Jpp 2017 16 issue 1 by Hogrefe

Volume 16 / Number 1 / 2017

Journal of

Personnel Psychology Editor-in-Chief Bernd Marcus Managing Editor Petra Gelléri Associate Editors Tanja Bipp Ian Gellatly Barbara Griffin Jonas Lang Laurenz Meier Sandra Ohly Xin-An Zhang

How to be more persuasive and successful in negotiations “Presented in a concise and even entertaining style, this book succeeds in demonstrating how to negotiate successfully and fairly at the same time. A clear recommendation.” Heinz Schuler, PhD, Hohenheim University, Stuttgart, Germany

Marco Behrmann

Negotiation and Persuasion

The Science and Art of Winning Cooperative Partners 2016, viii + 128 pp. US $34.80 / € 24.95 ISBN 978-0-88937-467-6 Also available as eBook Scientific research shows that the most successful negotiators analyze the situation thoroughly, self-monitor wisely, are keenly aware of interpersonal processes during the negotiation – and, crucially, enter negotiations with a fair and cooperative attitude. This book is a clear and compact guide on how to succeed by means of such goal-oriented negotiation and cooperative persuasion. Readers learn models to understand and describe what takes place during negotiations, while numerous figures, charts, and checklists clearly summarize effective

www.hogrefe.com

strategies for analyzing context, processes, competencies, and the impact of our own behavior. Real-life case examples vividly illustrate the specific measures individuals and teams can take to systematically improve their powers of persuasion and bargaining strength. The book also describes a modern approach to raising negotiation competencies as part of personnel development, making it suitable for use in training courses as well as for anyone who wants to be a more persuasive and successful negotiator.

Journal of

Personnel Psychology Volume 16, No. 1, 2017

Editor-in-Chief

Bernd Marcus, Organizational and Personnel Psychology, Institute of Business Administration, University of Rostock, Ulmenstr. 69, 18057 Rostock, Germany. Tel. +49 381 498-4080, Fax +49 381 498-4419, E-mail: bernd.marcus@uni-rostock.de

Managing Editor

Petra Gelle´ri, Work and Organizational Psychology, Faculty of the Humanities and Social Sciences, University of Hagen, Universita¨tsstr. 33, 58084 Hagen, Germany, Tel. +49 2331 987-2745, Fax +49 2331 987-2179, E-mail: jpp.editorial.ofﬁce@gmail.com

Associate Editors

Tanja Bipp, University of Wu¨rzburg, Germany Ian Gellatly, University of Alberta, Canada Barbara Grifﬁn, Macquarie University, Australia Jonas Lang, Ghent University, Belgium Laurenz Meier, University of Neuchaˆtel, Switzerland Sandra Ohly, University of Kassel, Germany Xin-An Zhang, Shanghai Jiao Tong University, China

Editorial Board

Mike Ashton, Canada Arnold Bakker, The Netherlands Gerhard Blickle, Germany Diana Boer, Germany John Campbell, USA Oliver Christ, Germany Neil Christiansen, USA Brian Connelly, Canada Jeremy Dawson, UK Nele de Cuyper, Belgium Filip De Fruyt, Belgium Evangelia Demerouti, The Netherlands Deanne den Hartog, The Netherlands Jo¨rg Felfe, Germany Steffen Giessner, The Netherlands Richard Gofﬁn, Canada Peter Harms, USA Alex Haslam, UK Sarah Hezlett, USA Giles Hirst, Australia Stefan Ho¨ft, Germany Astrid C. Homan, The Netherlands Thomas Jønsson, Denmark Rudolf Kerschreiter, Germany Ulla Kinnunen, Finland Martin Kleinmann, Switzerland Cornelius Ko¨nig, Germany Franciska Krings, Switzerland Jonas Lang, Belgium Kibeom Lee, Canada Klaus Melchers, Germany Bertolt Meyer, Germany John P. Meyer, Canada

Publisher

Hogrefe Publishing, Merkelstr. 3, 37085 Go¨ttingen, Germany, Tel. +49 551 99950-0, Fax +49 551 99950-425, E-mail publishing@hogrefe.com, Web http://www.hogrefe.com North America: Hogrefe Publishing, 7 Bulﬁnch Place, Suite 202, Boston, MA 02114, USA, Phone (866) 823-4726, Fax (617) 354-6875, E-mail publishing@hogrefe.com

Production

Christina Sarembe, Hogrefe Publishing, Merkelstr. 3, 37085 Go¨ttingen, Germany, Tel. +49 551 99950-0, Fax +49 551 99950-425, E-mail publishing@hogrefe.com

Subscriptions

Hogrefe Publishing, Herbert-Quandt-Str. 4, 37081 Go¨ttingen, Germany, Tel. +49 551 99950-900, Fax +49 551 99950-998

Advertising/Inserts

Marketing, Hogrefe Publishing, Merkelstr. 3, 37085 Go¨ttingen, Germany, Tel. +49 551 99950-423, Fax +49 551 99950-425, E-mail marketing@hogrefe.com

ISSN

ISSN-L 1866-5888, ISSN-Print 1866-5888, ISSN-Online 2190-5150

Ó 2017 Hogrefe Publishing. This journal as well as the individual contributions and illustrations contained within it are protected under international copyright law. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microﬁlming, recording or otherwise, without prior written permission from the publisher. All rights, including translation rights, reserved.

Publication

Published in 4 issues per annual volume.

Subscription Prices

Calendar year subscriptions only. Rates for 2017: Institutions: US $324.00/1249.00; Individuals: US $159.00/ 1114.00. Postage and handling: US $16.00/112.00. Single copies: US $81.00/162.50 + postage and handling.

Payment

Payment may be made by check, international money order, or credit card, to Hogrefe Publishing, Merkelstr. 3, 37085 Go¨ttingen, Germany, or, for North American customers, to Hogrefe Publishing, 7 Bulﬁnch Place, Suite 202, Boston, MA 02114, USA.

Abstracting Services

Abstracted/indexed in Social Sciences Citation Index (SSCI), Social SciSearch, Current Contents/Social and Behavioral Sciences (CC/S&BS), PsycINFO, PsyJOURNALS, PSYNDEX, Scopus, IBZ, IBR, ERIH.

Electronic Full-Text

The full-text of the Journal of Personnel Psychology is available online at www.psyjournals.com and also in PsycARTICLES.

Karin S. Moser, UK Klaus Moser, Germany Peter Muck, Germany Laetitia Mulder, The Netherlands Cornelia Niessen, Germany Ioannis Nikolaou, Greece Lisa Penney, USA Deborah Powell, Canada Floor Rink, The Netherlands Ann Marie Ryan, USA Paul R. Sackett, USA Jesus F. Salgado, Spain Niclas Schaper, Germany Bert Schreurs, The Netherlands Sebastian Schuh, China Birgit Schyns, UK Meir Shemla, The Netherlands Christiane Spitzmu¨ller, USA Daan Stam, The Netherlands Thomas Staufenbiel, Germany Sebastian Stegmann, Germany H. Canan Su¨mer, Turkey Klaus J. Templer, Singapore Robert Tett, USA Christian Vandenberghe, Canada Beatrice van der Heijden, The Netherlands Rolf van Dick, Germany Chockalingam Viswesvaran, USA S. Arzu Wasti, Turkey Juergen Wegge, Germany Despoina Xanthopoulou, Greece Ingo Zettler, Denmark

Impact Factor (2015): 0.925

Journal of Personnel Psychology (2017), 16(1)

Ó 2017 Hogrefe Publishing

Contents Original Articles

News and Announcements

Ó 2017 Hogrefe Publishing

Retention of Assessment Center Rater Training: Improving Performance Schema Accuracy Using Frame-of-Reference Training C. Allen Gorman and Joan R. Rentsch

Knowledge Contribution in Organizations via Social Media: The Interplay of Identiﬁcation and Perceived Usefulness Nicole Behringer, Kai Sassenberg, and Annika Scholl

Predicting Readiness for Diversity Training: The Inﬂuence of Perceived Ethnic Discrimination and Dyadic Dissimilarity Yunhyung Chung, Stanley M. Gully, and Kathi J. Lovelace

How Mindset Matters: The Direct and Indirect Effects of Employees’ Mindsets on Job Performance Matt Zingoni and Christy M. Corey

Work–Home Interface and Well-Being: A Cross-Lagged Analysis Audrey Babic, Florence Stinglhamber, Françoise Bertrand, and Isabelle Hansez

Changes Among Associate Editors Awards for Outstanding Achievements as Authors and Reviewers 2016

Journal of Personnel Psychology (2017), 16(1)

Original Article

Retention of Assessment Center Rater Training Improving Performance Schema Accuracy Using Frame-of-Reference Training C. Allen Gorman1 and Joan R. Rentsch2 1

Department of Management and Marketing, East Tennessee State University, Johnson City, TN, USA

School of Communication Studies, University of Tennessee, Knoxville, TN, USA

Abstract: The purpose of this research was to examine frame-of-reference (FOR) training retention in an assessment center (AC) rater training context. In this study, we extended Gorman and Rentsch’s (2009) research showing FOR training effects on performance schemas by examining the effects immediately after training and again after a two-week nonuse period. We examined the retention effects of FOR training on performance ratings and on performance schema accuracy. The results indicated that the FOR training condition, compared to a control condition, yielded performance ratings and performance schemas more similar to expert ratings and to an expert schema, respectively. FOR training also had positive effects on ratings and performance schema accuracy assessed two weeks after training. These results support and extend the theory of FOR training, which posits that the instructed theory of performance replaces the preexisting rater schemas (Lievens, 2001), and they contribute to the research on FOR training within AC contexts. Keywords: frame-of-reference training, training retention, rating accuracy, schema accuracy

Studies have demonstrated the positive effect of frame-ofreference (FOR) training on performance rating accuracy (e.g., Day & Sulsky, 1995; Gorman & Rentsch, 2009; Hoffman et al., 2012; Noonan & Sulsky, 2001; Roch, Woehr, Mishra, & Kieszczynska, 2011; Sulsky & Day, 1992, 1994) and assessment center (AC) rating accuracy (e.g., Lievens, 2001; Schleicher, Day, Mayes, & Riggio, 2002). Additionally, with respect to ACs, inter-rater reliability coefficients have been shown to be higher for FOR-trained raters than for untrained raters, and FOR training is related positively to AC construct-related validity (Lievens, 2001; Schleicher et al., 2002). However, FOR training research in the AC domain has highlighted a limitation of the AC literature, which is the lack of a research base for best practices. This is surprising because AC ratings are used primarily as predictors in personnel selection, and, typically, predictors receive greater legal scrutiny than criterion measures (SIOP Principles, 2003; Equal Employment Opportunity Commission, 1978).

In particular, we found no research investigating the optimal length of time between assessor training and participating as an AC rater. More specifically, researchers have not addressed the retention of assessor rater training. Therefore, one objective of the present study was to test performance schema retention over a two-week interval between FOR rater training and an AC rating task. Meeting this objective has practical and scientific implications for using FOR training in the AC domain. First, although researchers have studied cognitive aspects of FOR training (e.g., Roch & O’Sullivan, 2003; Schleicher & Day, 1998; Sulsky & Kline, 2007), few have examined how raters cognitively structure performance information. Gorman and Rentsch (2009) addressed this research gap by demonstrating that FOR training fosters the development of expert-like schemas, and that schema accuracy predicts rating accuracy over and above declarative knowledge learned during training. One scientific objective of the present study was to conduct a partial replication1

We use the term “partial replication” because we did not include the declarative knowledge measure that Gorman and Rentsch (2009) did in their original study. However, their finding that schema measures predict variance in training outcomes over and above that of declarative knowledge is not unique to their study, and this has been a common finding in studies of the utility of schema measures (e.g., Davis, Curtis, & Tschetter, 2003). Because of this, and the fact that declarative knowledge was not central to our hypotheses, we did not include the declarative knowledge measure in the present study.

Ó 2016 Hogrefe Publishing

Journal of Personnel Psychology (2017), 16(1), 1–11 DOI: 10.1027/1866-5888/a000167

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

of Gorman and Rentsch’s (2009) study, because the reputation of our science depends in part on the replication of findings across samples and contexts (Kepes & McDaniel, 2013). Second, researchers have left unexamined the retention of schemas after FOR training. An empirical test of the retention of schemas after FOR training would contribute to FOR training research in two ways. One, recall and recognition (learning) measures are typically assessed immediately at the conclusion of FOR training, but they reveal nothing about the transfer of training to applied contexts (Baldwin & Ford, 1988). The utility of FOR training is not in whether trainees provide accurate ratings immediately after training, but whether they provide accurate ratings later in another rating context. Two, Gorman and Rentsch (2009) provided evidence that a primary driver of rating accuracy resulting from FOR training is the development of an expert-like schema of performance information. In other words, a result of FOR training is that raters learn to think and rate like expert raters. However, what is unknown is whether FOR-trained raters continue to think and rate like expert raters after the initial training. Evidence of the retention of FOR-trained rater schemas would establish the robustness of FOR training’s effectiveness and utility as a worthwhile intervention for improving performance rating quality in ACs. Therefore, the present study was aimed at adding to the existing research base in support of FOR training for AC rater training and to advance the scientific research on FOR training by examining FOR training effects on the retention of cognitive structures of performance information acquired during FOR training after a two-week period of nonuse. Specifically, we offered the first test of whether performance schemas imparted through FOR training would be retained beyond the initial training session. In doing so, we responded to Roch et al. (2011), who stated that, “research investigating to what extent FOR training transfers out of the training session is needed” (p. 20, emphasis added).

FOR Training Research Foundations and Limitations FOR training is designed to increase rating accuracy and to reduce rater bias by providing (a) clear definitions of each performance dimension and the scales used to rate those dimensions, (b) examples of specific behaviors representing possible performance levels, (c) rating practice, and (d) discussion of discrepancies between practice ratings and expert ratings (Aguinis, Mazurkiewicz, & Heggestad, 2009; Cascio & Aguinis, 2005; Dierdorff, Surface, & Brown, 2010; McIntyre, Smith, & Hassett, 1984; Pulakos, Journal of Personnel Psychology (2017), 16(1), 1–11

1984; Sulsky & Day, 1992). Unfortunately, however, there is little consensus regarding the most appropriate cognitive theoretical explanation of FOR training effectiveness (e.g., Day & Sulsky, 1995; Gorman & Rentsch, 2009; Sulsky & Day, 1992; Woehr, 1994). Furthermore, perhaps a lack of consensus exists because the cognitive effects of FOR training have been tested using primarily recall and/or recognition measures. However, using recall and recognition measures in FOR training research limits understanding. Recall and Recognition Measures Researchers have typically examined the organization of knowledge imparted during FOR training using measures of recall and recognition. For example, Schleicher and Day (1998) derived the contents of raters’ cognitive representations using participants’ free recall statements. Day and Sulsky (1995) utilized the adjusted ratio of clustering (ARC; Roenker, Thompson, & Brown, 1971) to measure the organization of information in FOR trainees’ memories. The ARC measures the extent to which participants recalled performance information belonging to the same category in succession compared with the amount of clustering expected by chance alone. However, recall measures convey only a limited amount of information regarding the organization of knowledge. As Kraiger and Wenzel (1997) noted, measures other than traditional learning indicators (e.g., recall and recognition) are necessary to capture the criterion space of learning acquired in training. There are also several limitations to relying on clustering measures based on recalled information such as the ARC. First, participants will recall differing numbers of behaviors. These differences are due to errors in memory storage and retrieval and indicate that participants respond to different behaviors (i.e., standardization is lacking). Second, ARC scores are based on the number of behaviors and the order in which behaviors are recalled. If, for example, participants recall none or a limited number of performance behaviors, then meaningful ARC scores cannot be computed. Moreover, as Klein and Loftus (1990) pointed out, clustering scores can be improved by as much as a factor of 2 simply by instructing participants to recall behaviors from a given category. Thus, as Day and Sulsky (1995) aptly noted, “the implication is that the degree of clustering at recall may be more a function of retrieval processes than a reflection of memory organization” (p. 165). Third, recall and clustering measures do not fully address theoretical explanations for FOR training’s effectiveness. As Day and Sulsky (1995) correctly noted, FOR training conveys not only the relevant behaviors, but also the proper evaluation of each behavior. Recall and clustering measures do not appropriately capture both of these necessary components. Ó 2016 Hogrefe Publishing

