eds.), Designing and achieving competency. A competency-based approach to developing people and organizations, pp. 89-103. London: McGraw-Hill.

Ferguson, E. & Patterson, F. (1998). The five factor model of personality: Openness a distinct but related construct. Personality and Individual Differences, 24, 789-796.

Flint-Taylor, J., Graymja, & Robertson, I. T. (1999). The five-factor model of personality: levels of measurement and the prediction of managerial performance and attitudes. Paper presented at the Occupational Psychology Conference of the British Psychological Society, 5-7 January 1999, Blackpool, UK.

Foss, N. J. & Knusden, C. (1996). Towards a competence theory of the firm. London: Routledge.

Furnham, A. & Zacherl, M. (1986). Personality and job satisfaction. Personality and Individual Differences, 7, 453-459. George, J. M. (1992). The role of personality in organisational life: issues and evidence. Journal of Management, 18, 185-213.

Gibbons, P., Baron, H., Nyfield, G., & Robertson, I. T. (1995). The big five of personality, managerial performance, and competencies. Paper presented at the BPS Occupational Psychology Conference, Warwick, 3-5/1//1995.

209

References

Glaze, T. (1989). Cadbury's dictionary of competence. Personnel Management, 21, 44-47. Goldberg, L. (1990). An alternative "description of personality": The big five factor structure. Journal of Applied Psychology, 59, 1216-1229.

Goldberg, L. R., Sweeney, D., Merenda, P. F., & Hughes, J. E. (1998). Demographic variables and personality: The effects of gender, age, education, and ethnic/racial status on self-descriptions of personality attributes. Personality and Individual Differences, 24, 393-403.

Greatrex, J. & Philips, P. (1989). Oiling the wheels of competence. Personnel Management, 21, 36-39.

AL LO

Green, A. (1995). Verbal protocol analysis. The Psychologist, 8, 126-129.

Greenberg, J. (1993). Justice and organisational citizenship: A commentary on the state of the science. Special Issue: Organizational justice-citizenship behavior. Employee Responsibilities and Rights Journal, 6, 249-256.

Hackman, J. R. & Oldham, G. R. (1975). Development of the job diagnostic survey. Journal of Applied Psychology, 60, 159-170.

Hammer, M. & Champy, B. (1993). Re-engineering the corporation: a manifesto for business revolution. London: Nicolas Brealy Publishing.

Harris, M. M. & Schaubroeck, J. (1988). A meta-analysis of self-supervisor, selfpeer, and peer-supervisor ratings. Personnel Psychology, 41, 43-62.

Harris, M. M., Smith, D. E., & Champagne, D. (1995). A field study of performance appraisal purpose: research- versus administrative- based ratings. Personnel Psychology, 48, 151-160.

Hay, J. (1990). Managerial competences or managerial characteristics? Management Education and Development, 21, 305-315.

Hogan, J. & Ones, D. S. (1997). Conscientiousness and integrity at work. In R. Hogan, J. Johnson, & S. Briggs (Eds.), Handbook of personality psychology, pp. 849870. London: Academic Press. Hogan, J. & Roberts, B. W. (1996). Issues and non-issues in the fidelity-bandwidth trade-off. Journal of Organizational Behavior, 7, 627-637.

Hogan, J., Hogan, R., & Busch, C. (1984). How to measure service orientation. Journal of Applied Psychology, 69, 167-173. Hogan, R. (1986). Hogan Personality Inventory. Minneapolis: National computer systems.

210

References

Holmes, L. & Joyse, P. (1993). Rescuing the useful concept of managerial competence: from outcomes back to process. Personnel Review, 22, 37-52. Hornby, D. & Thomas, R. (1989). Towards a better standard of management. Personnel Management, 21, 52-55.

Hough, L. M. (1992). The "Big-Five" personality variables-construct confusion: description versus prediction. Human Performance, 5, 139-155.

Hough, L. M. (1998). Effects of intentional distortion in personality measurement and evaluation of suggested palliatives. Human Performance, 11, 209-244.

AL LO

Hough, L., Eaton, N., Dunnette, M., Kamp, J., & McCloy, R. (1990). Criterionrelated validities of personality constructs and the effect of response distortion of those validities (Monograph). Journal of Applied Psychology, 75, 581-595. Hunter, J. & Schmidt, F. (1989). Meta-analysis: facts and theories. In M. Smith & I. Robertson (Eds.), Advances in selection and assessment, pp. 203-216. Chichester: Wiley.

Iaffaldano, M. T. & Muchinsky, P. M. (1985). Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97, 251-273.

Iles, P. A. (1993). Achieving strategic coherence in HRD through competence-based management and organisation development. Personnel Review, 22, 63-80.

Jacobs, R. (1989). Getting the measure of management competence. Personnel Management, 21, 32-37.

Jacobs, R. & Solomon, T. (1977). Strategies for enhancing the prediction of job performance from job satisfaction. Journal of Applied Psychology, 62, 417-421.

James, L. R. & Brett, J. M. (1984). Mediators, moderators, and tests for mediation. Journal of Applied Psychology, 69, 307-321.

Kanji-Gopal, K. & Asher, M. (1996). 100 methods for total quality management. London: Sage. Kantas, A., Kalogera, S., & Nikolaou, I. (1997). Recruitment and selection: managerial practices and perceptions in the Greek private sector. In R. Pepermans, A. Buelens, C. J. Vinkenburg, & P. G. W. Jansen (Eds.), Managerial behaviour and practices: European research issues, pp. 81-90. Leuven: ACCO. King, L. A., Walker, L. M., & Broyles, S. J. (1996). Creativity and the five-factor model. Journal of Research in Personality, 30, 189-203. Kirkpatrick, B. (1994). The Oxford Paperback Thesaurus.

211

References

Kline, P. (1979). Psychometrics and psychology. London: Academic Press. Konovsky, M. A. & Organ, D. W. (1996). Dispositional and contextual determinants of organisational citizenship behavior. Journal of Organizational Behavior, 17, 253-266. Konovsky, M. A. & Pugh, S. D. (1994). Citizenship behavior and social exchange. Academy of Management Journal, 37, 656-669.

Lei, D., Hitt, M. A., & Bettis, R. (1996). Dynamic core competencies through metalearning and strategic context. Journal of Management, 22, 549-569.

AL LO

Lowery, C. M. & Krilowicz, T. J. (1996). Effects of organisational citizenship behaviours: Evidence from production supervisors. International Journal of Selection and Assessment, 4, 18-23.

Lundy, O. & Cowling, A. (1996). Strategic Human Resource Management. London: International Thomson Business Press.

Mackenzie, S. B., Podsakoff, P. M., & Fetter, R. (1991). Organizational citizenship behavior and objective productivity as determinants of managerial evaluations of salespersons' performance. Organizational Behavior and Human Decision Processes, 50, 123-150.

Management Charter Initiative (1993). Management Standards-Personal Competence. London: MCI.

Matthews, G. (1997). The big five as a framework for personality assessment. In N. Anderson & P. Herriot (Eds.), International handbook of selection and assessment, pp. 475-492. Chichester: Wiley.

McCrae, R. & Costa, P. (1982). Self-concept and the stability of personality: Crosssectional comparisons of self-reports and ratings. Journal of Personality and Social Psychology, 43, 1282-1292.

McCrae, R. & Costa, P. (1983). Joint factors in self-reports and ratings: neuroticism, extraversion and openness to experience. Personality and Individual Differences, 4, 245-255. McCrae, R. & Costa, P. (1987). Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology, 52, 8190. McCrae, R. & Costa, P. (1990). Personality in adulthood. New York: The Guilford Press.

212

References

McCrae, R. R. & Costa, P. T. (1996). Toward a new generation of personality theories: theoretical contexts for the five-factor model. In J. S. Wiggins (Ed.), The FiveFactor Model of Personality, pp. 51-87. London: The Guilford Press. McCrae, R. & Costa, P. (1997). Personality trait structure as a human universal. American Psychologist, 52, 509-516.

McCrae, R. & John, O. (1992). An introduction to the five-factor model and its applications. Journal of Personality, 60, 175-215.

Mischel, W. (1973). Toward a cognitive social learning reconceptualization of personality. Psychological Review, 80, 252-283.

AL LO

Mischel, W. (1993). Introduction to personality. London: Harcourt Brace College Publishers. Mischel, W. & Shoda, Y. (1995). A cognitive-affective system theory of personality: reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246-268.

Mischel, W. & Shoda, Y. (1998). Reconciling processing dynamics and personality dimensions. Annual Review of Psychology, 49, 229-258.

Moorman, R. H., Niehoff, B. P., & Organ, D. W. (1993). Treating employees fairly and organisational citizenship behavior: Sorting the effects of job satisfaction, organisational commitment, and procedural justice. Special Issue: Organizational justice-citizenship behavior. Employee Responsibilities and Rights Journal, 6, 209-225.

Morrison, E. W. (1994). Role definitions and organisational citizenship behavior: the importance of the employee's perspective. Academy of Management Journal, 37, 15431567.

Mount, M. K. & Barrick, M. R. (1995). The big five personality dimensions: implications for research and practice in human resources management. In G. R. Ferris (Ed.), Research in personnel and human resources management, pp. 153-200. London: JAI Press.

Mount, M., Barrick, M., & Strauss, P. (1994). Validity of observer ratings of the big five personality factors. Journal of Applied Psychology, 79, 272-280.

Mount, M. K., Barrick, M. R., & Stewart, G. L. (1998). Five-Factor Model of personality and performance in jobs involving interpersonal interactions. Human Performance, 11, 145-166. Necowitz, L. B. & Roznowski, M. (1994). Negative affectivity and job satisfaction: cognitive processes underlying the relationship and effects on employee behaviors. Journal of Vocational Behavior, 45, 270-294.

213

References

Nelson, A., Robertson, I. T., Walley, L., & Smith, M. (1999). Personality and work performance: some evidence from small and medium-sized firms. Paper presented at the Occupational Psychology Conference of the British Psychological Society, 5-7 January 1999, Blackpool, UK. Nikolaou, I. (1995). Personality and cognitive ability of personnel in a financial services organisation in Greece. Unpublished M.Sc. Dissertation, Manchester School of Management, UMIST, Manchester, UK.

Nunnally, J. (1978). Psychometric theory. New York: McGraw-Hill.

Norussis, M. J. (1994). SPSS Professional Statistics 6.1. Chicago: SPSS Inc.

AL LO

Nyfield, G. (1994). The big five versus specific traits. A practical evaluation. Paper presented at the 9th Annual Conference of the Society for Industrial and Organizational Psychology, Nashville, Tennessee, USA. Nyfield, G., Gibbons, P., Baron, H., & Robertson, I. T. (1995). The cross cultural validity of management assessment methods. Paper presented at the 10th Annual SIOP Conference, Orlando, USA, May 1995.

Ones, D. S. & Viswesvaran, C. (1996). Bandwidth-fidelity dilemma in personality measurement for personnel selection. Journal of Organizational Behavior, 7, 609-626.

Ones, D. S. & Viswesvaran, C. (1998). The effects of social desirability and faking on personality and integrity assessment for personnel selection. Human Performance, 11, 245-270.

Ones, D. S., Viswesvaran, C., & Schmidt, F. L. (1993). Comprehensive metaanalysis of integrity test validities: findings and implications for personnel selection and theories of job performance. Journal of Applied Psychology, 78, 679-703.

Ones, D., Mount, M., Barrick, M., & Hunter, J. (1994a-a). Personality and job performance: a critique of the Tett, Jackson, and Rothstein (1991) meta-analysis. Personnel Psychology, 47, 147-156.

Ones, D., Schmidt, F., & Viswesvaran, C. (1994b-b). Do broader personality variables predict job performance with higher validity? Paper presented at the 1994 Conference of the society for industrial and organisational psychology, Nashville, Tennessee, USA. Ones, D. S., Viswesvaran, C., & Reiss, A. D. (1996). Role of social desirability in personality testing for personnel selection: The red herring. Journal of Applied Psychology, 81, 660-679. O'Reilly, C. & Chatman, J. (1986). Organizational commitment and psychological attachment: the effects of compliance, identification, and internalization on prosocial behavior. Journal of Applied Psychology, 71, 492-499.

214

References

Organ, D. W. (1977). A reappraisal and reinterpretation of the satisfaction-causesperformance hypothesis. Academy of Management Review, 2, 46-53. Organ, D. W. (1988a). Organizational citizenship behavior. Massachusetts: Lexington Books.

Organ, D. W. (1988b). A restatement of the satisfaction-performance hypothesis. Journal of Management, 14, 547-557.

Organ, D. W. (1994a). Personality and organisational citizenship behavior. Journal of Management, 20, 465-478.

AL LO

Organ, D. W. (1994b). Organizational citizenship behaviour and the good soldier. In M. G. Rumsey, C. B. Walker, & J. H. Harris (Eds.), Personnel selection and classification, pp. 53-65. Hillsdale, New Jersey: Lawrence Erlbaum Ass. Organ, D. W. (1997). Toward an explication of "morale": in search of the m factor. In C. L. Cooper & S. E. Jackson (Eds.), Creating tomorrow's organizations. A handbook for future research in organisational behavior, pp. 375-383. Chichester: Wiley.

Organ, D. W. & Konovsky, M. (1989). Cognitive versus affective determinants of orgnizational citizenship behavior. Journal of Applied Psychology, 74, 157-164.