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Recognition measures also have limitations as measures of knowledge organization. For example, it is generally accepted that recognition measures can be manipulated by changing the characteristics of the distractors (Hall, 1983). Thus, seemingly trivial differences in the number and the quality of distractors can lead to nontrivial differences in recognition scores across studies. Furthermore, there is evidence that recognition is primarily recall-based (Guttentag & Carroll, 1997). This evidence challenges the assumed independence of recall and recognition processes. These limitations likely explain the generally disappointing results of studies using recall, recognition, and clustering measures (e.g., Day & Sulsky, 1995; Sulsky & Day, 1992), and the inconsistent support for social-cognitive person perception models as explanations for FOR training effectiveness. Relative to recall and recognition measures, schema measures are more appropriate measures of the knowledge structures acquired during FOR training (Gorman & Rentsch, 2009). Retention of Performance Schemas Following FOR Training Acquisition, retention, and transfer are essential and differentiated training outcomes, each revealing unique information about training effectiveness (Schmidt & Bjork, 1992). Furthermore, acquisition and retention are prerequisites for transfer. Although acquisition is evaluated routinely, retention requires additional investigation, especially in conjunction with acquisition (Arthur, Bennett, Edens, & Bell, 2003; Arthur, Bennett, Stanush, & McNelly, 1998; Schmidt & Bjork, 1992). Arthur et al. (1998) noted that knowledge learned during training can decay after a period of nonuse, and generally, the longer the period of nonuse, the greater the decay. This can be problematic in the case of FOR training because typically trainees lack an opportunity to use or practice the knowledge and skills learned during FOR training until they engage in a rating task outside of the training session (Ford, Quinones, Sego, & Speer Sora, 1992). Such delays between acquisition and transfer are common in AC settings where assessor training, such as FOR training, is typically followed by some period of nonuse (Roch & O’Sullivan, 2003). Clearly, the study of retention following a nonuse period has practical implications and is particularly relevant when delays exist between acquisition and transfer phases (Arthur et al., 1998). Moreover, what little research has been done on the retention of FOR training (only two published studies to our knowledge) is plagued by limitations. Sulsky and Day (1994), for example, examined the effects of FOR training on rating accuracy and found that FOR-trained raters provided significantly more accurate ratings (compared to a control group), even after a 48-hr delay. Although Sulsky and Day (1994) were the first to examine the effects of time Ó 2016 Hogrefe Publishing

on FOR training, their study had several limitations to be addressed by future research. First, 48 hr represents a short time period to examine retention and is atypical in the training retention literature (Arthur et al., 1998). Second, participants who received FOR training did not provide initial ratings immediately after watching the stimulus videotapes; they rated 48 hr after watching the videotapes. Use of initial ratings would minimize competing explanations (e.g., participants could have independently researched and prepared on their own time for the rating task during the 48-hr delay) and therefore strengthen inferences regarding the effectiveness of FOR training over time. Furthermore, Noble’s (1997) unpublished extended replication of Sulsky and Day’s (1994) study found that after a two-week delay, FOR-trained participants’ ratings were similar in accuracy to control-trained participants’ ratings, thus calling into question the generalizability of Sulsky and Day’s (1994) findings. Roch and O’Sullivan (2003) found that rating accuracy persisted in FOR-trained raters after a two-week nonuse period, and FOR-trained raters recalled more behaviors than control-trained raters. However, as mentioned previously, recall measures convey only limited information regarding knowledge organization, and there are serious concerns regarding reliance on these measures in cognitive research on FOR training. Moreover, both published studies used stimulus materials of limited generalizability. Roch and O’Sullivan (2003), for example, used videotapes from an unpublished study (McCauley et al., 1990) that depicted graduate students giving lectures on either sleep disorders or phobias. Although these tapes have been utilized in other FOR training studies (e.g., Woehr, 1994), the generalizability of graduate student lectures on abnormal psychology topics to performance appraisal or AC contexts is questionable. Sulsky and Day (1994) improved upon this somewhat, utilizing videotapes of graduate students role-playing as managers and problem subordinates engaged in fictitious meetings. In the present study, we use, as stimulus materials, videos of actual executives engaged in an exercise as part of an operational developmental AC. We believe that by using context-relevant stimuli in the present research, our study can improve the overall generalizability of FOR training retention research.

The Present Study Therefore, the objective of the present study was to advance FOR training research and its application to ACs by assessing the effects of FOR training on performance schemas over a period of nonuse. The study contributes to literature by examining FOR effects on cognitive structures over an extended nonuse period and by using more rigorous methodology (e.g., context-relevant stimuli, Journal of Personnel Psychology (2017), 16(1), 1–11

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

initial ratings) and data analyses than used in past research. Given the literature reviewed above in support of the efficacy of FOR training for improving rating accuracy, and given the previous findings of Gorman and Rentsch (2009), Roch and O’Sullivan (2003), and Sulsky and Day (1994) on the stability of performance ratings and schemas after FOR training, we tested the following hypotheses: Hypothesis (1a): FOR-trained raters will possess performance schemas immediately following training that are more similar to an expert schema (i.e., more accurate) than control-trained raters. Hypothesis (1b): FOR-trained raters will possess performance schemas that are more similar to an expert schema (i.e., more accurate) after a two-week period of nonuse than control-trained raters. Hypothesis (2a): Performance ratings from FORtrained raters will be more similar to expert ratings (i.e., more accurate) immediately following training than will performance ratings from control-trained raters. Hypothesis (2b): Performance ratings from FORtrained raters will be more similar to expert ratings (i.e., more accurate) after a two-week period of nonuse following training than will performance ratings from control-trained raters. Hypothesis (3a): Rating accuracy will be positively related to performance schema accuracy immediately following training. Hypothesis (3b): Rating accuracy will be positively related to performance schema accuracy after a two-week period of nonuse following training.

Method Participants Ninety-seven undergraduate students enrolled in psychology courses at a regional public university in the southwestern United States volunteered to participate in this study for extra course credit. Seven of the participants did not return for the second session, therefore their data were dropped. All analyses were conducted on the remaining participants (N = 90). The sample was 71% female, 61% Caucasian, 51% part-time workers, and 73% having no experience rating the job performance of another person. The mean age of participants was 19.17 years (SD = 2.21). Journal of Personnel Psychology (2017), 16(1), 1–11

There were no significant differences between the two study conditions on any of the demographic variables.

Procedure Participants completed two sessions with three to eight participants in each session. During Session 1, participants were randomly assigned to either a FOR training condition (n = 44) or to a control training condition (n = 46). Experimenters trained participants based on the condition (FOR training or control training). After training, participants completed a post-training schema measure. Next, while viewing two randomly presented videotaped performance episodes, participants recorded specific behaviors. After viewing each episode, participants recorded their ratings. Then, they completed a demographic questionnaire and received a reminder to return exactly two weeks later. Typically, retention testing involves specific task training and then testing task performance after a nonuse period (Arthur et al., 1998). Therefore, we used a two-week nonuse period for four reasons: a two-week interval (1) is typical in retention studies (Arthur et al., 1998), (2) is consistent with previous FOR training retention studies (e.g., Noble, 1997; Roch & O’Sullivan, 2003), (3) is a standard rule of thumb in psychometrics for determining the stability of a construct in a test-retest reliability framework (Cascio & Aguinis, 2005), and (4) represents a typical time interval between either initial assessor training or refresher/recalibration training and participation in an AC, according to the first author’s experience as an AC assessor. During Session 2, participants completed the posttraining schema measure, and viewed and rated the two randomly presented videotaped performance episodes used in the Session 1. Then, participants were debriefed and thanked.

Stimulus Materials and Comparison Scores We used two randomly selected episodes from Gorman and Rentsch’s (2009) stimulus performance episodes, which are videos featuring actual senior-level executives roleplaying in a developmental AC. The executives played the role of a manager holding a one-on-one meeting with a subordinate. In the present study, participants viewed the videos and rated the executive’s performance. The roleplaying exercises were designed to elicit behaviors reflecting the following performance dimensions: analysis, decisiveness, leadership, confrontation, and interpersonal sensitivity. Each performance episode was approximately 15 min. We assessed rating accuracy using Sulsky and Balzer’s (1988) procedures to develop comparison scores. Ó 2016 Hogrefe Publishing

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Three SMEs (subject matter experts), upper-level graduate students in industrial and organizational psychology who had completed intensive 30-hr training over 6 days and an annual day-long review training as assessors, independently observed and rated the video-recorded performance episodes. Then, they resolved rating differences through consensus to generate a set of comparison scores.

Dependent Variables

Rater Training

Rating Accuracy Using the formulas provided by Sulsky and Balzer (1988), we assessed rating accuracy using Cronbach’s (1955) four indices. Developed using a univariate analysis of variance (ANOVA) framework, elevation (E) is interpreted as an index of overall accuracy, differential elevation (DE) as an index of the accuracy with which a rater distinguishes between ratees across dimensions, stereotype accuracy (SA) as an index of the accuracy with which a rater discriminates among performance dimensions across rates, and differential accuracy (DA) as an index of the accuracy with which a rater identifies individual patterns of strengths and weaknesses (Jelley & Goffin, 2001; Sulsky & Balzer, 1988). Lower scores on these measures represent higher accuracy. We also assessed Borman’s (1977) differential accuracy (BDA), which measures the correlation between ratings on each dimension and the corresponding target scores across ratees. Higher BDA scores indicate better rating accuracy and rating validity (Sulsky & Day, 1994).

A trained graduate student conducted each training session using written standardized procedures. The training for both conditions lasted about 45 min. The FOR training was designed according to Pulakos’s (1984, 1986) protocol and entailed the trainer reading aloud the definition of each performance dimension and the scale anchor. Next, participants matched example behaviors to the dimensions. The trainer discussed how these examples of behaviors represented good performance or poor performance on each dimension. Participants practiced matching behaviors and dimensions again using a list of sample behaviors (similar to those seen in the videotapes) and matching each to a behavior. The trainer then discussed these behaviors and provided feedback as to the dimension and level of performance (weak or effective) represented by each behavior. Participants then observed and rated a video featuring a role play exercise using another AC candidate. To ensure that participants had exposure to examples of both weak and effective performance, the training video showed a candidate displaying a mix of positive and negative behaviors across the five dimensions. Next, the trainer discussed the ratings with the participants and explained the SMEs’ ratings. Trainers told participants in the control training that they would be evaluating assessee performance on the five performance dimensions, presented the rating form, read the definition of each dimension, and showed a broad training video on the evaluation of work performance. No additional training was provided. This type of control training is consistent with previous FOR training research (e.g., Day & Sulsky, 1995; Gorman & Rentsch, 2009; Hoffman et al., 2012).

Performance Schema Accuracy Performance schema accuracy (PSA), the degree to which individuals’ performance schemas are similar to SMEs’ performance schemas, was assessed using Gorman and Rentsch’s (2009) method. For each of the five dimensions, three behaviors were identified as most relevant to the videotaped exercise. Using an 11-point scale from 5 (very dissimilar) to +5 (very similar), participants rated the similarity of the 105 randomly presented pairwise comparisons of the 15 behaviors. We analyzed these data using individual differences Euclidean distance (INDSCAL) multidimensional scaling (MDS) analysis, which is useful for representing knowledge organization (e.g., Forgas, 1981; Rentsch, Heffner, & Duffy, 1994). MDS analysis provides an R2 value that indicates the variance accounted for by the dimensions produced in the MDS solution (Kruskal & Wish, 1978). R2 can be interpreted as a goodness-of-fit measure that ranges from 0 to 1 with higher values reflecting better fit (Rentsch, 1990). To measure PSA, MDS analyses were conducted using the SME similarity data matrix and each participant’s similarity data matrix. The resulting R2 value for each participant served as the operationalization of PSA in subsequent analyses. Again using Gorman and Rentsch’s (2009) methodology, the three SMEs generated similarity data matrices that we analyzed using MDS. The analyses yielded a fivedimensional solution that provided the best fit with a substantial R2 of .99. Consistent with previous research using expert similarity data matrices (e.g., Day, Arthur, & Gettman, 2001), we averaged the similarity ratings of the three SMEs to create the expert data matrix.

Rating Form While they viewed the videos, participants used spaces on a rating form to take notes, record ratee behaviors, and indicate whether each behavior was positive, negative, or neutral. Then, they recorded their dimension ratings on the form using an 11-point (1.0 = extremely weak to 5.0 = exceptional) Likert-type rating scale (Gorman & Rentsch, 2009).

Ó 2016 Hogrefe Publishing

Journal of Personnel Psychology (2017), 16(1), 1–11

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Results See Tables 1 and 2 for intercorrelations between the study variables at Time 1 and Time 2. As expected, FOR-trained raters produced ratings that did not differ significantly in terms of any of the rating accuracy indices from Time 1 to Time 2, and there was no significant difference in PSA from Time 1 to Time 2. Control-trained raters, however, evidenced a significant decrement in SA (Cohen’s d = .13, p < .05) and PSA (Cohen’s d = .28, p < .05) from Time 1 to Time 2.

significantly higher PSA than control-trained raters, t(88) = 2.31, p < .05 (one-tailed); Cohen’s d = .47. Hypothesis 1a was supported. Hypothesis 1b predicted that two weeks after training FOR-trained raters would produce higher PSA than control trained raters. We tested Hypothesis 1b using the same procedure used to test Hypothesis 1a and found that FOR-trained raters (M = .91, SD = .04) had significantly higher PSA two weeks after training than control-trained raters (M = .89, SD = .05), t(88) = 2.09, p < .05 (one-tailed); Cohen’s d = .45. Hypothesis 1b was supported.

Performance Schema Accuracy

Rating Accuracy

Hypothesis 1a predicted that, compared to control-trained raters, FOR-trained raters would produce greater PSA immediately following training. We conducted an independent-samples t-test on the means of the Fisher-z transformed square roots of the R2 values for FOR-trained raters (M = .91, SD = .05) and control-trained raters (M = .89, SD = .05), revealing the FOR-trained raters had

Hypothesis 2a predicted that FOR-trained raters would provide more accurate ratings immediately following training than control-trained raters. Due to conceptual overlap of the five accuracy indices (Schleicher et al., 2002) and their statistically significant intercorrelations (see Table 1), we conducted a multivariate analysis of variance (MANOVA), with training (FOR vs. control) as the

Table 1. Intercorrelations for study variables (Time 1) Variable 1. Training condition

2. Grade point average

.08

–

3. Rating experience

.09

.03

–

4. Elevationb

.31**

.08

.06

–

.09

.27*

.06

.53**

–

6. Stereotype accuracyb

.22*

.16

.06

.57**

.08

–

7. Differential accuracyb

.22*

.07

.12

.57**

.11

.24*

–

8. Borman’s differential accuracy

.33**

.07

.09

.32**

.13

.37**

.13

–

9. Performance schema accuracy

.23*

.14

.23*

.12

.03

.13

.19*

.15

5. Differential elevation

–

Notes. Rating experience = total number of times having rated the job performance of another person. 1 = control, 2 = frame-of-reference. Smaller values on these indices represent greater accuracy. *p < .05. **p < .01.

Table 2. Intercorrelations for study variables (Time 2) Variable

1. Training conditiona

–

2. Grade point average

.08

–

3. Rating experience

.09

.03

–

4. Elevationb

.29**

.06

.19*

–

.06

.03

.14

.48**

–

6. Stereotype accuracyb

.39**

.17

.03

.57**

.02

7. Differential accuracyb

.08

.11

.31**

.48**

.06

.47**

–

8. Borman’s differential accuracy

.15

.02

.32**

.20*

.42**

.16

–

9. Performance schema accuracy

.20*

.10

.12

.08

.12

.19*

.07

.08

5. Differential elevation

–

Journal of Personnel Psychology (2017), 16(1), 1–11

Ó 2016 Hogrefe Publishing

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Table 3. Analysis of variance results for the five accuracy components (Time 1)

Table 4. Analysis of variance results for the five accuracy components (Time 2)

Accuracy

FOR

Control

Cohen’s d

Accuracy

FOR

Control

Elevation

.55 (.20)

.68 (.21)

.65**

Elevation

.55 (.19)

.70 (.28)

Cohen’s d .60**

Differential elevation

.27 (.23)

.31 (.25)

.18

Differential elevation

.28 (.19)

.30 (.24)

.13

Stereotype accuracy

.29 (.10)

.35 (.18)

.45*

Stereotype accuracy

.24 (.11)

.38 (.20)

.85**

Differential accuracy

.21 (.08)

.26 (.14)

.44*

Differential accuracy

.20 (.10)

.22 (.15)

.16

Borman’s differential accuracy

.85 (.49)

.51 (.51)

.68**

Borman’s differential accuracy

.77 (.47)

.61 (.56)

.30

Notes. N = 90. Items in parentheses are standard deviations. For elevation, differential elevation, stereotype accuracy, and differential accuracy, small numbers represent greater accuracy. For Borman’s differential accuracy, larger numbers represent greater accuracy. FOR = frame-of-reference. *p < .05. **p < .01.

independent variable and the five rating accuracy indices as multiple dependent variables to test Hypothesis 2a.2 The results revealed that FOR-trained raters were significantly more accurate than control-trained raters, F(5, 84) = 3.27, p < .01; Wilks’s Λ = .84; η2p = .16. We estimated effect sizes associated with each accuracy variable using ANOVA (see Table 3). Hypothesis 2a was supported. Hypothesis 2b predicted that, compared to controltrained raters, FOR-trained raters would provide more accurate ratings two weeks after training. We tested Hypothesis 2b using MANOVA with training (FOR vs. control) as the independent variable and the five rating accuracy indices as multiple dependent variables. The results revealed that FOR-trained raters were significantly more accurate than control-trained raters, F(5, 84) = 3.66, p < .01; Wilks’s Λ = .82; η2p = .18. Using ANOVA, we estimated the effect size associated with each accuracy index. A summary of these results is provided in Table 4. Hypothesis 2b was supported.

Hypothesis 3b predicted that PSA would be positively related to rating accuracy two weeks after training. Hypothesis 3b was partially supported because PSA correlated significantly with SA (r = .19, p < .05, one-tailed). All other rating accuracy-schema accuracy correlations were in the prediction direction (see Table 2).

Performance Schema Accuracy – Rating Accuracy Relationship Hypothesis 3a predicted that PSA would be positively related to rating accuracy immediately following training. Hypothesis 3a was partially supported, as PSA correlated significantly with DA (r = .19, p < .05, one-tailed). All other rating accuracy-PSA correlations were in the prediction direction (see Table 1).

Discussion In the present study, we examined FOR training effects on the retention of cognitive structures of performance information acquired during FOR training after a two-week period of nonuse. Despite somewhat smaller effect sizes in the present study, our results partially replicated Gorman and Rentsch’s (2009) findings because FOR-trained raters’ PSA was greater than control-trained raters’ PSA. Furthermore, our results indicated that FOR training effects on PSA retention were maintained after two weeks of nonuse. Our results also supported predictions that FOR training would yield greater rating accuracy relative to control training and that higher accuracy would be maintained two weeks after training. These results contribute to research that has shown FOR training’s effects on rating accuracy, and our results corroborate Roch and O’Sullivan’s (2003) finding that the effect is stable after two weeks. Our results also showed that PSA was correlated positively with rating accuracy immediately after FOR training and again two weeks later.

Although previous research has recommended including Cronbach’s four accuracy indices and BDA in a single multivariate analysis due to the conceptual overlap of the five accuracy indexes and their statistically significant intercorrelations (e.g., Schleicher et al., 2002), an anonymous reviewer recommended that separate analyses be conducted for Cronbach’s indices and for BDA. For Cronbach’s indices, the results revealed that FOR-trained raters were significantly more accurate than control-trained raters both immediately following training, F(4, 85) = 2.67, p < .05; Wilks’s Λ = .89; η2p = .11, and again two weeks later, F(4, 85) = 4.54, p < .01; Wilks’s Λ = .82; η2p = .18. For BDA, the results also revealed that FOR-trained raters were significantly more accurate than control-trained raters immediately following training, F(1, 88) = 10.98, p < .01; η2p = .11, but not two weeks later, F(1, 88) = 2.01, ns. For the sake of parsimony and ease of interpretation, we report the results of the full MANOVAs including all five accuracy indices in the manuscript.

Ó 2016 Hogrefe Publishing

Journal of Personnel Psychology (2017), 16(1), 1–11

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Together the results contribute to the FOR training literature by supporting and extending Gorman and Rentsch’s (2009) and others’ work by providing evidence that PSA persists after a two-week nonuse period between training and the rating task. One explanation for FOR training’s positive effects on rating accuracy, even two weeks after training, is that FOR training develops effective and relatively stable schemas that are resistant to cognitive demands encountered during and after training. These results support the notion from rater training research that the imparted theory of performance replaces the preexisting schema of the raters (Lievens, 2001). Specifically, the present study provided additional evidence that FOR training alters trainees’ schemas, which leads to increased accuracy of performance ratings. Furthermore, the evidence that FOR training effects are stable over a twoweek period of nonuse supports FOR training’s practicality and usefulness in AC contexts and other organizational applications.

Implications and Future Research Assessment Centers FOR training has been applied primarily in performance appraisal contexts (Woehr & Huffcutt, 1994), but, more recently, it has been shown to increase rater accuracy in AC settings (e.g., Lievens, 2001; Schleicher et al., 2002). Thus, one objective of the present study was to further highlight the relevance of FOR training research for the AC domain. The results of the present study have important implications for AC applications. Other than the recommendation in the International Taskforce on Assessment Center Guidelines (2015) that no more than six months should elapse between assessor training and the first use of those assessors in an operational AC, there is no research in the AC literature that addresses the optimal length of time between assessor training and participation in the AC. Our results suggest that novice raters should maintain the performance schemas acquired during FOR training and should see no significant decrement in rating accuracy for at least two weeks, and perhaps longer, although trainees will likely evidence decrements in rating accuracy as the period of nonuse increases (Arthur et al., 1998). We see a fertile area for future research, and we encourage AC researchers to examine time between assessor training and participation in operational ACs. Moreover, a two-week delay is a useful time period for AC contexts and may be relevant to some other rating settings such as performance appraisal settings. Sulsky and Day (1994) demonstrated that delays between training and observation compared to delays between observation and ratings have different implications. In AC contexts, Journal of Personnel Psychology (2017), 16(1), 1–11

raters may be trained once and then after a minimal delay of a day or so observe and make ratings. Afterward, they may not rate again for a more extended time period (e.g., weeks or months) without refresher training. Thus, many aspects of the timing between training and the rating task require further study. Performance Schemas Although field research has demonstrated the effectiveness of FOR training on increasing rater accuracy (e.g., Noonan & Sulsky, 2001), no field studies have been conducted to examine the connection between FOR training and PSA. Therefore, further laboratory research in this area should be complemented by field research and should be conducted with larger and more demographically diverse samples. The results of the present study, in addition to the results of previous studies (e.g., Gorman & Rentsch, 2009), suggest that PSA should be considered a meaningful outcome variable in FOR training research in addition to other traditional indices of rating accuracy. Moreover, despite the fact that Bernardin and Buckley (1981) originally proposed FOR training as a method for identifying idiosyncratic raters, this suggestion has largely been ignored (Hauenstein & Foti, 1989). PSA could be utilized, in addition to rating accuracy, as a novel method for identifying rater idiosyncrasy, particularly because paired comparison ratings are effective in eliciting idiosyncratic understanding of stimuli (Mohammed, Klimoski, & Rentsch, 2000; Rentsch, 1990; Rentsch & Klimoski, 2001). Additional research should also be devoted to understanding how, and if, FOR training protocols can be adjusted to streamline the creation of expert-like schemas while still maintaining the integrity of the training principles. Rating Accuracy Our results also shed light on the debate regarding the appropriate rating accuracy index to use in FOR training research. Overall, we found that FOR-trained raters maintained similar levels of rating accuracy on all five rating accuracy indices two weeks after the initial training (although not all of the indices were significantly different than those for control-trained raters at both Time 1 and Time 2). However, we found that control-trained raters evidenced a significant decrement in SA after the two week nonuse period (although not on any of the other accuracy indices), and the largest difference between conditions at Time 2 was for SA (Cohen’s d = .85) but there was no significant difference for SA at Time 1. Moreover, FORtrained raters were significantly more accurate with respect to BDA at Time 1, but there was no significant difference for BDA between the two conditions at Time 2. Ó 2016 Hogrefe Publishing

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Thus, control-trained raters’ ability to identify differences among performance dimensions significantly declined over the two-week interval, but their ability to rank order ratees remained relatively unchanged. Perhaps this suggests that the real contribution of FOR training is that FOR-trained raters retain their ability to detect differences among performance dimensions, whereas non-FOR-trained raters will experience a decrement in this ability if not used or practiced. Thus, distance accuracy indices, such as SA, may be more relevant for measuring the retention of FOR training than a correlational accuracy index such as BDA. Future research should methodically evaluate factors, such as the number of ratees, number of dimensions, inter-rater reliability, variance in performance levels, or type of rating format used, that may differentially influence rating accuracy indices at both the conclusion of training and after a period of nonuse.

Limitations As with any study, one should be cautious in generalizing the present results. The present study involved a sample of novice raters in a laboratory setting. Experienced raters in an organization may have existing and well-established schemas that may be less malleable (Borman, 1987; Lievens, 2001; Schleicher & Day, 1998). Despite the likelihood that the raters in the present study had less experience than raters in organizations and their motivations were likely different, our results were consistent with past findings and indicated that the participants benefited from the FOR training. In addition, an area to be explored in future research is the effect of same versus different rating episodes for initial and later ratings. In the present study, participants rated the same rating episodes after the initial training and again two weeks later. Although using the same episodes may be hindered by memory effects, using different episodes may be hindered by equivalency problems. In the present study, raters did not receive feedback on their initial ratings. We recommend future researchers examine the effects of using same or different rating episodes for initial and ratings after a nonuse period.

Conclusion FOR training research and practice has moved beyond the performance appraisal domain into other areas, such as ACs. Yet, little research has examined the influence of FOR training on AC rater cognition, especially whether the cognitive effects of FOR training persist beyond the training itself. Improving upon the limitations of previous

Ó 2016 Hogrefe Publishing

cognitive research on FOR training, the results of the present study indicated that FOR training fosters the development of expert-like schemas that persist even after a two-week period of nonuse between the training and an AC rating task, and that schema accuracy predicted rating accuracy both after training and again two weeks later. Our findings fill a void in the AC rater training research, and clearly suggest that FOR training has the potential to be a sufficient, if not necessary, component to the implementation of a successful AC. Acknowledgments A version of this paper was presented at the 2011 conference of the Society for Industrial and Organizational Psychology, Chicago, IL. We thank Katy Gaddis, Steven Apodaca, Josh Collins, Lauren Felton, Garolyn Jergins, Ashley McIntyre, Ben Overstreet, Kenneth Smith, Jessica Stoner, and Jennifer Thorndike for their assistance with data collection. We also thank Sheila List for providing valuable comments on an earlier version of this manuscript.

References Aguinis, H. D., Mazurkiewicz, M. D., & Heggestad, E. D. (2009). Using web-based frame-of-reference training to decrease biases in personality-based job analysis: An experimental field study. Personnel Psychology, 62, 405–438. doi: 10.1111/j.17446570.2009.01144.x Arthur, W., Bennett, W., Edens, P. S., & Bell, S. T. (2003). Effectiveness of training in organizations: A meta-analysis of design and evaluation features. Journal of Applied Psychology, 88, 234–245. doi: 10.1037/0021-9010.88.2.234 Arthur, W., Bennett, W., Stanush, P. L., & McNelly, T. L. (1998). Factors that influence skill decay and retention: A quantitative review and analysis. Human Performance, 11, 57–101. doi: 10.1207/s15327043hup1101_3 Baldwin, T. T., & Ford, J. K. (1988). Transfer of training: A review and directions for future research. Personnel Psychology, 41, 63–105. doi: 10.1111/j.1744-6570.1988.tb00632.x Bernardin, H. J., & Buckley, M. R. (1981). Strategies in rater training. Academy of Management Review, 6, 205–212. doi: 10.5465/AMR.1981.4287782 Borman, W. C. (1977). Consistency of rating accuracy and rating errors in the judgment of human performance. Organizational Behavior and Human Performance, 20, 238–252. doi: 10.1016/ 0030-5073(77)90004-6 Borman, W. C. (1987). Personal constructs, performance schemata, and “folk theories” of subordinate effectiveness: Explorations in an Army office sample. Organizational Behavior and Human Decision Processes, 57, 338–357. doi: 10.1016/ 0749-5978(87)90018-5 Cascio, W. F., & Aguinis, H. (2005). Applied psychology in human resource management (6th ed.). Upper Saddle River, NJ: Pearson-Prentice Hall. Cronbach, L. (1955). Processes affecting scores on “understanding of others” and “assumed similarity”. Psychological Bulletin, 52, 177–193. doi: 10.1037/h0044919

Journal of Personnel Psychology (2017), 16(1), 1–11

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

Davis, M. A., Curtis, M. B., & Tschetter, J. D. (2003). Evaluating cognitive training outcomes: Validity and utility of structural knowledge assessment. Journal of Business and Psychology, 18, 191–206. doi: 10.1023/A:1027397031207 Day, D. V., & Sulsky, L. M. (1995). Effects of frame-of-reference training and information configuration on memory organization and rating accuracy. Journal of Applied Psychology, 80, 001–009. doi: 10.1037/0021-9010.80.1.158 Day, E. A., Arthur, W., & Gettman, D. (2001). Knowledge structures and the acquisition of a complex skill. Journal of Applied Psychology, 86, 1022–1033. doi: 10.1037/0021-9010.86.5.1022 Dierdorff, E. C., Surface, E. A., & Brown, K. G. (2010). Frame-ofreference training effectiveness: Effects of goal orientation and self-efficacy on affective, cognitive, skill-based, and transfer outcomes. Journal of Applied Psychology, 95, 1181–1191. doi: 10.1037/a0020856 Equal Employment Opportunity Commission. (1978). Uniform guidelines on employee selection procedures. Federal Register, 43, 38290–38315. Ford, J. K., Quinones, M., Sego, D., & Speer Sora, J. S. (1992). Factors affecting the opportunity to perform trained tasks on the job. Personnel Psychology, 45, 511–527. doi: 10.1111/ j.1744-6570.1992.tb00858.x Forgas, J. P. (1981). Social episodes and group milieu: A study in social cognition. British Journal of Social Psychology, 20, 77–87. doi: 10.1111/j.2044-8309.1981.tb00478.x Gorman, C. A., & Rentsch, J. R. (2009). Evaluating frame-ofreference rater training effectiveness using performance schema accuracy. Journal of Applied Psychology, 94, 1336–1344. doi: 10.1037/a0016476 Guttentag, R. E., & Carroll, D. (1997). Recollection-based recognition: Word frequency effects. Journal of Memory and Language, 37, 502–516. doi: 10.1006/jmla.1997.2532 Hall, J. F. (1983). Recall versus recognition: A methodological note. Journal of Experimental Psychology: Learning, Memory, and Cognition, 92, 346–349. doi: 10.1037/0278-7393.9.2.346 Hauenstein, N. M. A., & Foti, R. J. (1989). From laboratory to practice: Neglected issues in implementing frame-of-reference rater training. Personnel Psychology, 42, 359–379. doi: 10.1111/ j.1744-6570.1989.tb00663.x Hoffman, B. J., Gorman, C. A., Blair, C. A., Meriac, J. P., Overstreet, B. L., & Atchley, E. K. (2012). Evidence for the effectiveness of an alternative multisource performance rating methodology. Personnel Psychology, 65, 531–563. doi: 10.1111/j.17446570.2012.01252.x International Taskforce on Assessment Center Guidelines. (2015). Guidelines and ethical considerations for assessment center operations. Journal of Management, 41, 1244–1273. doi: 10.1177/0149206314567780 Jelley, R. B., & Goffin, R. D. (2001). Can performance-feedback accuracy be improved? Effects of rater priming and ratingscale format on rating accuracy. Journal of Applied Psychology, 86, 134–144. doi: 10.1037/0021-9010.86.1.134 Kepes, S., & McDaniel, M. A. (2013). How trustworthy is the scientific literature in I-O psychology? Industrial and Organizational Psychology: Perspectives on Science and Practice, 6, 252–268. doi: 10.1111/iops.12045 Klein, S. B., & Loftus, J. (1990). Rethinking the role of organization in person memory: An independent trace storage model. Journal of Personality and Social Psychology, 59, 400–410. doi: 10.1037/0022-3514.59.3.400 Kraiger, K., & Wenzel, L. H. (1997). Conceptual development and empirical evaluation of measures of shared mental models as indicators of team effectiveness. In M. T. Brannick, E. Salas, & C. Prince (Eds.), Team Performance Assessment and Measurement (pp. 63–84). Mahwah, NJ: Erlbaum.