Organ, D. W. & Lingl, A. (1995). Personality, satisfaction, and organisational citizenship behavior. The Journal of Social Psychology, 135, 339-350.

Organ, D. W. & Paine, L. B. (in press). "A new kind of performance" for industrial and organisational psychology: recent contributions to the study of organisational citizenship behaviour. In C. L. Cooper & I. T. Robertson (Eds.), International Review of Industrial and Organizational Psychology. Chichester, UK: Wiley.

Organ, D. W. & Ryan, K. (1995). A meta-analytic review of attitudinal and dispositional predictors of organisational citizenship behavior. Special Issue: Theory and literature. Personnel Psychology, 33, 775-802.

Papalexandris, N. (1992a). Human Resource Management in Greece. Employee Relations, 14, 38-52.

Papalexandris, N. (1992b). Greece. In C. Brewster, A. Hegewisch, T. Lockhart, & L. Holden (Eds.), The European Human Resource Management Guide, pp. 229-260. London: Academic Press. Parks, J. M. & Kidder, D. L. (1991). "Till death us do part..." Changing work relationships in the 1990s. In C. L. Cooper & D. M. Rousseau (Eds.), Trends in Organizational Behavior, pp. 111-136. Chichester: Wiley. Parry, S. B. (1996). The quest for competencies. Training, 33, 48-56.

215

References

Patiniotis, N. & Stavroulakis, D. (1997). The development of vocational education policy in Greece: a critical approach. Journal of European Industrial Training, 21, 192202. Paunonen, S. V. (1998). Hierarchical organisation of personality and prediction of behavior. Journal of Personality and Social Psychology, 74, 538-556.

Pervin, L. (1993). Personality theory and research. New York: Wiley.

Petty, M. M., McGee, G. W., & Cavender, J. W. (1984). A meta-analysis of the relationship between individual job satisfaction and individual performance. Academy of Management Review, 9, 712-721.

AL LO

Piedmont, R. L. & Weinstein, H. P. (1994). Predicting supervisor ratings of job performance using the NEO Personality Inventory. Journal of Psychology, 128, 255265. Ree, M., Earles, J., & Teachout, M. (1994). Predicting job performance: not much more than g. Journal of Applied Psychology, 79, 518-524.

Revelle, W. (1995). Personality processes. Annual Review of Psychology, 46, 295328.

Robertson, I. (1992). Personnel selection methods. Workshop on psycho-social aspects of employment. Sofia, September 16-20, 1992.

Robertson, I. (1993). Personality assessment and personnel selection. European Review of Applied Psychology, 43, 187-194.

Robertson, I. T. (1996). Personnel selection and assessment. In P. Warr (Ed.), Psychology at work, pp. 121-160. London: Penguin.

Robertson, I. T. & Callinan, M. (1998). Personality and work behaviour. European Journal of Work and Organizational Psychology, 7, 317-336.

Robertson, I. T. & Fairweather, J. (1998). The role of personality in work performance: understanding the results of research. Paper presented at the joint SIOP/EAWOP symposium: Five-factor model in personnel selection and testinginternational perspectives. American Psychological Association Annual Convention, San Francisco, CA, USA, August 1998.

Robertson, I. T. & Gibbons, P. (1996). Understanding management performance. Paper presented to The British Academy of Management Conference, University of Aston, Birmingham, September 1996.

216

References

Robertson, I. & Kinder, A. (1993). Personality and job competences: The criterionrelated validity of some personality variables. Journal of Occupational and Organizational Psychology, 66, 225-244. Robertson, I. T., Iles, P. A., Gratton, L., & Sharpley, D. (1991). The impact of personnel selection and assessment methods on candidates. Human Relations, 44, 963982.

Robertson, I. T., Baron, H., Gibbons, P., MacIver, R., & Nyfield, G. (1997). Conscientiousness and job success for managers. Paper presented at the Annual Occupational Psychology Conference, British Psychological Society, Blackpool, 1997.

AL LO

Rosse, J. G., Stecher, M. D., Miller, J. L., & Levin, R. A. (1998). The impact of response distortion on pre-employment personality testing and hiring decisions. Journal of Applied Psychology, 83, 634-644. Sackett, P. R., Burris, L. R., & Callahan, C. (1989). Integrity testing for personnel selection: an update. Personnel Psychology, 42, 491-529.

Salgado, J. F. (1996). Personality and job competences: A comment on the Robertson and Kinder (1993) study. Journal of Occupational and Organizational Psychology, 69, 373-375.

Salgado, J. F. (1997). The five factor model of personality and job performance in the European Community. Journal of Applied Psychology, 82, 30-43.

Salgado, J. F. (1998). Big five personality dimensions and job performance in army and civil occupations: a European perspective. Human Performance, 11, 271-288.

Salgado, J. F. & Rumbo, A. (1997). Personality and job performance in financial services managers. International Journal of Selection and Assessment, 5, 91-100.

Saville & Holdsworth (1990). Occupational Personality Questionnaire Manual. Surrey: Esher.

Saville, P., Sik, G., Nyfield, G., Hackston, J., & MacIver, R. (1996). A demonstration of the validity of the Occupational Personality Questionnaire (OPQ) in the measurement of job competencies across time and in separate organizations. Applied Psychology: An International Review, 45, 243-262. Schmidt, F. L. & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262-274.

Schmidt, F., Ones, D., & Hunter, J. (1992). Personnel selection. Annual Review of Psychology, 43, 627-670.

217

References

Schmit, M. & Ryan, A. (1993). The big five in personnel selection: factor structure in applicant and non-applicant populations. Journal of Applied Psychology, 78, 966974. Schmitt, N. & Chan, D. (1998). Personnel selection. A theoretical approach. London: SAGE.

Schmitt, N., Gooding, R., Noe, R., & Kirsch, M. (1984). Meta-analyses of validity studies published between 1964 and 1982 and the investigation of study characteristics. Personnel Psychology, 37, 407-422.

Schnake, M. (1991). Organizational citizenship: a review, proposed model and research agenda. Human Relations, 44, 735-759.

AL LO

Schneider, R. & Hough, L. (1995). Personality and Industrial/Organizational Psychology. In C. Cooper & I. Robertson (Eds.), International Review of Industrial and Organizational Psychology, pp. 75-129. Chichester: Wiley.

Schneider, R. J., Hough, L. M., & Dunnette, M. D. (1996). Broadsided by broad traits: how to sink science in five dimensions or less. Journal of Organizational Behavior, 7, 639-655.

Schroder, H. M. (1989). Managerial competence: the key to excellence. New York: Kendall/Hunt.

Schroder, H. M. (1992). Managerial competence and style. In M. Kirton (Ed.), Adaptors and innovators, pp. 97-124. London: Routledge.

Schroder, H. M. & Cockerill, A. P. (1990). Managerial Competence and cognitive style: its implications for the management of creative scientific work. London Business School, Centre For Organisational Research, Working Paper No.002.