Journal of Personnel Psychology (2017), 16(1), 1–11

Kruskal, J. B., & Wish, M. (1978). Multidimensional scaling. Beverly Hills, CA: Sage. Lievens, F. (2001). Assessor training strategies and their effects on accuracy, interrater reliability, and discriminant validity. Journal of Applied Psychology, 86, 255–264. doi: 10.1037/00219010.86.2.255 McCauley, D. P., Jago, I. A., Gore, B., Lance, C. E., Quarles, F. K., Sledge, L., . . . Guest, F. D. (1990). The development of more ecologically valid videotapes for use in performance appraisal research. Paper presented at the annual meeting of the Southeastern Psychological Association, Atlanta, GA. McIntyre, R. M., Smith, D. E., & Hassett, C. E. (1984). Accuracy of performance ratings as affected by rater training and perceived purpose of rating. Journal of Applied Psychology, 69, 147–156. doi: 10.1037/0021-9010.69.1.147 Mohammed, S., Klimoski, R., & Rentsch, J. R. (2000). The measurement of team mental models: We have no shared schema. Organizational Research Methods, 3, 123–165. doi: 10.1177/ 109442810032001 Noble, S. A. (1997). Effect of a time delay on frame-of-reference training. Paper presented at the 12th annual conference for the Society of Industrial and Organizational Psychology, St. Louis, MO. Noonan, L. E., & Sulsky, L. M. (2001). Impact of frame-ofreference and behavioral observation training on alternative training effectiveness criteria in a Canadian military sample. Human Performance, 14, 3–26. doi: 10.1207/ S15327043HUP1401_02 Pulakos, E. D. (1984). A comparison of rater training programs: Error training and accuracy training. Journal of Applied Psychology, 69, 581–588. doi: 10.1037/0021-9010. 69.4.581 Pulakos, E. D. (1986). The development of training programs to increase accuracy with different rating tasks. Organizational Behavior and Human Decision Processes, 38, 78–91. doi: 10.1016/0749-5978(86)90027-0 Rentsch, J. R. (1990). Climate and culture: Interaction and qualitative differences in organizational meanings. Journal of Applied Psychology, 75, 668–681. doi: 10.1037/0021-9010. 75.6.668 Rentsch, J. R., Heffner, T. S., & Duffy, L. T. (1994). What you know is what you get from experience: Team experience related to teamwork schemas. Group and Organization Management, 19, 450–474. doi: 10.1177/1059601194194004 Rentsch, J. R., & Klimoski, R. J. (2001). Why do “great minds” think alike? Antecedents of team member schema agreement. Journal of Organizational Behavior, 22, 107–120. doi: 10.1002/ job.81 Roch, S. G., & O’Sullivan, B. J. (2003). Frame of reference rater training issues: Recall, time and behavior observation training. International Journal of Training and Development, 7, 93–107. doi: 10.1111/1468-2419.00174 Roch, S. G., Woehr, D. J., Mishra, V., & Kieszczynska, U. (2011). Rater training revisited: An updated meta-analytic review of frame-of-reference training. Journal of Occupational and Organizational Psychology, 85, 370–395. doi: 10.1111/j.20448325.2011.02045.x Roenker, D. L., Thompson, C. P., & Brown, S. C. (1971). Comparison of measures for estimation of clustering in free recall. Psychological Bulletin, 76, 45–48. doi: 10.1037/h0031355 Schleicher, D. J., & Day, D. V. (1998). A cognitive evaluation of frame-of-reference training: Content and process issues. Organizational Behavior and Human Decision Processes, 38, 78–91. doi: 10.1006/obhd.1998.2751 Schleicher, D. J., Day, D. V., Mayes, B. T., & Riggio, R. E. (2002). A new frame for frame-of-reference training: Enhancing the

Ó 2016 Hogrefe Publishing

C. A. Gorman & J. R. Rentsch, Retention of Assessment Center Rater Training

construct validity of assessment centers. Journal of Applied Psychology, 87, 735–746. doi: 10.1037/0021-9010.87.4.735 Schmidt, R. A., & Bjork, R. A. (1992). New conceptualizations of practice: Common principles in three paradigms suggest new concepts for training. Psychological Science, 3, 207–217. doi: 10.1111/j.1467-9280.1992.tb00029.x Society for Industrial and Organizational Psychology. (2003). Principles for the validation and use of personnel selection procedures (4th ed.). Bowling Green, OH: Society for Industrial and Organizational Psychology. Sulsky, L. M., & Balzer, W. K. (1988). Meaning and measurement of performance rating accuracy: Some methodological and theoretical concerns. Journal of Applied Psychology, 73, 497–506. doi: 10.1037/0021-9010.73.3.497 Sulsky, L. M., & Day, D. V. (1992). Frame-of-reference training and cognitive categorization: An empirical investigation of rater memory issues. Journal of Applied Psychology, 77, 501–510. doi: 10.1037/0021-9010.77.4.501 Sulsky, L. M., & Day, D. V. (1994). Effect of frame-of-reference training on rater accuracy under alternative time delays. Journal of Applied Psychology, 79, 535–543. doi: 10.1037/ 0021-9010.79.4.535 Sulsky, L. M., & Kline, T. B. (2007). Understanding frame-ofreference training success: A social learning theory perspective. International Journal of Training and Development, 11, 121–131. doi: 10.1111/j.1468-2419.2007.00273.x

Ó 2016 Hogrefe Publishing

Woehr, D. J. (1994). Understanding frame of reference training: The impact of training on the recall of performance information. Journal of Applied Psychology, 79, 525–534. doi: 10.1037/00219010.79.4.525 Woehr, D. J., & Huffcutt, A. I. (1994). Rater training for performance appraisal: A quantitative review. Journal of Occupational and Organizational Psychology, 67, 189–205. doi: 10.1111/ j.2044-8325.1994.tb00562.x

Received August 17, 2015 Revision received February 17, 2016 Accepted March 22, 2016 Published online September 22, 2016

C. Allen Gorman Department of Management and Marketing East Tennessee State University PO Box 70625 128 Sam Wilson Hall Johnson City, TN 37614 USA gormanc@etsu.edu

Journal of Personnel Psychology (2017), 16(1), 1–11

Original Article

Knowledge Contribution in Organizations via Social Media The Interplay of Identiﬁcation and Perceived Usefulness Nicole Behringer,1 Kai Sassenberg,1,2 and Annika Scholl1 1

Leibniz-Institut für Wissensmedien, Tübingen, Germany

Faculty of Science, University of Tübingen, Germany

Abstract: Knowledge exchange via social media is crucial for organizational success. Yet, many employees only read others’ contributions without actively contributing their knowledge. We thus examined predictors of the willingness to contribute knowledge. Applying social identity theory and expectancy theory to knowledge exchange, we investigated the interplay of users’ identification with their organization and perceived usefulness of a social media tool. In two studies, identification facilitated users’ willingness to contribute knowledge – provided that the social media tool seemed useful (vs. not-useful). Interestingly, identification also raised the importance of acquiring knowledge collectively, which could in turn compensate for low usefulness of the tool. Hence, considering both social and media factors is crucial to enhance employees’ willingness to share knowledge via social media. Keywords: social identification, knowledge exchange, participation, social media, usefulness

Exchanging new information and sharing experience-based knowledge among co-workers is vital for organizations to progress and be prepared for the fast-changing society and markets. A growing number of organizations have recognized the potential of social media tools (e.g., blogs, wikis) for capturing, organizing, and transferring such knowledge. These tools make knowledge easily available to other employees. Indeed, organizations’ social media adoption advances at a rapid pace (Pfisterer, Streim, & Hampe, 2013). According to a company survey by McKinsey, 83% of the companies involved reported the use of social media technologies (Bughin & Chui, 2013). Yet, research on contributions via social media indicates that, in many cases, employees are not motivated to engage in knowledge exchange (e.g., Kimmerle, Wodzicki, & Cress, 2008). Most users participate passively by reading their coworkers’ posts, but only a small fraction actively engages by contributing content (Matschke, Moskaliuk, Bokhorst, Schümmer, & Cress, 2014; Preece & Shneiderman, 2009). It is thus essential to understand the determinants of users’ knowledge contribution (i.e., information sharing with fellow users) via social media in the organizational context.

Journal of Personnel Psychology (2017), 16(1), 12–24 DOI: 10.1027/1866-5888/a000169

In this regard, two fields of research are relevant. First, the (perception of the) social medium is central. Approaches to technology acceptance (Davis, Bagozzi, & Warshaw, 1989; Venkatesh & Davis, 2000; Venkatesh, Morris, Davis, & Davis, 2003) highlight perceived usefulness as the crucial prerequisite for the use of a medium. Beyond perceived usefulness, however, we argue that the social context in which a social medium is embedded deserves more attention. Here, approaches to engagement in social context (i.e., social identity theory; Tajfel & Turner, 1986) propose that users are more willing to contribute to others they identify with. The current research takes a novel approach by integrating these ideas. We argue that the perceived usefulness of a social media tool and users’ social identification with the people they share their knowledge with interact with one another – such that a combination of both factors should facilitate users’ willingness to contribute their knowledge. In doing so, we follow recent calls to apply established theories to social media and to investigate interactions with specific features of the social media context (see McFarland & Ployhart, 2015). The present research, thereby, seeks to contribute to an understanding of how organizations can foster knowledge contribution.

Ó 2016 Hogrefe Publishing

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

The Expectancy of Contributing: Perceived Usefulness of Social Media The dominant approaches to technology acceptance (Davis et al., 1989; Venkatesh et al., 2003) consider perceived ease of use and perceived usefulness as the main preconditions for technology adoption. Indeed, perceived usefulness influences technology usage (e.g., Venkatesh & Davis, 2000). Not surprisingly, employees will only use a technology if they believe that this technology does serve the purpose it was designed for (i.e., if it is seen as useful; Behringer & Sassenberg, 2015; Chang & Yang, 2013; McGowan et al., 2012). In short, employees are willing to use a technology (e.g., a tool) if they expect it to help them reach their goals.

The Value of Contributing: Social Identification and Group-Serving Behavior Beyond the usefulness of the technology, the social context should play a key role: knowledge contribution via social media implies sharing information with other users. How do social relations to these other users affect the willingness to contribute? Social identity theory (Tajfel & Turner, 1979) suggests that individuals engage in behavior that serves a group (e.g., other users in their organization) they identify with. Social identification implies that individuals internalize the group’s norms and interests (Sassenberg, Matschke, & Scholl, 2011; Täuber & Sassenberg, 2012), and show engagement on behalf of the group (Brewer & Kramer, 1986; Ouwerkerk & Ellemers, 2002). Similarly, organizational identification (i.e., social identification with an organization) is essential for voluntary engagement in organizations (for a meta-analysis see Riketta, 2005). In short, social identification can elicit the willingness to take an extra step. What does this imply for knowledge contribution? Sharing knowledge is often not an explicit part of employee’s job description. Hence, knowledge contribution in an organization via social media can be regarded as voluntary behavior. One can, therefore, expect social identification to predict knowledge contribution. Indeed, social identification positively predicts participation in online communities (Bagozzi, Dholakia, & Pearo, 2007; Ren, Kraut, & Kiesler,

Ó 2016 Hogrefe Publishing

2007; Sassenberg, 2002). Furthermore, identification with a community of practice increases the number of individual contributions (Kalman, Monge, Fulk, & Heino, 2002; Matschke et al., 2014). To conclude, social identification seems an important motivator to contribute knowledge to one’s group or organization.

Expectancy Value: The Interplay Between Perceived Usefulness and Social Identification Social identification, however, does not always lead to higher effort in favor of the group; sometimes it even lowers effort and performance (Haslam, 2004). Ouwerkerk, de Gilder, and de Vries (2000) found, for instance, that identification enhanced effort on behalf of one’s group, but only when the group’s status was low and, thus, could be improved (not when its status was high). Similarly, Woltin and Sassenberg (2015) found that highly identified members engaged more in group tasks only if they could expect their behavior to promote goal achievement. In contrast, when they had reasons to expect that their effort would not support goal achievement (because a deadline had passed), higher identification no longer increased effort, but promoted goal disengagement. Both findings suggest that social identification only heightens the willingness to contribute to a group, if contributions are perceived to be useful. Therefore, we argue that the perceived usefulness of a medium (i.e., the “expectancy”) might moderate the effect of social identification on knowledge contribution. To be more precise, we expect social identification and the perceived usefulness of a social media tool to interact in their impact on knowledge contribution: when perceiving a social media tool as not useful for knowledge exchange, it would not be in the best interest of the group or organization to still use this tool. Here, identification should rather help employees to reduce their effort (i.e., not to contribute). However, when a tool is seen as useful for knowledge exchange, using this tool seems functional, as it supports attaining the group’s goal. In this case, identification should promote contributions by means of this tool: Hypothesis 1 (H1): The more users identify, the more knowledge they will contribute, if the perceived usefulness of the social media tool is high (but not if it is low).

Journal of Personnel Psychology (2017), 16(1), 12–24

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

The Mechanism: Social Identification and Goal Transformation How, exactly, should identification influence users’ motivation to contribute to group outcomes in case of high perceived usefulness? By definition, social identification is self-perception as a group member, rather than a distinct individual (Tajfel & Turner, 1986). Consequently, individuals become motivated to achieve positive outcomes for their group, rather than for them personally, promoting effort on behalf of the group. Accordingly, social identification gives rise to a goal transformation, by which the selfinterest (on the personal level) is redefined on the collective level. The outcomes for oneself and others in the group become practically interchangeable (De Cremer & Van Vugt, 1999). Building on this argument, in knowledge exchange via social media, social identification should promote the perceived importance of collective knowledge acquisition – that is, it should promote the “value” a user attributes to his/her fellow users’ knowledge acquisition (i.e., retrieval and learning of information by other group members). The more a user (e.g., employee) identifies with a group (e.g., team), the more s/he considers it important that all group members acquire the knowledge they need to contribute to the group’s success; hence, the possibility that other group members may acquire his/her knowledge is considered valuable. Hypothesis 2 (H2): The more users identify, the more important collective knowledge acquisition is to them. Importance of collective knowledge acquisition should, in turn, predict contributions. Here, two competing predictions can be derived. First, the more important a goal (i.e., the higher the “value”), the more willing people usually are to engage in behavior to reach that goal – if they expect this behavior to be useful in reaching the goal (i.e., the “expectancy” is, also, high; Vroom, 1964). Accordingly, in the context of knowledge contribution via social media, the more users subjectively consider it important to acquire knowledge together, the more willing they should be to actually engage in contributing – provided that the tool seems useful (not if it is not useful). This is because, only if the social media tool is perceived as useful, users can expect their contributions to help them reach their goal and may, thus, be willing to adopt the tool (here, by contributing their knowledge). In short, this suggests that a useful (rather than a not-useful) tool might enhance the

Journal of Personnel Psychology (2017), 16(1), 12–24

relation between importance of collective knowledge acquisition and knowledge contribution (“enhancement hypothesis”). Hypothesis 3a (H3a): The more important collective knowledge acquisition is for users, the more knowledge they will contribute – provided that the perceived usefulness of the social media tool is high (vs. low). Yet, there might be a second, alternative prediction for the role of perceived usefulness of the tool. If there are any doubts about the tool’s usefulness, users’ willingness to contribute knowledge via this tool may most critically depend on the ultimate value of the anticipated outcome – here, on the importance of collective knowledge acquisition for the user. In case of low perceived usefulness of the tool, users do not expect that contributing knowledge via this tool is very effective for achieving their group’s goals. Hence, to have a reason to contribute nonetheless, some additional value needs to be anticipated. In other words, only a high “value” of contributing (here, high importance of collective knowledge acquisition) may compensate for low usefulness of the tool (i.e., low “expectancy”). Accordingly, a different interaction pattern than predicted in H3a might occur; users may be more willing to contribute when the tool’s perceived usefulness is low (rather than high), but the importance of collective knowledge acquisition is high. This is in line with the idea that, when the group’s situation is (so-far) unfavorable, group members at times exert more effort on behalf of the group to improve their group’s situation (e.g., as a means to restore their positive social identity; Ouwerkerk et al., 2000) – here, this might imply more contribution especially when the tool’s usefulness seems low, but collective knowledge acquisition is considered as important. Accordingly, low (rather than high) perceived usefulness of a tool might facilitate the relation between importance of collective knowledge acquisition and knowledge contribution (“compensation hypothesis”). Hypothesis 3b (H3b): The more important collective knowledge acquisition is for users, the more knowledge they contribute – provided that the perceived usefulness of the social media tool is low (vs. high). Combining Hypotheses 1, 2, and 3a/3b, we assume a moderated mediation (see Figure 1). Specifically, we expect perceived usefulness to moderate the relation of identification and knowledge contribution (H1), and to do so via users’ importance of collective knowledge

Ó 2016 Hogrefe Publishing

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

Study 1 Method

Figure 1. The moderated mediation model of knowledge contribution (H4a vs. H4b).

acquisition: The more users identify with their (group within an) organization, the more important they consider collective knowledge acquisition (H2). This relation between identification and importance of collective knowledge acquisition will be constant (i.e., not moderated). The more important collective knowledge acquisition is, however, the more users will contribute – either provided that the perceived usefulness of the tool is high (rather than low; “enhancement hypothesis” H3a) or that it is low (rather than high; “compensation hypothesis” H3b). This results in two competing moderated mediation hypotheses: Hypothesis 4a (H4a): The more users identify, the more important collective knowledge acquisition is for them and, thus, the more knowledge they contribute – provided that the perceived usefulness of the social media tool is high (vs. low).

Hypothesis 4b (H4b): The more users identify, the more important collective knowledge acquisition is for them and, thus, the more knowledge they contribute – provided that the perceived usefulness of the social media tool is low (vs. high).

The Present Research Two studies tested these predictions in a university and a business setting, to provide evidence that the hypotheses hold across different contexts. In Study 1, we informed undergraduates that their university institute considers implementing a social learning platform. In this context, we assessed their behavioral intentions to contribute own knowledge on behalf of their fellow students under highly controlled conditions. By assessing intentions to contribute, we sought to gain first evidence that identification with one’s group in the organization (here, at university) and perceived usefulness jointly predict the willingness to contribute. In Study 2, employees of a company reported their knowledge contributions in an existing internal wiki to examine whether the effects also apply to subsequent actual contributions in a real work context.

Ó 2016 Hogrefe Publishing

Fifty-four psychology students filled out an online survey (44 female; Mage = 22.30 years; SD = 3.76) distributed via the university mailing list. Participation was voluntary and compensated with course credits or a €5 voucher. As part of the cover story (257 words), participants read that parts of their future classes might be held online and that online learning will take place via a social learning platform. To make it more plausible, they saw 10 screenshots of this platform, including explanations of its tools (e.g., forum, library, wiki). To be able to generalize across these tools, we coined the term “social learning platform.” Afterwards, social identification as a psychology student of this university was assessed with a five-item scale (e.g., “I identify as being psychology student at my university,” scales ranging from 1 = I don’t agree at all to 7 = I fully agree; α = .85) from Hinkle, Taylor, and Fox-Cardamone (1989, three items) and Kessler and Hollbach (2005, two items) capturing the cognitive and the affective components of identification. For this and all following multi-item measures, we averaged responses across items (for all items and order of measures, see Appendix). Perceived usefulness of the learning platform was measured with three items (e.g., “The learning platform provides a lot of interesting functions,” from 1 = I don’t agree at all to 7 = I fully agree; α = .64). We assessed importance of collective knowledge acquisition with fellow students with four items (α = .98) (e.g., “It is important to me that my fellow students deepen their psychological knowledge”; 1 = does not apply at all to 7 = applies completely). The intentions for knowledge contribution scale consisted of four items focusing on the different functionalities of the social learning platform (e.g., “I would participate in discussions in the forum,” “I would post links with relevant study-related information,” from 1 = very unlikely to 7 = very likely; α = .83). Control variables. Notably, our hypotheses all exclusively focus on the social relation between users (i.e., social identification with their group) and other group-related variables. In contrast, approaches to technology acceptance have mostly examined and demonstrated the influence of individual-level variables for knowledge contribution. To be able to control for a significant amount of the mere individual-level processes, we included an individual-level control variable that taps into effort expectancy (see Venkatesh et al., 2003) and perceived ease of use (see Davis et al., 1989 – namely, users’ technological self-efficacy (Hertel, Niedner, & Herrmann, 2003); in line with these approaches, users’ technological self-efficacy should

Journal of Personnel Psychology (2017), 16(1), 12–24

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

predict more knowledge contribution. It was measured with three items on the confidence in using the platform functionalities (e.g., “I think that I would get on well with the technological functionalities of the social learning platform,” 1 = I don’t agree at all to 7 = I fully agree). Finally, we also assessed the overall evaluation (i.e., liking vs. disliking) of the tool as potential control variable.1

Table 1. Means, standard deviations, and bivariate correlations (Study 1, N = 54) Mean SD

Means, standard deviations, and bivariate correlations between all variables included in the analyses are presented in Table 1. Due to the small and homogenous sample size, we decided to not control for demographic variables and performed all analyses with bootstrapping with the PROCESS macro for SPSS (Hayes, 2013).2 H1 predicted that the higher participants’ identification is, the higher their intentions for knowledge contribution, but only in case of high (compared to low) perceived usefulness of the social learning platform. A simple moderation analysis (Hayes, 2013; Model 1) including intentions for knowledge contribution (dependent variable), technological self-efficacy (control variable), identification (predictor), and perceived usefulness (moderator) tested this. As expected, the higher participants rated their technological self-efficacy, the stronger their intentions to contribute knowledge on the learning platform were (B = .60, SE = .13, p < .001). There was neither a main effect of identification nor of perceived usefulness (see Table 2). Importantly, results yielded the predicted Identification Perceived usefulness interaction (B = .30, SE = .15, p = .045). The conditional effects showed that when perceived usefulness was high (+1 SD), identification positively predicted intentions for knowledge contribution (B = .48, SE = .19, p = .014). In contrast, when perceived usefulness was low ( 1 SD), identification and intentions for knowledge contribution were not related (B = .09, SE = .19, p = .640). Hence, in line with H1, the more users identified, the more knowledge they were willing to contribute, if the perceived usefulness of the social media tool was high (but not if it was low, see Figure 2). We further expected that identification predicts a higher importance of collective knowledge acquisition (H2). Indeed, identification predicted the importance of collective knowledge acquisition (B = .54, SE = .17, p = .003); the more users identified, the more important collective knowledge acquisition was to them. 1

1. Technological self-efficacy

5.82 1.08

Predictor variables 2. Social identification

5.10 1.09 .13

3. Perceived usefulness

5.46 0.95 .15

.10

4. Importance of collective KA 4.80 1.48 .17

Results

Control variable

.42** .14

Outcome variable 5. Knowledge contribution intention

4.64 1.28 .55*** .25

.20 .28*

Notes. KA = knowledge acquisition. *p < .05. **p < .01. ***p < .001.

Table 2. Unstandardized coefficients from bootstrapping analysis for intentions for knowledge contribution, testing H1 (Study 1, N = 54) Intentions for knowledge contribution Variable

Constant

1.11

.79

.167

Technological self-efficacy

0.60

.13

< .001

Social identification

0.19

.13

.153

Perceived usefulness

0.08

.15

.604

Social identification Perceived usefulness

0.30

.15

.045

Notes. B = unstandardized effect size. Bootstrap sample size = 1,000. All variables are mean-centered.

In addition, we hypothesized that the perceived usefulness of the tool moderates the relation between importance of collective knowledge acquisition and intentions for knowledge contribution (H3a vs. H3b); combined with H1, this implied that social identification should predict higher importance of collective knowledge acquisition and, thereby, more intentions for knowledge contribution – depending on the perceived usefulness of the tool (H4a vs. H4b). To test these two hypotheses, we conducted a moderated mediation analysis with bootstrapping (Hayes, 2013; Model 15) with social identification (predictor), importance of knowledge acquisition (mediator), perceived usefulness (moderator), intentions for knowledge contribution (dependent variable), and technological selfefficacy (control variable). All results are presented in Table 3. The Perceived usefulness Importance of collective knowledge acquisition interaction testing H3a/b was

Results of both studies do not change substantially when controlling for overall evaluation of the tool; hence, we do not discuss this control variable in more detail. Please note that this analysis procedure is more reliable than conventional testing for smaller samples, as in the present studies (Shrout & Bolger, 2002). As suggested by a reviewer, for both studies, all variables were mean-centered; bootstrapping results without the mean-centering of all variables are highly similar (and for the interactions identical, as the macro uses mean-centered variables for an interaction term).

Journal of Personnel Psychology (2017), 16(1), 12–24

Ó 2016 Hogrefe Publishing

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

(B = .24, SE = .12, CI = [.0853; .5892]). In contrast, if the moderator perceived usefulness was high (+1 SD), the indirect effect of identification via importance of collective knowledge acquisition on intentions to contribute was reversed (B = .16, SE = .09, CI = [ .4213; .0294]). In short, in line with H4b, users’ higher identification resulted in higher importance of collective knowledge acquisition and, in turn, in stronger intentions to contribute – provided that the tool’s perceived usefulness was low (vs. high; as implied in the “compensation hypothesis”).

Discussion Figure 2. Knowledge contribution intentions as a function of identification and perceived usefulness (Study 1, N = 54). For reasons of readability, the figure depicts results from simple slope analyses via multiple regression with mean-centered variables; accordingly, values for the dependent variable here differ from the values from bootstrapping analyses reported in the table.

Table 3. Unstandardized coefficients from bootstrapping analysis for intentions for knowledge contribution, testing H3a/b and H4a/b (Study 1, N = 54) Intentions for knowledge contribution Variable

Constant

1.33

.71

.066

Technological self-efficacy

0.57

.12

< .001

Social identification

0.23

.13

.077

Perceived usefulness

0.11

.14

.413

Importance of collective KA

0.07

.10

.480

Social identification Perceived usefulness

0.53

.14

< .001

Importance of collective KA Perceived usefulness

0.39

.10

< .001

Notes. KA = knowledge acquisition; B = unstandardized effect size. Bootstrap sample size = 1,000. All variables are mean-centered.

significant (B = .39, SE = .10, p < .001). Supporting H3b (the “compensation hypothesis”), the conditional effects showed that importance of collective knowledge acquisition predicted stronger intentions to contribute when the moderator perceived usefulness of the tool was low (B = .38, SE = .14, p = .011), but not when it was high (B = .02, SE = .13, p = .872; slopes from a simple moderation analysis, Model 1). Furthermore, supporting H4b (rather than H4a), the conditional indirect effects revealed that, if the moderator perceived usefulness of the platform was low ( 1 SD), identification led to stronger intentions to contribute via higher importance of collective knowledge acquisition

Ó 2016 Hogrefe Publishing

In Study 1, we investigated how identification and perceived usefulness jointly predict individuals’ intentions to contribute their knowledge, and whether these relations can be explained by an increased importance of collective knowledge acquisition. As predicted in H1, the more users identified with the group, the more they intended to contribute knowledge, but only in case of high (compared to low) perceived usefulness of the platform. In addition, supporting H2, the more users identified with the group, the more important collective knowledge contribution was for them. Moreover, in line with H3b (not H3a), the more important collective knowledge acquisition was for them, the more knowledge users intended to contribute – provided that the perceived usefulness of the platform was low (vs. high). This indicates that importance of collective knowledge acquisition may have served to compensate for low usefulness (i.e., users may have been especially willing to contribute here, compensating for the platform’s low usefulness). In sum, this supported a moderated mediation as predicted in H4b (not H4a). One might criticize that the current study context was, to some extent, hypothetical – it did not investigate the actual introduction of a social media platform and, as a consequence, focused on anticipated usefulness and intentions to contribute. However, to make this situation as realistic as possible, all participants studied under the same program, we assessed their identification with this program, and the cover story explained that such a platform may be implemented within their program. Therefore, they were led to believe that the platform would eventually become real (please note that we, of course, thoroughly debriefed participants right after the study session). Accordingly, the results should similarly apply to real contexts in which an actual social media tool is implemented. Nonetheless, Study 2 set out to go beyond intentions to contribute knowledge in the future and to replicate the findings for the actual knowledge contributions to an organizational wiki that employees report.

Journal of Personnel Psychology (2017), 16(1), 12–24

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

Study 2 Method This study investigated employees of a department on technical product development of an internationally operating company. This department had been using an internal wiki for knowledge exchange for more than one year. One hundred seven employees of the department were invited via e-mail to participate in this online study. Participation was voluntary and not compensated. Overall, 58 employees filled out the online survey (corresponding to a response rate of 54%). Three participants were excluded because they were the wiki managers and, thus, their contributions to the wiki were not voluntary. For the remaining 55 participants (42 female, 13 male), age was assessed in five groups: 4% were younger than 25 years, 40% were between 25 and 35 years, 31% were between 36 and 45 years, 22% were between 46 and 55 years, and 2 employees were older than 55 years. Twenty percent held a leadership position. Average tenure was 3.73 years (SD = 4.56); average work experience 18.15 years (SD = 10.63). We assessed social identification at the workgroup (rather than organizational) level, because the behavior of interest refers to contributing to a wiki shared only by this workgroup. Indeed, Van Knippenberg and van Schie (2000) showed that workgroup identification is more strongly related to certain team outcomes than organizational identification. In our study, participants indicated identification on five items (e.g., “I identify as being a member of my workgroup,” α = .89) from Hinkle et al. (1989, three items) and Mael and Ashforth (1992, two items) qualified for the working context. Perceived usefulness of the wiki refers to the expectation that contributing knowledge to the wiki is helpful to increase workgroup success and improve one’s reputation. It was measured with six items (e.g., “Exchanging knowledge in the wiki can contribute to workgroup success,” α = .90). Again, for all multi-item measures, responses were averaged. Technological self-efficacy was measured with three items (e.g., “I get on well with the technological functionalities of the wiki,” α = .71; all measures used scales from 1 = I don’t agree at all to 7 = I fully agree). Again, we also assessed overall evaluation (i.e., liking vs. disliking) of the tool as control variable. Importance of collective success was assessed with four items. It assessed the subjective value of the workgroup’s success as potential outcome of knowledge exchange

among members (e.g., “It is important to me that we, as a workgroup, do a good job.”; scale ranging from 1 = does not apply at all to 7 = applies completely; α = .96). The knowledge contribution scale consists of three items covering the number of contributions, optimizations/ updates on contributions by colleagues, and comments as reaction to colleagues’ contributions that participants themselves had added to the wiki (α = .77). Participants filled in concrete numbers. We summed up these three indicators, which were coded as 0 = no contribution versus 1 = contribution took place.

Results Means, standard deviations, and bivariate correlations are presented in Table 4. All analyses followed the pattern from Study 1 with the PROCESS macro (Hayes, 2013).3 We expected that perceived usefulness should moderate the effect of social identification on knowledge contribution (H1). All main effects are reported in Table 5. The Identification Perceived usefulness interaction predicted knowledge contribution (B = .06, SE = .03, p = .040). Table 4. Means, standard deviations, and bivariate correlations (Study 2, N = 55) Mean

Control variable 1. Technological self-efficacy

4.20 1.19

Predictor variables 2. Social identification

5.87 1.24

.12

3. Perceived usefulness

4.64 1.30

.25

.14

4. Importance of collective success Outcome variable

6.69 0.86

.17

.52*** .19

5. Knowledge contribution

0.22 0.34

.26

.07

.02 .11

Table 5. Unstandardized coefficients from bootstrapping analysis for intentions for knowledge contribution, testing H1 (Study 2, N = 55) Intentions for knowledge contribution Variable

Constant

.07

.17

.689

Technological self-efficacy

.07

.04

.095

Social identification

.03

.04

.523

Perceived usefulness

.02

.04

.593

Social identification Perceived usefulness

.06

.03

.040

Notes. B = unstandardized effect size. Bootstrap sample size = 1,000. All variables are mean-centered.