Shackleton, V. & Newell, S. (1997). International assessment and selection. In N. Anderson & P. Herriot (Eds.), International handbook of selection and assessment, pp. 81-95. Chichester: Wiley.

Shoda, Y. & Mischel, W. (1996). Toward a Unified, Intra-individual Dynamic Conception of Personality. Journal of Research in Personality, 30, 414-428.

Short, E. C. (1984). Competence: Inquiries into its meaning and acquisition in educational settings. Lanham: University Press of America. Cited in Collin (1989).

Singh, J., Goolsby, J. R., & Rhoads, G. K. (1994). Behavioral and psychological consequences of boundary spanning burnout for customer service representatives. Journal of Marketing Research, 31, 558-569. Smith, C. A., Organ, D. W., & Near, J. P. (1983). Organizational citizenship behavior: its nature and antecedents. Journal of Applied Psychology, 68, 653-663.

218

References

Smith, M. & Robertson, I. (1992). Assessing competencies. In R. Boam & P. Sparrow (Eds.), Designing and achieving competency, pp. 50-75. London: McGrawHill. Smith, M. & Robertson, I. (1993). The theory and practice of systematic personnel selection. London: Macmillan.

Smith, P. C., Kendall, L. M., & Hulin, C. L. (1969). The measurement of satisfaction in work and retirement. Chicago: Rand McNally.

AL LO

Sparrow, P. R. (1997). Organizational competencies: creating a strategic behavioural framework for selection and assessment. In N. Anderson & P. Herriot (Eds.), International Handbook of Selection and Assessment, pp. 343-368. Chichester: John Wiley & Sons. Sparrow, P. R. & Bognanno, M. (1993). Competency Requirement forecasting: issues for international selection and assessment. International Journal of Selection and Assessment, 1, 50-58.

Spector, P. E. (1997). Job satisfaction. Application, assessment, causes, and consequences. London: SAGE.

Stajkovic, A. D. & Luthans, F. (1998). Self-efficacy and work-related performance: a meta-analysis. Psychological Bulletin, 124, 240-261.

Stone, E. F. (1988). Moderator variables in research: a review and analysis of conceptual and methodological issues. Research in Personnel and Human Resources Management, 6, 191-229.

Stone-Romero, E. F. & Anderson, L. E. (1994). Relative power of moderated multiple regression and the comparison of subgroup correlation coefficients for detecting moderating effects. Journal of Applied Psychology, 79, 354-359.

Tang, T. L. P., Tollison, P. H., & Whiteside, H. D. (1996). The case of active and inactive quality circles. The Journal of Social Psychology, 136, 57-67.

Tansky, J. W. (1993). Justice and organisational citizenship behavior: What is the relationship? Special Issue: Organizational justice-citizenship behavior. Employee Responsibilities and Rights Journal, 6, 195-207.

Tett, R., Jackson, D., & Rothstein, M. (1991). Personality measures as predictors of job performance: a meta-analytic review. Personnel Psychology, 44, 703-742. Tett, R., Jackson, D., Rothstein, M., & Reddon, J. (1994). Meta-analysis of personality-job performance relations: a reply to Ones, Mount, Barrick and Hunter (1994). Personnel Psychology, 47, 157-172.

219

References

Thierry, H. (1998). Motivation and satisfaction. In P. J. D. Drenth, H. Thierry, & C. J. De Wolf (Eds.), Handbook of Work and Organizational Psychology, vol. 4. Organizational Psychology, pp. 253-290. Hove, East Sussex: Psychology Press. Thoms, P., Moore, K. S., & Scott, K. S. (1996). The relationship between selfefficacy for participating in self-managed work groups and the big five personality dimensions. Journal of Organizational Behavior, 17, 349-362.

Tixier, M. (1994). Management and communication style in Europe: can they be compared and matched? Employee Relations, 16, 8-26.

Tixier, M. (1996). Employers' recruitment tools across Europe. Employee Relations, 18, 69-80.

AL LO

Tokar, D. M. & Mezyldo Subich, L. (1997). Relative Contributions of Congruence and Personality Dimensions to Job Satisfaction. Journal of Vocational Behavior, 50, 482-491.

Tokar, D. M., Fischer, A. R., & Subich, L. M. (1998). Personality and vocational behavior: a selective review of the literature, 1993-1997. Journal of Vocational Behavior, 53, 115-153.

Torrington, D. & Blandamer, W. (1992). Competence, pay and performance management. In R. Boam & P. Sparrow (Eds.), Designing and achieving competency. A competency-based approach to developing people and organizations, pp. 137-145. London: McGraw-Hill.

Torrington, D. & Hall, L. (1995). Personnel management. HRM in action. London: Prentice Hall.

Tsaousis, I. (1999). The traits personality questionnaire (TPQue): A Greek measure for the five factor model. Personality and Individual Differences, 26, 271-283.

Tsaousis, Y. (1996). The psychometric assessment of personality in Greek spoken population. Unpublished Ph.D. Thesis, University of London.

Fons Van de Vijver & Hambleton, R. K. (1996). Translating tests: Some practical guidelines. European Psychologist, 1, 89-99.

Van Dyne, L., Graham, J. W., & Dienesch, R. M. (1994). Organizational citizenship behavior: construct redefinition, measurement and validation. Academy of Management Journal, 37, 765-802. Vecchio, R. P. (1998). Leader-member exchange, objective performance, employment duration, and supervisor ratings: Testing for moderation and mediation. Journal of Business and Psychology, 12, 327-341.

220

References

Viswesvaran, C., Ones, D. S., & Schmidt, F. L. (1996). Comparative analysis of the reliability of job performance ratings. Journal of Applied Psychology, 81, 557-574. Warr, P. (1996). Employee well-being. In P. Warr (Ed.), Psychology at work, pp. 224-253. London: Penguin.

Weightman, J. (1994). Competencies in action. London: IPD.

Warr, P., Cook, J., & Wall, T. (1979). Scales for the measurement of some work attitudes and aspects of psychological well-being. Journal of Occupational Psychology, 52, 129-148.

AL LO

Williams, L. J. & Anderson, S. E. (1991). Job satisfaction and organisational commitment as predictors of organisational citizenship and in-role behaviors. Journal of Management, 17, 601-617. Woodruffe, C. (1993). What is meant by a competency? Leadership and Organisation Development Journal, 14, 29-36.

Yang, Z. (1997). The influence of job characteristics and personality on job satisfaction. Unpublished Ph.D. Thesis, Manchester School of Management, UMIST, Manchester, UK.

221

Appendices

Appendix 1 Sample items of the TPQue (Tsaousis, 1999; 1996)

Warmth: • Many people think of me as a friendly and warm person. • I do not like to talk with other people (R). • Some times, I have the impression that I am a cold and distant person (R). • I am usually involved emotionally in my friends' problems.