Please note that bootstrapping with PROCESS is applicable for both continuous and categorical dependent measures (Hayes, 2013).

Journal of Personnel Psychology (2017), 16(1), 12–24

Ó 2016 Hogrefe Publishing

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

Figure 3. Knowledge contribution as a function of social identification and perceived usefulness (Study 2, N = 55). For reasons of readability, the figure depicts results from simple slope analyses via multiple regression with mean-centered variables; accordingly, values for the dependent variable here differ from the values from bootstrapping analyses reported in the table.

Table 6. Unstandardized coefficients from bootstrapping analysis for knowledge contribution, testing H3a/b and H4a/b (Study 2, N = 55) Intentions for knowledge contribution Variable

Constant

.00

.17

.998

Technological self-efficacy

.06

.04

.126

Social identification

.02

.04

.719

Perceived usefulness

.00

.04

.986

Importance of collective success

.13

.10

.206

Social identification Perceived usefulness

.07

.03

.013

Importance of collective success Perceived usefulness

.20

.08

.017

Notes. B = unstandardized effect size. Bootstrap sample size = 1,000. All variables are mean-centered.

Supporting H1 and replicating the findings from Study 1, the conditional effects demonstrated that, when the moderator perceived usefulness was high (+1 SD), identification positively predicted intentions for knowledge contribution (though marginally; B = .10, SE = .06, p = .06; one-tailed). In contrast, when the moderator perceived usefulness was low ( 1 SD), this relation between identification and intentions for knowledge contribution did not occur (B = .05, SE = .04, p = .13; one-tailed; see Figure 3). In short, the more users identified, the more knowledge they were willing to contribute – provided that the perceived usefulness of the wiki was high (vs. low). Moreover, we expected identification to predict higher importance of collective success (H2). Replicating findings from Study 1, the more users identified with their

Ó 2016 Hogrefe Publishing

workgroup, the more important collective success was to them (B = .38, SE = .08, p < .001). Furthermore, importance of collective success should predict higher knowledge contribution – depending on the perceived usefulness of the wiki (H3a vs. H3b); finally, identification should predict higher importance of collective success and, thereby, more knowledge contribution – depending on the perceived usefulness of the wiki (H4a vs. H4b). Results of the moderated mediation analysis (again, see procedure from Study 1) are presented in Table 6. As expected, the Perceived usefulness Importance of collective success interaction was significant (B = .20, SE = .08, p = .017). In line with H3b (the “compensation hypothesis”), the conditional effects showed that importance of collective success tended to predict more knowledge contribution when the moderator perceived usefulness of the wiki was low (B = .09, SE = .06, p = .143), but a trend to the opposite relation occurred when the moderator perceived usefulness was high (B = .40, SE = .20, p = .051; slopes from simple moderation analysis, Model 1). Moreover, in line with H4b (rather than H4a) and results from Study 1, the conditional indirect effects revealed that, if the moderator perceived usefulness was low ( 1 SD), identification predicted more knowledge contribution via higher importance of collective success (B = .05, SE = .04, 95% CI = [.0004; .1962]). In contrast, if the moderator perceived usefulness was high (1 SD), this indirect effect of identification on knowledge contribution via importance of collective success did not occur and, if anything, pointed in the opposite direction (B = .15, SE = .14, 95% CI = [ .6209; .0034]). Taken together, results supported H4b that the more users identify, the more important collective success is for them and, thus, the more they contribute – provided that the wiki’s perceived usefulness is low (vs. high; as implied in the “compensation hypothesis”).

Discussion Study 2 almost completely replicated the findings of Study 1 in an organizational context, assessing participants’ reports on their actual contribution behavior. We, again, found that the more users identified with their group, the more they contributed their knowledge – provided that the tool was perceived as highly useful (confirming H1; though the slopes did not reach conventional levels of statistical significance, the pattern of results was identical to Study 1). In addition, we again found that higher identification predicted higher importance of collective knowledge acquisition (here, collective success), regardless of the

Journal of Personnel Psychology (2017), 16(1), 12–24

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

perceived usefulness of the wiki (confirming H2). Studies 1 and 2, together, suggest this effect to occur for identification on the organizational as well as on the workgroup level. Perceived usefulness moderated the second part of the mediation, again supporting the “compensation hypothesis” H3b and H4b (rather than 3a and 4a) – indicating that, when collective success is highly important for them, users might compensate for the low usefulness of the tool by contributing a lot of their knowledge. Note that technological self-efficacy predicted contributions in Study 1, but not (significantly) in Study 2 – potentially due to social desirability, as employees rarely have the courage to rate their self-efficacy as low within organization studies.

General Discussion The current research studied the interplay between users’ social identification and the perceived usefulness of a social media tool – as well as the importance of collective knowledge acquisition as a process explaining these effects – in predicting the willingness to contribute knowledge. Supporting our hypothesis (H1), knowledge contribution was a function of the interplay between identification and perceived usefulness. In both studies, identification facilitated knowledge contribution when the tool was considered useful (not when it was not useful). These results emphasize the importance of both (a) users’ identification with the group they are contributing knowledge to and (b) the perceived usefulness of the available tool to foster users’ knowledge exchange. Furthermore, we found a positive relation between identification and importance of collective knowledge acquisition (H2) in both studies. In more abstract terms, social identification seemed to heighten the “value” (i.e., importance) of contributing knowledge on behalf of the group. In addition, we expected the tool’s usefulness to moderate the effect of identification via importance of collective knowledge acquisition on knowledge contribution, more precisely, the path from importance to contribution. We found support for a “compensation hypothesis” (H3b and H4b) – namely, that high importance of collective knowledge acquisition (i.e., a high value of sharing knowledge) could compensate for low perceived usefulness of the tool (i.e., low expectancy to be successful). In both studies, identification predicted higher importance of collective knowledge acquisition which, in turn, more knowledge contribution among those perceiving low usefulness of the tool; no such indirect effect occurred if usefulness was high. This implies that our results did not support an “enhancement hypothesis” (H3a and H4a), such Journal of Personnel Psychology (2017), 16(1), 12–24

that (high) perceived usefulness of the tool might boost the relation between importance of collective knowledge acquisition and knowledge contribution. In sum, the results clearly outline that identification, importance of collective knowledge acquisition, and perceived usefulness jointly predict contributions. Notably, both studies used single-source measures, raising concerns about common method bias. As the present research, however, in large parts investigated interaction effects, which cannot be explained by common method bias (cf. Siemsen, Roth, & Oliveira, 2010), this strengthens confidence in our findings. Moreover, the sample size was relatively small in both studies. Yet, the results almost perfectly replicated across both studies. Still, a replication of these findings with a larger sample is indicated to rule out final doubts. The current research is that the findings were replicated in two different contexts (Study 1: in a learning context; Study 2 in a business context). On the one hand, these differences between studies may be considered as limitation, because the two settings might not seem directly comparable. On the other hand, however, we counter this critique by highlighting that we examined the very same concepts in both studies. The identical theoretical framework was used, which should, in fact, apply to very different contexts. Indeed, the findings are strikingly consistent. Therefore, the current research provides support for the theoretical framework and the predictions derived from it across different settings – which we consider a clear strength of this pair of studies. Additional support for the validity of our assumptions results from the fact that the effects were found with regard to different tools, like a forum or blog (only in Study 1) and a wiki (in both studies). In other words, the motivational factors discussed here seem to affect (intended and actual) knowledge contribution via social media – independently of the cognitive and behavioral affordances provided by a specific technology. At the same time, further studies should explicitly test for moderators of the findings by varying contextual/technological features more systematically and may seek to replicate the effects in a longitudinal study. Future research might also take an even more finegrained approach, as suggested by the Integrative Model of Behavior Prediction (Fishbein & Yzer, 2003). This suggests that perceived usefulness, intentions (here, intentions to contribute), and behavior (here, actual contribution) should be assessed more specifically. Regarding knowledge contribution contexts, this implies that all three variables should be assessed separately for each tool. This is beyond the present research, which took a first step toward bringing technology acceptance approaches and social identity theory together. Ó 2016 Hogrefe Publishing

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

The findings revealed that high identifiers seemed to compensate the low perceived usefulness of a tool by means of more effort (i.e., more contributions). In case of low perceived usefulness of a tool, the expected effect of one’s contributions on goal achievement is low – users here do not contribute to a tool (only) because they expect their contribution will help them reach a goal; thus, users need to anticipate some additional value to have a reason to contribute nonetheless. More abstractly speaking, a high “value” seems needed to compensate for low “expectancy” (i.e., low perceived usefulness of the tool). In the current research, such a high “value” is given, for instance, by (the importance of) collective knowledge acquisition. The surprisingly rather low knowledge contribution from employees who rated high on both – the importance of collective knowledge acquisition and perceived usefulness – might be due to alternative tools to share their knowledge with others, like the phone or e-mail. It is certainly conceivable that individuals perceive a wiki as very useful, but, at the same time, may have other useful tools at their disposal to communicate important information. However, this attempt to explain a pattern that occurred only in one of our studies is, inevitably, very speculative. Hence, future research should test whether this result replicates and includes other possible or competing tools for knowledge exchange. One other aspect that might have influenced the results is that the importance of collective knowledge acquisition scale might be biased by social desirability. The mean score of 6.7 (on a 1–7 scale; vs. 4.8 in Study 1) in Study 2 was very high. Yet, as socially desirable responses only cause an upward shift of the response distribution, this is not a serious concern when interpreting correlations involving the scale (Podsakoff & Organ, 1986). In terms of theoretical implications, a particular strength of the current research is that it considered social identification and different facets of value (i.e., importance of collective knowledge acquisition or collective success) and expectancy (i.e., perceived usefulness of tools) in the context of knowledge contribution via social media. In doing so, it took the social embeddedness of social media into account. Specifically, it brought together social identity theory (Tajfel & Turner, 1986) and approaches to technology acceptance (Davis et al., 1989), two so-far unrelated approaches as to when individuals adopt a technology and/or engage on behalf of their group. The interplay between two concepts of these approaches explained a meaningful proportion of variance of knowledge contribution. Hence, combining these approaches seems to be fruitful to understand users’ willingness to contribute knowledge and, potentially, also other facets of behavior. In particular, research on social media in knowledge management might profit from considering social identification and other Ó 2016 Hogrefe Publishing

variables capturing the quality of social relationships, rather than merely focusing on expectancy theory (or a more differentiated cost-benefit approach). If our results are replicated across other media or contexts, it might be useful to add social identification to the Technology Acceptance Model or the Unified Theory of Acceptance and Usage of Technologies (Venkatesh et al., 2003). The current research also adds to social identity research. Earlier research considered group norms as a moderator of the impact of social identification on effort in favor of the group (e.g., Louis, Davies, Smith, & Terry, 2007). The current research focused on the expectation that one’s own behavior will contribute to the good of the group, because only useful tools will warrant that the effort in favor of the group will translate into positive outcomes. This moderator of the impact of social identification on behavior might also be relevant for social identity research in other domains. At the same time, communicating the norm within a workgroup or organization on how and when to use a tool may, likewise, guide highly identified users’ contributions. The findings also have implications for management and practitioners. As identification predicts greater importance of collective knowledge acquisition (for similar findings in other domains, see Riketta, 2005), one promising approach seems to invest in measures that foster identification in the work context. On the organizational level, organizational support theory suggests that employees, who feel that the organization cares for their well-being, are likely to reciprocate. To do so, these employees psychologically invest into the organization and develop a stronger sense of attachment to and identification with the organization itself (Eisenberger, Armeli, Rexwinkel, Lynch, & Rhoades, 2001). Possible routes to ensure that employees perceive the organization as supportive and identify with it are to implement fair management procedures (Tyler & Blader, 2003) or provide opportunities for informal exchange, like a “tea kitchen” or joint team events. Yet, going beyond prior approaches to social identification, the present research highlights the significance of perceived usefulness of a knowledge exchange tool and its interplay with social identification. Our results highlighted that social identification only predicts higher engagement in contributing one’s knowledge, if the tool is indeed perceived as useful. Accordingly, one approach to foster knowledge contributions may be to combine measures boosting social identification with highly useful tools for knowledge exchange. Perceived usefulness can be facilitated by adapting tools to users’ needs and enabling users to realize the tool’s potential benefits, for instance, by means of trainings in which participants elaborate the benefits of social media for knowledge exchange, and enact measures to make collective knowledge acquisition more transparent. Journal of Personnel Psychology (2017), 16(1), 12–24

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

Is high perceived usefulness, however, always preferable? Considering results for H1 and H4a together, our findings suggest that (a) users’ willingness to contribute may be highest when (a) users identify and the tool is perceived as useful and/or when (b) users find collective knowledge acquisition important and the tool is perceived as not useful. In short, this raises the question about a potentially optimal level of perceived usefulness of a social media tool in organizations. Our findings do not suggest a clear preference for a tool which is (not) perceived as useful, but rather indicate the importance of considering perceived usefulness and the social context (i.e., social identification and importance of collective knowledge acquisition) to fully understand the impact on users’ willingness to take the extra step and actively share their knowledge with others. To conclude, these suggestions can contribute to a longer lasting continuation of vivid and inspiring knowledge exchange in organizations by means of social media. The current research demonstrates how social identification guides knowledge contribution via social media, namely by affecting the related benefits. Furthermore, this research indicates that if the expectancy of contributing (i.e., perceived usefulness of a tool) is low, high value (i.e., a high importance of collective success or knowledge acquisition) is needed to compensate for this low expectancy and, nevertheless, motivates users to contribute their knowledge via this tool.