Neuroticism

Extroversion

AL LO

Anxiety: • I am an anxious person. • I am rarely worried about things that might happen in the future (R). • I have a lot of fears. • I am known as a person who is not afraid easily (R).

Openness to experience

Agreeableness

Fantasy: • I consider myself as a person with rich and active imagination. • I rarely daydream (R). • I believe that people who daydream usually are people who nobody can take them seriously (R). • When I was a kid I used to have many imaginary friends.

Trust: • I believe that most people are honest and with god intentions. • I believe that even bad people have some good and honest feelings. • Some people think of me as a cynical person (R). • I rarely trust other people (R).

Conscientiousness

Competence: • Quite often I feel unprepared to deal with a difficult situation (R). • I am a very effective person. • I am consider as a prudent person. • I usually avoid to take the responsibility to carry out difficult or crucial cases (R).

(R) denotes reversed scored items

222

Appendices

Appendix 2 The Job Satisfaction Scale (Warr, Cook & Wall, 1979) Instructions: The following statements concern various aspects of your job. Please tell us how satisfied or dissatisfied you feel with each of the features of your present job by putting in a circle one of the numbers following each statement. Give one answer for every statement. There are no right or wrong answers. Please give your honest opinion

on each one of the statements. Please do not discuss your replies with others. It is your opinion which matters. Even though it may be hard to decide, be sure not to miss any

I am not sure

I am moderately satisfied

I am very satisfied

I am extremely satisfied

1 1 1

2 2 2

3 3 3

4 4 4

5 5 5

6 6 6

7 7 7

7. Your rate of pay 8. The opportunity to use your abilities 9. Industrial relations between management and workers in your firm

1 1

2 2

3 3

4 4

5 5

6 6

7 7

10. Your chance of promotion 11. The way your firm is managed 12. The attention paid to the suggestions you make

1 1

2 2

3 3

4 4

5 5

6 6

7 7

13. Your hours of work 14. The amount of variety in your job 15. Your job security

1 1 1

2 2 2

3 3 3

4 4 4

5 5 5

6 6 6

7 7 7

1. The physical work conditions 2. The freedom to choose your own method of working

I am very dissatisfied 2

I am extremely dissatisfied

I am moderately dissatisfied

AL LO

questions out. Thank you for your help.

Your fellow workers The recognition you get for good work Your immediate boss The amount of responsibility you are given

3. 4. 5. 6.

223

Appendices

Appendix 3 Measure of Organisational Citizenship Behaviour (Smith et al., 1983) Instructions: Circle the figure (5,4,3,2,1) that best describes how characteristic this behaviour is of the employee you are rating: 4

Not at all characteristic

5 5

4 4

3 3

2 2

1 1

5 5

4 4

3 3

2 2

1 1

5 5

4 4

3 3

2 2

1 1

AL LO

1. Helps other employees with their work when they have been absent 2. Exhibits punctuality in arriving at work on time in the morning and after lunch breaks 3. Volunteers to do things not formally required by the job 4. Takes undeserved work breaks (r) 5. Takes the initiative to orient new employees to the department even though it is not part of his/her job description 6. Exhibits attendance at work beyond the norm, for example, takes less days off than most individuals or less than allowed 7. Helps others when their work load increases (assists others until they get over their hurdles) 8. Coasts towards the end of the day (r) 9. Gives advance notice if unable to come to work 10. Spends a great deal of time in personal telephone conversations (r) 11. Does not take unnecessary time off work 12. Assists me with my duties 13. Makes innovative suggestions to improve the overall quality of the department 14. Does not take extra breaks 15. Willingly attends functions not required by the organisation, but helps in its overall image 16. Does not spend a great deal of time in idle conversation Note: (r) denotes reverse scoring

Very characteristic

224

Appendices

Appendix 4 The job proficiency index (Robertson and Gibbons, 1996)

Please rate this employee on each of the following statements using the scale: Unsure

Agree

Strongly agree

Disagree

This employee: 1. Achieves the objectives of the job

AL LO

2. Demonstrates expertise in all aspects of the job

Strongly disagree 1

3. Fulfils all the requirements of the job 4. Is competent in all areas of the job

5. Performs well in the job overall

6. Accomplishes all that is required in the post

225

Appendices

Appendix 5 The questionnaire booklet used in the UK pilot study IOANNIS NIKOLAOU, BSc, MSc (PhD cand.) Organizational Psychologist Manchester January 1997 Dear Sir / Ms,

I am a Doctoral student with Manchester School of Management, UMIST. My project is in the field of personnel selection, examining specifically the moderating role of generic work competencies in the personality - job performance relationship. A part of my research is the development of an instrument measuring work competencies; here is where I need your assistance.

AL LO

I would be grateful if you could spare some time - not more than 20 minutes - in order to complete the following questionnaire with regard to a few of your subordinates. Please rate each subordinate in each statement using a 0 to 7 scale. You should put the appropriate number to the corresponding cell for each employee on the right of every statement. An example of an item is the following : This employee displays self-confidence during his/her work duties

More than 98% of his/her colleagues in the same role in the organisation More than 84% of his/her colleagues in the same role in the organisation More than 67% of his/her colleagues in the same role in the organisation More than 50% of his/her colleagues in the same role in the organisation More than 33% of his/her colleagues in the same role in the organisation More than 16% of his/her colleagues in the same role in the organisation More than 2% of his/her colleagues in the same role in the organisation Less than 2% of his/her colleagues in the same role in the organisation

7 6 5 4 3 2 1 0

In order to rate your employees to this competency in a 0-7 scale you should bear in mind the following table, where each number stands for a different category of the employeeâ&#x20AC;&#x2122;s performance in comparison to his/her colleagues in the same role in the organisation:

More specifically, the higher the score of the employee in that competency, the fewer people in the same role in the organisation possess that competency more than the ratee, and the lower the score of the employee in that competency, the more people in the same role in the organisation possess that competency more than the ratee. Finally, at the last page of the questionnaire I ask you to complete a very short job proficiency measurement, for the same employees you have already rated. I have to assure you that the results of this study are entirely confidential and that the employees you are assessing will not become by any means aware of your ratings. If you wish to receive feedback of the results please fill in your details at the end of the questionnaire. Thank you very much for your co-operation

Ioannis Nikolaou (PhD student) tel. 0161-2363311-2213 (ext.)

Prof. Ivan T. Robertson Head of Manchester School of Management, UMIST

226

Appendices

This employee: More than 98% of his/her colleagues in the same role in the organisation

More than 84% of his/her colleagues in the same role in the organisation

AL LO

Employee A B C D E presents information in a well-organised and logical way demonstrates clarity of thought both in writing and speaking maintains awareness of the degree of the task accomplishment shows energy during the accomplishment of his/her tasks shows decisiveness in decision-making situations shows positive feelings towards his/her colleagues speaks clearly remains focused on his/her tasks responds with spontaneity to new conditions takes the role of the leader in group activities interacts with other people in a sensitive way delivers effective oral presentations examines an issue thoroughly before reaching a decision behaves dynamically at work is capable of persuading his/her colleagues has good relations with most of his/her colleagues uses correct grammar in writing follows the rules and principles of the organisation is proactive (i.e. spontaneously takes action in order to achieve something at work) remains calm in difficult and stressful situations s/he tries to help her/his colleagues with their problems speaks fluently works with proficiency looks for stimulation at work influences his/her colleagues

More than 33% of his/her colleagues in the same role in the organisation

More than 16% of his/her colleagues in the same role in the organisation

More than 2% of his/her colleagues in the same role in the organisation

Less than 2% of his/her colleagues in the same role in the organisation

More than 50% of his/her colleagues in the same role in the organisation

More than 67% of his/her colleagues in the same role in the organisation

227

Appendices

This employee: More than 98% of his/her colleagues in the same role in the organisation

More than 84% of his/her colleagues in the same role in the organisation

More than 50% of his/her colleagues in the same role in the organisation

More than 33% of his/her colleagues in the same role in the organisation

More than 16% of his/her colleagues in the same role in the organisation

More than 2% of his/her colleagues in the same role in the organisation

does not include irrelevant material both in writing and speaking directs his/her colleagues positively shows consideration for his/her colleagues uses correct spelling in writing shows friendly behaviour within the organisation

More than 67% of his/her colleagues in the same role in the organisation

AL LO

Employee A B C D E takes into account the suggestions of his/her colleagues uses an appropriate style in writing follows tasks through to completion is active most of the time motivates his/her colleagues at work respects his/her colleagues uses suitable language both in writing and speaking attempts to reach the required standards of his/her position is willing to commit himself/herself to new tasks leads his/her colleagues effectively cares about the problems of his/her colleagues adapts a different style of language depending on the target audience accomplishes his/her duties in a systematic way looks for new activities within his/her work environment inspires his/her colleagues will do some extra-work in order to help his/her colleagues accomplish their tasks covers the main points of an issue (e.g. during an oral presentation) organises in advance his/her plans to achieve goals remains lively throughout his/her time at work "shows the way" to his/her colleagues displays kindness towards his/her colleagues

Less than 2% of his/her colleagues in the same role in the organisation

228

Appendices

Job Proficiency Measurement Please rate the same employees, in the same order please, on each of the following statements using the scale: Disagree

Unsure

Agree

Strongly Agree

Strongly Disagree

You should mark your responses next to each statement:

AL LO

This employee: Employee 1. achieves the objectives of the job 2. demonstrates expertise in all aspects of the job

3. fulfils all the requirements of the job

4. is competent in all areas of the job

5. performs well in the job overall

6. accomplishes all that is required in the post

Thank you very much for your assistance

229

Appendices

Appendix 6 The employees’ questionnaire booklet (cover letter and job satisfaction measure) of the study ΙΩΑΝΝΗΣ ΝΙΚΟΛΑΟΥ, Μ.Sc. Διδακτορικός Φοιτητής Ψυχολογίας Ευριπίδου 39, 185 32, Πειραιάς Αγαπητέ(ή) κύριε(α),

AL LO

ονομάζομαι Ιωάννης Νικολάου. Είμαι Εργασιακός Ψυχολόγος, απόφοιτος του τμήματος Ψυχολογίας του πανεπιστημίου Κρήτης κι από το 1995, με υποτροφία του ΙΚΥ, κάνω το διδακτορικό μου στη Σχολή Διοίκησης Επιχειρήσεων του Μάντσεστερ στην Αγγλία, μελετώντας το ρόλο των ερωτηματολογίων προσωπικότητας και της επαγελματικής ικανοποίησης στο χώρο της επιλογής και αξιολόγησης προσωπικού σε ελληνικές εταιρείες. Η εταιρεία σας μου προτάθηκε, σαν μία επιχείρηση που ίσως να ενδιαφέρονταν να λάβει μέρος στην πραγματοποίηση αυτής της έρευνας. Εφόσον επιθυμείτε να συμμετάσχετε στην έρευνα θα ήθελα να συμπληρώσετε τα δύο ερωτηματολόγια που ακολουθούν. Κάθε ένα έχει διαφορετικές οδηγίες για τη συμπλήρωσή του, τις οποίες θα ήθελα να διαβάσετε προσεκτικά πριν απαντήσετε. Ζητώντας σας να λάβετε μέρος σε αυτή την έρευνα επιθυμώ να αποκομίσετε και εσείς κάποιο όφελος από αυτή και όχι απλά να πάρω εγώ τα στοιχεία που χρειάζομαι για την έρευνά μου χωρίς να σας προσφέρω τίποτα. Για το λόγο αυτό είναι απαραίτητο να συμπληρώσετε τα στοιχεία που ακολουθούν. Με τον τρόπο αυτό στη συνέχεια θα έρθω σε επικοινωνία μαζί σας ώστε να σας ενημερώσω ατομικά για τα αποτελέσματα της έρευνας που σας αφορούν. Θα ήθελα να απαντήσετε στις ερωτήσεις των ερωτηματολογίων με ειλικρίνια. Θεωρείται αυτονόητο ότι θα τηρηθεί το απόρρητο σχετικά με τις ατομικές σας απαντήσεις. Η διεύθυνση της εταιρείας θα λάβει γνώση των αποτελεσμάτων σε συνολικό-συλλογικό επίπεδο και μόνο.

Ονοματεπώνυμο: ___________________________________________________ Τίτλος θέσης: ______________________________________________________ Τμήμα: ____________________________________________________________ Ηλικία: 20-30 31-40 41-50 51-60 61+ Πόσο καιρό εργάζεστε στην τωρινή σας θέση; ________________

Όταν συμπληρώσετε τα ερωτηματολόγια βάλτε τα μέσα στο φάκελο που σας δόθηκε, σφραγίστε τον και επιστρέψτε τον σε μένα. Για οτιδήποτε πληροφορία ή βοήθεια χρειαστείτε θα σας παρακαλούσα να μη διστάσετε να έρθετε σε επικοινωνία μαζί μου. Σας ευχαριστώ πολύ για τη συμμετοχή σας

230

Appendices

Από τους συναδέλφους σας Από την αναγνώριση που παίρνετε για την απόδοσή σας Από τον/την άμεσα προϊστάμενό/ή σας Από το μέγεθος των αρμοδιοτήτων που σας δίνονται Από το μισθό σας Από τη δυνατότητα να χρησιμοποιείτε τις ικανότητές σας Από τις εργασιακές σχέσεις ανάμεσα στη διοίκηση και στους εργαζόμενους στην επιχείρηση Από τις πιθανότητες προαγωγής σας Από τον τρόπο που διοικείται η επιχείρηση Από την προσοχή που δίνεται στις προτάσεις που κάνετε Από τις ώρες εργασίας σας Από την ποικιλία στην εργασία σας Από τη σιγουριά που σας δίνει η εργασία σας (π.χ. ότι δε θα απολυθείτε ξαφνικά)