References Bagozzi, R. P., Dholakia, U. M., & Pearo, L. R. K. (2007). Antecedents and consequences of online social interactions. Media Psychology, 9, 77–114. doi: 10.1080/15213260709336804 Behringer, N., & Sassenberg, K. (2015). Introducing social media for knowledge management: Determinants of employees’ intentions to adopt new tools. Computers in Human Behavior, 48, 290–296. doi: 10.1016/j.chb.2015.01.069 Brewer, M. B., & Kramer, R. M. (1986). Choice behavior in social dilemmas: Effects of social identity, group size, and decision framing. Journal of Personality and Social Psychology, 50, 543–549. doi: 10.1037/0022-3514.50.3.543 Bughin, J., & Chui, M. (2013). Evolution of the networked enterprise: McKinsey Global Survey results. Retrieved from http:// www.mckinsey.com/business-functions/business-technology/ our-insights/evolution-of-the-networked-enterprise-mckinseyglobal-survey-results Chang, Y.-S., & Yang, C. (2013). Why do we blog? From the perspectives of technology acceptance and media choice factors. Behaviour and Information Technology, 32, 371–386. doi: 10.1080/0144929x.2012.656326 Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35, 982–1003. doi: 10.1287/mnsc.35.8.982 De Cremer, D., & Van Vugt, M. (1999). Social identification effects in social dilemmas: A transformation of motives. European Journal of Social Psychology, 29, 871–893. doi: 10.1002/(SICI) 1099-0992(199911)29:7<871::AID-EJSP962>3.0.CO;2-I

Journal of Personnel Psychology (2017), 16(1), 12–24

Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades, L. (2001). Reciprocation of perceived organizational support. Journal of Applied Psychology, 86, 42–51. doi: 10.1037/0021-9010.86.1.42 Fishbein, M., & Yzer, M. C. (2003). Using theory to design effective health behavior interventions. Communication Theory, 13, 164–183. doi: 10.1093/ct/13.2.164 Haslam, S. A. (2004). Psychology in organizations: The social identity approach (2nd ed.). London and Thousand Oaks, CA: Sage. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: Guilford. Hertel, G., Niedner, S., & Herrmann, S. (2003). Motivation of software developers in open source projects: An Internet-based survey of contributors to the Linux kernel. Research Policy, 32, 1159–1177. doi: 10.1016/s0048-7333(03)00047-7 Hinkle, S., Taylor, L. A., & Fox-Cardamone, D. L. (1989). Intragroup identification and intergroup differentiation: A multicomponent approach. British Journal of Social Psychology, 28, 305–317. doi: 10.1111/j.2044-8309.1989.tb00874.x Kalman, M. E., Monge, P., Fulk, J., & Heino, R. (2002). Motivations to resolve communication dilemmas in database-mediated collaboration. Communication Research, 29, 125–154. doi: 10.1177/0093650202029002002 Kessler, T., & Hollbach, S. (2005). Group-based emotions as determinants of ingroup identification. Journal of Experimental Social Psychology, 41, 677–685. doi: 10.1016/j.jesp. 2005.01.001 Kimmerle, J., Wodzicki, K., & Cress, U. (2008). The social psychology of knowledge management. Team Performance Management, 14, 381–401. doi: 10.1108/13527590810912340 Louis, W., Davies, S., Smith, J., & Terry, D. (2007). Pizza and pop and the student identity: The role of referent group norms in healthy and unhealthy eating. Journal of Social Psychology, 147, 57–74. doi: 10.3200/SOCP.147.1.57-74 Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior, 13, 103–123. doi: 10.1002/job.4030130202 Matschke, C., Moskaliuk, J., Bokhorst, F., Schümmer, T., & Cress, U. (2014). Motivational factors of information exchange in social information spaces. Computers in Human Behavior, 36, 549–558. doi: 10.1016/j.chb.2014.04.044 McFarland, L. A., & Ployhart, R. E. (2015). Social media: A contextual framework to guide research and practice. Journal of Applied Psychology, 100, 1653–1677. doi: 10.1037/a0039244s McGowan, B. S., Wasko, M., Vartabedian, B., Miller, R. S., Freiherr, D. D., & Abdolrasulnia, M. (2012). Understanding the factors that influence the adoption and meaningful use of social media by physicians to share medical information. Journal of Medical Internet Research, 14, 210–220. doi: 10.2196/jmir.2138 Ouwerkerk, J. W., de Gilder, D., & de Vries, N. K. (2000). When the going gets tough, the tough get going: Social identification and individual effort in intergroup competition. Personality and Social Psychology Bulletin, 26, 1550–1559. doi: 10.1177/ 01461672002612009 Ouwerkerk, J. W., & Ellemers, N. (2002). The benefits of being disadvantaged: Performance-related circumstances and consequences of intergroup comparisons. European Journal of Social Psychology, 32, 73–91. doi: 10.1002/ejsp.62 Pfisterer, S., Streim, A., & Hampe, K. (2013). Arbeit 3.0. Arbeiten in der digitalen Welt. Retrieved from http://www.bitkom. org/files/documents/Studie_Arbeit_3.0.pdf Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12, 531–544. doi: 10.1177/014920638601200408

Ó 2016 Hogrefe Publishing

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

Preece, J., & Shneiderman, B. (2009). The reader-to-leader framework: motivating technology-mediated social participation. AIS Transactions on Human-Computer Interaction, 1, 13–32. Ren, Y., Kraut, R., & Kiesler, S. (2007). Applying common Identity and bond theory to design of online communities. Organization Studies, 28, 377–408. doi: 10.1177/0170840607076007 Riketta, M. (2005). Organizational identification: A meta-analysis. Journal of Vocational Behavior, 66, 358–384. doi: 10.1016/ j.jvb.2004.05.005 Sassenberg, K. (2002). Common bond and common identity groups on the Internet: Attachment and normative behavior in on-topic and off-topic chats. Group Dynamics: Theory, Research, and Practice, 6, 27–37. doi: 10.1037/1089-2699. 6.1.27 Sassenberg, K., Matschke, C., & Scholl, A. (2011). The impact of discrepancies from ingroup norms on group members’ wellbeing and motivation. European Journal of Social Psychology, 41, 886–897. doi: 10.1002/ejsp.833 Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and non-experimental studies: New processes and recommendations. Psychological Methods, 7, 422–445. Siemsen, E., Roth, A., & Oliveira, P. (2010). Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods, 13, 456–476. doi: 10.1177/1094428109351241 Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA: Brooks/Cole. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. Austin (Eds.), Psychology of intergroup relations (pp. 7–24). Chicago, IL: Nelson-Hall. Täuber, S., & Sassenberg, K. (2012). Newcomer Conformity – How self-construal affects the alignment of cognition and behavior

Ó 2016 Hogrefe Publishing

with group goals in novel groups. Social Psychology, 43, 138–147. doi: 10.1027/1864-9335/a000092 Tyler, T. R., & Blader, S. L. (2003). The group engagement model: Procedural justice, social identity theory, and cooperative behavior. Personality and Social Psychology Review, 7, 349–361. doi: 10.1177/1368430201004003003 Van Knippenberg, D., & van Schie, E. M. (2000). Foci and correlates of organizational identification. Journal of Occupational and Organizational Psychology, 73, 137–147. doi: 10.1348/096317900166949 Venkatesh, V., & Davis, F. D. (2000). A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science, 46, 186–204. Venkatesh, V., Morris, M. G., Davis, F. D., & Davis, G. B. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27, 425–478. Vroom, V. H. (1964). Work and motivation. New York, NY: Wiley. Woltin, K.-A., & Sassenberg, K. (2015). Showing engagement or not: The influence of social identification and group deadlines on individual control strategies. Group Processes and Intergroup Relations, 18, 24–44. doi: 10.1177/ 1368430214542254 Received October 30, 2015 Revision received April 20, 2016 Accepted April 22, 2016 Published online September 22, 2016 Annika Scholl Leibniz-Institut für Wissensmedien Schleichstr. 6 72076 Tübingen Germany a.scholl@iwm-tuebingen.de

Journal of Personnel Psychology (2017), 16(1), 12–24

N. Behringer et al., Knowledge Contribution in Organizations via Social Media

Appendix Complete Lists of Items Used in the Studies Study 1 Social identification (1 = I don’t agree at all; 7 = I fully agree) I identify as being psychology student at my university. I feel strong ties to the psychology students at my university. I am happy to belong to the psychology students at my university. I appreciate being psychology student at my university. I perceive myself as psychology student at my university. Usefulness of the learning platform (1 = I don’t agree at all; 7 = I fully agree) The learning platform provides a lot of interesting functions. The structure of the learning platform with its different functionalities is clear and comprehensible. I like the layout of the learning platform. Importance of collective knowledge acquisition (1 = does not apply at all; 7 = applies completely) It is important to me, that my fellow students . . . . . . acquire key competencies. . . . expand their psychological knowledge. . . . acquire skills that are relevant for their studies. . . . deepen their psychological knowledge. Technological self-efficacy (1 = I don’t agree at all; 7 = I fully agree) I think that I possess the required skills, to become engaged in the learning platform. I think that I could handle the technological functionalities of the learning platform effectively. I think that I would get on well with the technological functionalities of the learning platform. Knowledge contribution intention (1 = very unlikely; 7 = very likely) I would post links with relevant study-related information. I would answer questions in the forum. I would participate in discussions in the forum. I would write comments on my fellow students’ contributions.

Journal of Personnel Psychology (2017), 16(1), 12–24

Study 2 Social identification (1 = I don’t agree at all; 7 = I fully agree) I am happy to belong to my workgroup. I feel strong ties to my workgroup. I identify as being a member of my workgroup. When someone criticizes my workgroup it feels like a personal insult. When I talk about my workgroup, I usually say ‘we’ rather than ‘they’. Usefulness of the wiki (1 = I don’t agree at all; 7 = I fully agree) Exchanging knowledge in the wiki can contribute to workgroup success We as a workgroup can be more efficient, if we share our knowledge in the wiki. The wiki is useful for the work of our workgroup. With a high quality contribution in the wiki you can prove your own professional expertise. The wiki offers the opportunity to prove your own knowledge and gain high reputation. By providing high quality contributions to the wiki you can gain reputation among your coworkers. Importance of collective success (1 = does not apply at all; 7 = applies completely) It is important to me, that we as a workgroup . . . . . . work efficiently. . . . are successful. . . . can do our job well. . . . expand work-related skill. Technological self-efficacy (1 = I don’t agree at all; 7 = I fully agree) Sometimes I am not able to edit content because of technological challenges. (recoded). I find it difficult to edit content in the wiki. (recoded). I get on well with the technological functionalities of the wiki. Knowledge contribution (Please fill in the number.) How often did you submit an own contribution in the wiki? How often did you edit a contribution of a coworker in the wiki? How often did you comment on a contribution of a coworker in the wiki?

Ó 2016 Hogrefe Publishing

Original Article

Predicting Readiness for Diversity Training The Inﬂuence of Perceived Ethnic Discrimination and Dyadic Dissimilarity Yunhyung Chung,1 Stanley M. Gully,2 and Kathi J. Lovelace3 1

College of Business and Economics, University of Idaho, Moscow, ID, USA

School of Labor and Employment Relations, College of the Liberal Arts, Pennsylvania State University, University Park, PA, USA

School of Business Administration, Menlo College, Atherton, CA, USA

Abstract: Using data collected from 160 employed professionals in the US, we performed multivariate and univariate multiple regression analyses to examine the joint effect of perceived ethnic discrimination and ethnic dyadic dissimilarity on trainee readiness for diversity training (pre-training motivation to learn, self-efficacy, intention to use, and perceived utility). A significant interaction effect showed that individuals displayed stronger pre-training motivation to learn, intention to use, and perceived utility when they perceived discrimination based on ethnic background and when they were ethnically dissimilar to their supervisor. However, perceived ethnic discrimination was not associated with these three readiness variables when subordinate-supervisor ethnic backgrounds were the same. Implications for theory and practice are discussed. Keywords: diversity training, pre-training readiness, ethnic discrimination, dyadic dissimilarity

Although many organizations emphasize the importance of fair and inclusive work climates, racial and ethnic discrimination charges reported to the Equal Employment Opportunity Commission (EEOC) have continued to rise over the last decade (e.g., from 80,680 charges in 1997 to 88,778 charges in 2014, approximately 35% of racial discrimination charges; EEOC, 2015). Because racial discrimination deteriorates various individual, group, and organization outcomes (see Goldman, Gutek, Stein, & Lewis, 2006 for a review), diversity initiatives in the workplace, especially diversity training, have increased in popularity: roughly 68% of US companies provide diversity training (Society for Human Resource Management, 2010). Diversity training can be an effective means of improving employees’ reactions toward racial diversity and increasing specific diversity management skills and knowledge (see Buzrukova, Jehn, & Spell, 2012; Kalinoski et al., 2013 for reviews). Nevertheless, cynicism toward diversity programs is still widespread (Kulik, Pepper, Roberson, & Parker, 2007). This cynicism may be rooted in employees’ low pretraining interests in diversity training (Kulik et al., 2007).

Ó 2016 Hogrefe Publishing

Employees come to training programs with different backgrounds and knowledge regarding what they have experienced at work and what they will confront when they return to their jobs (Mathieu & Martineau, 1997). Because these individual backgrounds and contextual knowledge may influence how they respond to diversity training initiatives (Mor Barak, Cherin, & Berkman, 1998; Roberson, Kulik, & Tan, 2013; Wiethoff, 2004), employees enter training with varying levels of motivation and receptivity toward new concepts and ideas and this pre-training readiness may greatly affect the success of training processes and outcomes (Mathieu & Martineau, 1997). As such, understanding how individual and contextual factors influence pre-training readiness toward diversity training is important for the effectiveness of diversity training (Chung, 2013; Wiethoff, 2004). In this study, we define pre-training readiness for diversity training as employees’ pre-training attitudes and beliefs that enable them to be receptive and motivated to learn the content of diversity training and to actively engage in and be committed to the training while taking a diversity training program. Pre-training readiness for diversity

Journal of Personnel Psychology (2017), 16(1), 25–35 DOI: 10.1027/1866-5888/a000170

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

training is a higher-order construct of the multiple variables that may influence trainee receptiveness to training. Specifically, we examine four dimensions of attitudes/ beliefs related to the employee’s pre-training readiness for diversity training: pre-training motivation to learn, selfefficacy, intention to use, and perceived utility of knowledge and skills learned from diversity training. There is strong theoretical and empirical evidence on the influence of these four readiness dimensions on posttraining reactions, learning, and training transfer in the general training literature (e.g., Alliger, Tannenbaum, Bennett, Traver, & Shotland, 1997; Colquitt, LePine, & Noe, 2000; Mathieu & Martineau, 1997; Sitzmann, Brown, Casper, Ely, & Zimmerman, 2008 for reviews). However, less research has focused on the pre-training factors that influence an employee’s readiness for diversity training (Roberson et al., 2013). Previous research has focused on attitudes toward diversity itself (see Corobot-Mason, Konrad, & Linnehan, 2006 for a review) and voluntary participation in diversity training as proxy variables of trainees’ receptiveness toward the content of diversity training (Kulik et al., 2007). Contributing to this stream of literature, we investigate how individual and contextual factors affect the employee’s pre-training motivation to learn, self-efficacy, intention to use, and perceived utility of diversity training. We contend that this information can aid in the development and delivery of effective diversity training programs, especially the needs assessment of diversity training, which can improve diversity training outcomes and strengthen the connection between workplace diversity and improved company products and services. We argue that perceived ethnic discrimination and ethnic dyadic dissimilarity are pivotal individual and contextual predictors of readiness for diversity training. Perceived ethnic discrimination and ethnic dissimilarity between a subordinate and a supervisor have been identified as major sources of negative work outcomes (see King, Dawson, Kravitz, & Gulick, 2012; Goldman et al., 2006; Avery, Volpone, Mckay, King, & Wilson, 2012). Perceived ethnic discrimination is an individual’s reaction from social categorization and intergroup comparison processes based on ethnicity. Also, because supervisors are a major force in defining the realities of subordinates’ jobs (Wesolowski & Mossholder, 1997), tensions and conflicts caused by ethnic dissimilarity may become crucial issues to employees and may make employees eager to find solutions. Hence, perceived ethnic discrimination and ethnic dyadic dissimilarity may stimulate a desire for learning about diversity issues, thus improving readiness for diversity training. The hypothesized model for this study is illustrated in Figure 1.

Journal of Personnel Psychology (2017), 16(1), 25–35

Ethnic Dyadic Dissimilarity H3 Perceived Ethnic Discrimination

H2 Readiness for Diversity Training - Pre-training motivation to learn - Pre-training self-efficacy - Pre-training intention to use - Pre-training utility

Figure 1. Hypothesized model.