Είμαι μέτρια ικανοποιημένος/η Είμαι πολύ ικανοποιημένος/η Είμαι πάρα πολύ ικανοποιημέος/η

Δεν είμαι βέβαιος/η

Από τις συνθήκες εργασίας σας Από την ελευθερία να επιλέγετε οι ίδιοι τον τρόπο εργασίας σας

AL LO

Πόσο ικανοποιημένος/η είστε:

Είμαι πάρα πολύ δυσαρεστημένος/η Είμαι πολύ δυσαρεστημένος/η Είμαι μέτρια δυσαρεστημένος/η

Οδηγίες: Οι προτάσεις που ακολουθούν αναφέρονται σε μερικές πλευρές της εργασίας σας. Σημειώστε πόσο ικανοποιημένος/η ή δυσαρεστημένος/η αισθάνεστε για κάθε μία από αυτές σχετικά με την τωρινή σας εργασία βάζοντας σε κύκλο έναν από τους αριθμούς που ακολουθούν σε κάθε πρόταση. Δώστε μία απάντηση για κάθε πρόταση. Δεν υπάρχουν σωστές και λάθος απαντήσεις. Σας παρακαλώ να δώσετε την ειλικρινή σας γνώμη για κάθε μία από τις προτάσεις και να μη συζητήσετε τις απαντήσεις σας με άλλους. Η δική σας γνώμη είναι αυτή που μας ενδιαφέρει. Ακόμα κι αν είναι δύσκολο να αποφασίσετε, βεβαιωθείτε ότι δεν παραλείψατε καμία ερώτηση. Σας ευχαριστώ πολύ για τη βοήθειά σας.

1 1

2 2

3 3

4 4

5 5

6 6

7 7

1 1

2 2

3 3

4 4

5 5

6 6

7 7

231

Appendices

Appendix 7 The supervisors’ questionnaire booklet (cover letter, OCB measure, job proficiency index and the competencies measure) of the study ΙΩΑΝΝΗΣ ΝΙΚΟΛΑΟΥ, Μ.Sc. Διδακτορικός Φοιτητής Ψυχολογίας Ευριπίδου 39, 185 32, Πειραιάς Αγαπητ

AL LO

Όπως ίσως γνωρίζετε, η εταιρεία που εργάζεστε είναι μία από τις ελληνικές επιχειρήσεις που συμμετέχουν στην έρευνα που κάνω στα πλαίσια του διδακτορικού μου, που πραγματοποιώ στη Σχολή Διοίκησης Επιχειρήσεων του Μάντσεστερ, UMIST, στην Αγγλία και το θέμα του οποίου περιλαμβάνεται στο χώρο της επιλογής και αξιολόγησης προσωπικού. Πιο συγκεκριμένα μελετάω την επίδραση που έχουν οι γενικές επαγγελματικές δεξιότητες στη σχέση ανάμεσα στα χαρακτηριστικά της προσωπικότητας των εργαζομένων και την αποδοτικότητα στην εργασία τους, όπως αυτή αξιολογείται από τους άμεσα προϊσταμένους τους. Στην πρώτη φάση αυτής της έρευνας ζήτησα από έναν αριθμό εργαζομένων σας να συμπληρώσουν δύο ερωτηματολόγια, προσωπικότητας και επαγγελματικής ικανοποίησης. Τα ερωτηματολόγια αυτά μου έχουν τώρα επιστραφεί. Για την ολοκλήρωση όμως της έρευνας, είναι όχι απλώς απαραίτητο αλλά αναγκαίο, οι άμεσα προϊστάμενοι αυτών των εργαζομένων, δηλαδή εσείς, να τους αξιολογήσετε σε μία σειρά ικανοτήτων-δεξιοτήτων και συμπεριφορών που θα βρείτε στα μικρά ερωτηματολόγια που ακολουθούν. Κατανοώ ότι ο χρόνος σας είναι πολύτιμος αλλά πραγματικά ο κόπος και τα στοιχεία που έχω συγκεντρώσει από τους υφιστάμενούς σας, θα είναι άχρηστα αν δεν συνοδευτούν και από τις αξιολογήσεις τους. Θ α ή θ ε λ α ν α σ α ς δ ι α β ε β α ι ώ σ ω ό τ ι τ α α π ο τ ε λέ σ μ α τ α α υ τ ώ ν τ ω ν αξιολογήσεων είναι απολύτως εμπιστευτικά και οι εργαζόμενοι που αξιολογούνται δεν πρόκειται σε καμμία περίπτωση να πληροφορηθούν τα αποτελέσματα αυτών των αξιολογήσεων. Μόνο εσείς ή/και η διεύθυνση της εταιρείας σας θα λάβετε γνώση αυτών των αποτελεσμάτων, σε λεπτομερή μορφή και ανά εργαζόμενο/η. Τα ονόματα των υφισταμένων σας, αναφορικά με τους οποίους πρέπει να συμπληρώσετε τα ερωτηματολόγια, είναι τα εξής:

Εργαζόμενος Α:

Εργαζόμενος Β: Εργαζόμενος Γ:

Εργαζόμενος Δ: Εργαζόμενος Ε:

Αφού συμπληρώσετε τα ερωτηματολόγια, βάλτε τα μέσα στον φάκελο που σας δίνεται, σφραγίστε τον και επιστρέψτε τον σε μένα. Σας ευχαριστώ πολύ για το χρόνο και τη συνεργασία σας

232

Appendices

Πολύ χαρακτηριστική

Καθόλου χαρακτηριστική

Στο ερωτηματολόγιο αυτό καλείστε να αξιολογήσετε τους εργαζόμενους σε μία σειρά συμπεριφορών που μπορούν να πραγματοποιούν και οι οποίες ναι μεν είναι βοηθητικές για την εταιρεία ή/και το τμήμα και συνήθως χαρακτηρίζουν τον "καλό" εργαζόμενο, αλλά δεν επιβάλλονται από τη θέση του εργαζόμενου ούτε είναι καθορισμένες από τις αρμοδιότητες του ή αμοίβεται παραπάνω όταν τις πραγματοποιεί. Θα χρειαστεί να αξιολογήσετε κάθε εργαζόμενο σε κάθε πρόταση χρησιμοποιώντας μία κλίμακα από το 1-5, όπου το “1” δηλώνει “καθόλου χαρακτηριστική συμπεριφορά” και το “5” δηλώνει “πολύ χαρακτηριστική συμπεριφορά”. Πρέπει να βάλετε το κατάλληλο νούμερο στο αντίστοιχο κουτάκι για κάθε εργαζόμενο, δεξιά από κάθε πρόταση.

AL LO

Ο/Η εργαζόμενος/η … Α Β Γ Δ Ε 1. βοηθάει τους άλλους εργαζόμενους στη δουλειά τους όταν αυτοί έχουν απουσιάσει 2. έρχεται στην ώρα του/της για δουλειά το πρωί και μετά τα μεσημεριανά διαλείμματα 3. αναλαμβάνει εθελοντικά εργασίες που δεν αποτελούν τυπικά μέρος της δουλειάς του/της 4. κάνει αδικαιολόγητα διαλείμματα από την εργασία του/της 5. παίρνει την πρωτοβουλία να καθοδηγήσει τους νέους εργαζόμενους στο τμήμα, ακόμη κι αν αυτό δεν αποτελεί μέρος της δουλειάς του/της 6. εργάζεται περισσότερο χρόνο από το αναμενόμενο, για παράδειγμα, παίρνει λιγότερες μέρες άδεια απ’ ότι οι περισσότεροι, ή λιγότερες απ’ ότι δικαιούται 7. βοηθάει τους άλλους εργαζόμενους όταν αυξάνεται ο φόρτος εργασίας τους (τους βοηθάει μέχρι να ξεπεράσουν τις δυσκολίες τους) 8. επιβραδύνει το ρυθμό εργασίας του/της προς το τέλος της ημέρας 9. ειδοποιεί εγκαίρως όταν δεν μπορεί να έρθει στη δουλειά 10. ξοδεύει πολύ χρόνο σε προσωπικές τηλεφωνικές συνομιλίες 11. δεν παίρνει άδεια αν δεν είναι απαραίτητο 12. με βοηθάει στα καθήκοντά μου 13. κάνει πρωτότυπες υποδείξεις για τη συνολική βελτίωση του τμήματος 14. δεν κάνει επιπλέον διαλείμματα 15. παρακολουθεί πρόθυμα δραστηριότητες που δεν απαιτούνται από την εταιρεία, αλλά συμβάλλουν στη συνολική της εικόνα 16. δεν ξοδεύει πολύ χρόνο σε άσκοπες συζητήσεις

233

Appendices

Σ’ αυτό το ερωτηματολόγιο οι ίδιοι εργαζόμενοι αξιολογούνται αναφορικά με τη γενικότερη αποδοσή τους στην εργασία τους. Θα σας παρακαλούσα να αξιολογήσετε κάθε εργαζόμενο, με την ίδια σειρά, σε κάθε μία από τις ακόλουθες προτάσεις σύμφωνα με την κλίμακα που ακολουθεί:

Διαφωνώ

Δεν είμαι σίγουρος/η

Συμφωνώ

Συμφωνώ απολύτως

1. εκπληρώνει τους στόχους της δουλειάς

AL LO

2. δείχνει επιδεξιότητα σε όλους τους τομείς της δουλειάς

Ο/Η εργαζόμενος/η …

Διαφωνώ απολύτως

3. εκπληρώνει όλες τις απαιτήσεις της εργασίας του/της 4. είναι ικανός σε όλες τις πλευρές της δουλειάς

5. γενικά αποδίδει ικανοποιητικά στην εργασία του/της

6. φέρνει εις πέρας ό,τι απαιτεί η θέση του/της

Τέλος, στο τρίτο και τελευταίο ερωτηματολόγιο που ακολουθεί στην τελευταία σελίδα, οι ίδιοι εργαζόμενοι αξιολογούνται αναφορικά με την κατοχή ή μη μιας σειράς επαγγελματικών δεξιοτήτων. Θα χρειαστεί να αξιολογήσετε κάθε εργαζόμενο σε κάθε πρόταση χρησιμοποιώντας μία 0-7 κλίμακα, όπου κάθε νούμερο αντιστοιχεί σε διαφορετικό επίπεδο κατοχής της δεξιότητας από τον/την εργαζόμενο/η σε σύγκριση με τους συναδέλφους του/της στην ίδια θέση στην επιχείρηση, όπως μπορείτε να δείτε και στον πίνακα που ακολουθεί: Περισσότερο από το 98% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 84% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 67% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 50% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 33% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 16% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

3 1

Περισσότερο από το 2% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Λιγότερο από το 2% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Πιο απλά, όσο υψηλότερη είναι η βαθμολογία του εργαζόμενου σε αυτή την επαγγελματική δεξιότητα, τόσο λιγότεροι συνάδελφοί του/της, κατέχουν αυτή τη δεξιότητα περισσότερο από τον αξιολογούμενο, ενώ όσο χαμηλότερη είναι η βαθμολογία του/της σε αυτή την επαγγελματική δεξιότητα, τόσο περισσότεροι συνάδελφοί του/της κατέχουν αυτή τη δεξιότητα περισσότερο από τον αξιολογούμενο.

234

Appendices

Ο /Η εργαζόμενος/η … A B Γ Δ E

Περισσότερο από το 84% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 67% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

AL LO

Περισσότερο από το 50% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 33% των συναδέλφων του/της στην ίδια θέση στην εταιρεία Περισσότερο από το 16% των συναδέλφων του/της στην ίδια θέση στην εταιρεία Περισσότερο από το 2% των συναδέλφων του/της στην ίδια θέση στην εταιρεία Λιγότερο από το 2% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

Περισσότερο από το 98% των συναδέλφων του/της στην ίδια θέση στην εταιρεία

μιλάει καθαρά αντιμετωπίζει με θετικά αισθήματα τους συναδέλφους του/της φέρεται δυναμικά στην εργασία του/της χρησιμοποιεί σωστά τη γραμματική όταν γράφει έχει καλές σχέσεις με τους περισσότερους συναδέλφους του/της αναζητεί νέα ερεθίσματα στη δουλειά του/της χρησιμοποιεί κατάλληλο τρόπο γραφής σέβεται τους συναδέλφους του/της είναι πρόθυμος να αναλάβει νέες εργασίες χρησιμοποιεί κατάλληλη γλώσσα τόσο όταν μιλάει όσο και όταν γράφει φέρεται με καλωσύνη προς τους συναδέλφους του/της ψάχνει για νέες δραστηριότητες στη δουλειά του/της χρησιμοποιεί σωστή ορθογραφία όταν γράφει ενδιαφέρεται για τους συναδέλφους του/της αναλαμβάνει πρωταγωνιστικό ρόλο σε ομαδικές δραστηριότητες δείχνει φιλική διάθεση μέσα στην επιχείρηση παρακινεί τους συναδέλφους του/της στη δουλειά

Σας ευχαριστώ πολύ για τη συμμετοχή σας. Μόλις ολοκληρωθεί η έρευνα θα σας ενημερώσω για τα αποτελέσματα των αξιολογήσεων.

235