Background and Hypotheses Pre-Training Readiness for Diversity Training The effectiveness of diversity training can be largely influenced by the trainee’s readiness for the training. Within the general training literature, theoretical and empirical evidence with regard to the connections between pre-training readiness variables and training outcomes is well documented (e.g., Colquitt et al., 2000; Gully & Chen, 2010; Sitzmann et al., 2008). Likewise, existing research on diversity training has highlighted the importance of assessing pre-training readiness for diversity training to enhance diversity training effectiveness (see Roberson et al., 2013 for a review). Drawing on both groups of research, this study identifies and explains the following four key dimensions that form readiness for diversity training as follows. Pre-training motivation to learn refers to the trainees’ desire to learn the content of training programs before participating in a training program. Trainees who are not motivated prior to training to learn training content will learn less and have less desire to apply the training in the workplace. In line with this argument, previous research has found that pre-training motivation to learn improves trainee learning and utility reactions (Tracey, Hinkin, Tannenbaum, & Mathieu, 2001), learning outcomes (Mathieu, Tannenbaum, & Salas, 1992), and training transfer (Patrick, Smy, Tombs, & Shelton, 2012). Hence, it is also reasonable to conclude that pre-training motivation is an important dimension of trainee readiness for diversity training because a higher level of pre-training motivation to learn would result in greater learning about diversity during the training and greater transfer of training at the posttraining phase. Pre-training self-efficacy refers to the degree to which people believe they can learn and apply training content before attending a training program (Mathieu et al., 1992). Drawing on Bandura’s (1986) social cognitive theory, previous research found that pre-training self-efficacy has a strong positive effect on learning, training transfer, and job performance (Colquitt et al., 2000; Patrick et al., 2012;

Ó 2016 Hogrefe Publishing

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

Tracey et al., 2001). In diversity training research, little attention has been paid to pre-training self-efficacy about learning diversity. However, drawing on the findings of general training research, a trainee’s belief in his or her capabilities to learn diversity-related knowledge and skills should encourage the trainee to more actively engage in diversity training programs. Pre-training intention to use training refers to the trainee’s plan to use the knowledge and skills learned in the training program. The theory of planned behavior (Ajzen, 1991) posits that an individual’s intention to perform a given behavior can strongly predict actual behavior. By extension, the trainee’s intention to use diversity training content can predict whether the trainee learns to perform the specific behaviors on the job or acquires knowledge and skills related to training transfer. Diversity training has been criticized because transfer of learning to the job is not easy to demonstrate (Roberson, Kulik, & Pepper, 2009). However, if an individual has strong intentions to use diversity training content during the pre-training phase, then the resulting increased intention is likely to enhance learning during the training and lead to changes in behavior during the post-training phase (Combs & Luthans, 2007; Wiethoff, 2004). Therefore, pre-training intention to use diversity-related skills and knowledge constitutes trainee readiness for diversity training. Pre-training perceived utility of training refers to a trainee’s opinion or perception regarding whether the training program content will be useful on the job (Alliger et al., 1997). The general training literature has found that the trainee’s belief that training is useful affects willingness to learn and transfer training content to the job (see Alliger et al., 1997 for a review). In the pre-training phase, people judge their situation to determine whether diversity training has practical value. If they find diversity training useful, they are more likely to be motivated to learn diversity training content. As a result, pre-training perceived utility may help trainees improve readiness for learning diversity subjects. In summary, among various factors that can determine pre-training readiness for diversity training, we focus on four dimensions: pre-training (a) motivation to learn, (b) self-efficacy, (c) intention to use, and (d) perceived utility. These four dimensions are positively associated with each other. Trainees who believe their capabilities to learn are also motivated to learn training content (Colquitt et al., 2000) and are more likely to engage in positive diversity initiatives (Combs & Luthans, 2007). In addition, trainees who perceive a high level of utility also react to the training positively (Alliger et al., 1997). We propose that these four dimensions may be influenced by (i) a trainee’s experiences regarding ethnic discrimination and (ii) ethnic dyadic dissimilarity between the trainee and the supervisor, which is addressed as follows. Ó 2016 Hogrefe Publishing

Perceived Ethnic Discrimination and Readiness for Diversity Training Social identity theory (SIT) provides the theoretical rationale for the consequences of perceived ethnic discrimination. SIT posits that people categorize themselves into in-groups and others into out-groups. Through intergroup comparisons, individuals regard their group as superior to other groups, which may lead to perceived discrimination of out-group members (Ashforth & Mael, 1989; Turner & Haslam, 2001). Through the lens of SIT, extant research has expounded on the negative effects of perceived ethnic discrimination on various work outcomes (King et al., 2012; Goldman et al., 2006 for reviews). However, building on expectancy theory, this negative relationship may positively influence pre-training readiness for diversity training. Expectancy theory suggests that expectancy (i.e., a belief about the link between trying to learn and handling diversity issues more effectively), instrumentality (i.e., a belief of the link between handling diversity issues and better job performance or relationships), and valence (i.e., a belief of the value of higher job performance or better relationships) will help trainees actively engage in training programs and be motivated to learn the content of the programs (Noe, 2010). In the context of diversity training, when trainees believe that they have the ability to learn diversity training content (expectancy), learning the content is linked to outcomes such as working well with ethnically different others (instrumentality), and they value working well with supervisors (valence), they are more likely to be motivated to learn the content of diversity training programs. Diversity training is designed to remove work-related obstacles to enable people to interact well with others from different backgrounds. Thus, trainees who have experienced ethnic discrimination are likely to positively react to diversity training and are willing to use knowledge and skills learned from diversity training (Kulik et al., 2007; Mor Barak et al., 1998). Accordingly, they are more likely to believe that learning the content of diversity training is linked to outcomes such as working well with people who have different ethnicity (i.e., instrumentality) and the outcomes are practically valuable to them (i.e., valence). Because trainees are aware of a personal need for training (Noe, 2010), they are more motivated to learn the diversity training material. Therefore, perceived ethnic discrimination may be positively related to a trainee’s pre-training motivation to learn diversity training content. In line with this logic, perceived ethnic discrimination may positively affect pre-training self-efficacy. If trainees have experienced ethnic discrimination, they may have delved into discrimination issues and may better understand when they encounter illustrations of discrimination

Journal of Personnel Psychology (2017), 16(1), 25–35

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

during the diversity training. This suggests that the trainees will be less anxious about learning, thus resulting in increased confidence for learning relevant and related concepts (i.e., higher pre-training self-efficacy) (Holladay, Knight, Paige, & Quiñones, 2003), which increases expectancy. In addition, perceived ethnic discrimination may positively influence pre-training intention to use diversity training content. The theory of planned behavior (Ajzen, 1991) suggests that behavioral intentions enhance pre-training readiness for diversity training and therefore increases active participation in diversity training programs (Wiethoff, 2004). A trainee who perceives ethnic discrimination is more likely to have experiences related to diversity issues and to believe that knowledge and skills from diversity training may be used when they encounter similar problems. Thus, the trainee will be more willing to use the diversity training content on the job and is more likely to express confidence in using knowledge and skills learned from diversity training at work (Holladay et al., 2003). Likewise, perceived ethnic discrimination may be positively associated with pre-training perceived utility of diversity training. Adults tend to have higher readiness for training when they recognize the practical value of learning (Blanchard & Thacker, 1998). Since diversity training is designed to promote fair treatment toward employees regardless of background, trainees who observe and experience discriminatory events may perceive a greater value of diversity training (Kulik et al., 2007). In addition, if trainees favorably evaluate the practical value of diversity training efforts and think that diversity training will help them perform more effectively, they will have a more positive evaluation of the utility of the training program (i.e., valence). Based on the preceding discussion, we hypothesize that Hypothesis 1: Perceived ethnic discrimination is positively related to four trainee readiness dimensions toward diversity training: pre-training (a) motivation to learn, (b) self-efficacy, (c) intention to use, and (d) perceived utility.

Ethnic Dyadic Dissimilarity and Readiness for Diversity Training Dyadic dissimilarity/similarity refers to the comparative demographic characteristics between a subordinate and a supervisor at the individual level of analysis (Tsui & O’Reilly, 1989). Ethnic dissimilarity focuses on the subordinate-supervisor difference based on ethnicity. According to the similarity-attraction paradigm (Byrne, 1971) and self-categorization theory (Turner & Haslam, Journal of Personnel Psychology (2017), 16(1), 25–35

2001), individuals psychologically categorize themselves based on distinguishable social characteristics (e.g., ethnicity) and are more likely to become attached to and interact with in-group members. Hence, ethnic dyadic dissimilarity between a supervisor and a subordinate can reduce the quality and quantity of interactions and exacerbate feelings of discrimination and isolation, which consequently can lead to perceptions of an unfavorable work environment (Avery, McKay, & Wilson, 2008) and negative work outcomes (Avery et al., 2012). However, building on expectancy theory, we suggest that ethnic dissimilarity is positively associated with readiness for diversity training. If an individual perceives lower supervisor support or has experienced a relationship challenge due to the greater ethnic dissimilarity, he or she may deem diversity training necessary to reduce perceived or experienced social conflict and to better respond to an unpleasant situation due to ethnic discrimination. Hence, ethnic dyadic dissimilarity may create a perceived need for diversity training (i.e., valence in the expectancy theory) and enable trainees to believe that learning the content of diversity training may help them to work well with coworkers and supervisors who are demographically dissimilar with themselves (i.e., instrumentality). As a result, it may be positively associated with pre-training motivation to learn diversity training content. Similarly, pre-training self-efficacy to learn diversity-related content may be greater for trainees who are ethnically dissimilar from their supervisor because they may have had to handle relationship challenges and, thus, may be more familiar with the key issues of diversity and have greater confidence in their ability to learn related materials (i.e., higher expectancy). Moreover, the trainee’s pre-training intention to use would also be greater for those individuals familiar with issues associated with ethnic dyadic dissimilarity because diversity training content may be viewed as helpful for developing the skills needed to solve the encountered issues (i.e., higher valence). Likewise, trainees who experience ethnic dyadic dissimilarity are more likely to recognize the usefulness or utility of diversity training. If they have experienced negative outcomes from dyadic dissimilarity, their need to improve the situation and their desire to learn skills and knowledge for coping will be stronger. As such, supervisor-subordinate ethnic dyadic dissimilarity is likely to increase individual needs for and interests in diversity training and to develop a stronger desire to attend diversity training (Kulik et al., 2007; Roberson, Kulik, & Pepper, 2001). In this respect, we propose the following hypothesis: Hypothesis 2: Ethnic dyadic dissimilarity between a supervisor and a subordinate will positively relate to Ó 2016 Hogrefe Publishing

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

four trainee readiness dimensions toward diversity training: pre-training (a) motivation to learn, (b) selfefficacy, (c) intention to use, and (d) perceived utility.

Interaction Between Perceived Ethnic Discrimination and Ethnic Dissimilarity on Readiness for Diversity Training In the previous sections, we argued that perceived ethnic discrimination and ethnic dyadic dissimilarity, independently, are likely to positively affect readiness for diversity training: however, these variables may interact with each other. Mathieu and Martineau’s (1997) conceptual framework of training motivation proposes that the extent to which individual characteristics influence training readiness may depend on work contexts that facilitate or constrain the influence of individual characteristics. In addition, research on diversity has suggested work contexts related to demographic characteristics may trigger or hibernate social categorization based on the characteristics (e.g., Avery et al., 2008; Chung et al., 2015). Presence of social settings or work contexts related to ethnicity may promote social identification and increase the salience of discrimination (Avery et al., 2008). On the contrary, if work contexts are not relevant to perceived ethnic discrimination, they may not be perceived as consequential. Following this logic, we argue that ethnic dyadic dissimilarity is a contextual factor that can facilitate or impede the influence of perceived ethnic discrimination on trainee readiness for diversity training. Because ethnic dyadic dissimilarity may elicit social categorization based on ethnicity and have trainees feel a greater need for diversity training, it may heighten the positive influence of perceived ethnic discrimination on training readiness for diversity training. On the contrary, because ethnic dyadic similarity may suppress social categorization based on ethnicity and reduce the importance of finding a remedy for discrimination issues, it may weaken the positive influence of perceived ethnic discrimination on training readiness for diversity training. We suggest that dyadic similarity or dissimilarity functions as a trigger for the cue of whether perceived ethnic discrimination will apply to the current work context. Specifically, we propose that Hypothesis 3: Ethnic dyadic dissimilarity positively moderates the relationship between perceived ethnic discrimination and readiness toward diversity training such that the relationships with pre-training (a) motivation to learn, (b) self-efficacy, (c) intention to use, and (d) perceived utility will be stronger when the subordinate’s ethnicity is different from the supervisor’s ethnicity. Ó 2016 Hogrefe Publishing

Method Sample and Procedures A total of 203 surveys were collected by 26 students enrolled in a Master’s level Human Resource Management program. After explaining the purpose and procedure of the survey, students were asked to provide survey questionnaires and envelopes to employed friends and colleagues in the US Participants were told that the survey was completely voluntary and anonymous. Students received one extra credit point for each survey they submitted. Respondents returned completed surveys by mail to the researchers to ensure confidentiality. Removal of missing values yielded a final sample of 160 subjects.

Measures Trainee Readiness for Diversity Training Respondents were asked to report their pre-training readiness for diversity training given the following hypothetical situation: “Suppose you are going to take diversity training. This set of questions asks you to describe how you feel about each of the following statements, before attending the diversity training program.” Four dimensions of readiness for diversity training were measured using existing measures (1 = “strongly disagree” to 7 = “strongly agree”): (a) Noe and Schmitt’s (1986) pretraining motivation to learn (four items) (e.g., “I am willing to exert considerable effort in diversity programs in order to improve my skills and knowledge.”). (b) Phillips and Gully’s (1997) pre-training self-efficacy (three items) (e.g., “I think that I can learn many skills and acquire knowledge through diversity programs.”). (c) Ajzen and Fishbein’s (1980) pre-training intention to use (two items) (e.g., “If given the opportunity, I intend to use what I learn in diversity programs in my job.”). (d) Ford and Noe’s (1987) pre-training perceived utility (four items) (e.g., “Diversity programs I might attend would be relevant to skills I had hoped to develop.”). To verify that the measure of trainee readiness for diversity training includes the four dimensions, we conducted confirmatory factor analyses (CFA). CFA results indicated that the four factor model fits the data reasonably well (w2df=59, n=160 = 104.59, RMSEA = .071, CFI = .95, TLI = .94, SRMR = .05) and fits significantly better than a one factor model (Δw2Δdf=6, n=160 = 80.66, p < .001, RMSEA = .11, CFI = .87, TLI = .84, SRMR = .06). We also found that the model with the higher-order factor adequately fits the data (w2df=61, n=160 = 108.22, RMSEA = .071, Journal of Personnel Psychology (2017), 16(1), 25–35

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

CFI = .95, TLI = .94, SRMR = .05) and fits significantly better than a one factor model (Δw2Δdf=4, n=160 = 77.31, p < .001). However, we did not find significant difference between the four factor model and the four factor model with the higher order (p > .1). Nevertheless, the CFA results show that both models fit the data much better than one factor solution. Also, the four factor model with a higher-order construct is more parsimonious than the four factor model. In addition, the four dimensions were highly correlated to each other and conceptually related to each other, and exhibit similar relationships with post-training outcomes. Therefore, trainee readiness for diversity training appears to be a higher-order construct that includes four dimensions. Perceived Ethnic Discrimination Four items of perceived ethnic discrimination were adopted from James, Lavato, and Cropanzano’s (1994) Workplace Prejudice/Discrimination Inventory (e.g., “At work I feel socially isolated because of my racial/ethnic background.”; 1 = “strongly disagree” to 7 = “strongly agree”). Ethnic Dyadic Dissimilarity Between a Supervisor and a Subordinate Ethnicity was categorized into four groups (white, black, Asian, and Hispanic). Following Tsui, Porter, and Egan (2002), and Avery et al. (2008), a value of 0 was assigned when ethnicity of a supervisor and a subordinate belonged to the same group, whereas a value of 1 was assigned when ethnicity of a supervisor and a subordinate belonged to different groups. Control Variables Control variables included gender (0 = male; 1 = female), citizenship (0 = US; 1 = non-US), industry (0 = others; 1 = manufacturing), tenure (the number of years working for the company), experience in diversity training (0 = no; 1 = yes), education (1 = high school graduate or less, 2 = college undergraduate, no degree, 3 = associate degree, 4 = four-year college degree, and 5 = graduate or professional degree). Using the Standard Occupational Classification, five dummy variables for job type were included: management, business and financial operations, computer and engineering occupations, and office and administrative support. The rationale for including these control variables is available from the authors.

Results The means, standard deviations, correlations among all key variables, and internal consistency reliabilities are shown in Table 1. All pre-training readiness items were factor-analyzed to evaluate the structure of the measures. The results from a principal component analysis with a varimax rotation yielded a four factor solution that matched the trainee readiness measures (74.7% of the variance). From the CFA results of readiness for diversity training and literature review, we argue that readiness for diversity training is a higher-order construct with four factors. Therefore, we performed multivariate multiple regression to examine how perceived ethnic discrimination and ethnic dyadic dissimilarity influence the linear combination of the four dimensions of trainee readiness for diversity training (Dattalo, 2013).1 This technique is part of the general linear model which takes into account intercorrelations among the multiple dependent variables and keeps Type I error from inflating (Tabachnick & Fidell, 2013). We also performed univariate multiple regressions to examine how these predictors influence each of the four dimensions of trainee readiness for diversity training (see Table 2). Overall, results from both methods were consistent with each other. The multivariate regression results showed that perceived ethnic discrimination was associated with the linear combination of the four dimensions of trainee readiness for diversity training (Wilks’ L = .94; Pillai’s Trace = .06; F(4, 144) = 2.39, p < .10, η2p = .06 for both). The univariate regression results indicated that perceived ethnic discrimination was positively associated with pretraining motivation to learn (β = .15, p < .10) and positively and significantly associated with pre-training utility (β = .17, p < .05), providing partial support for Hypothesis 1. Multivariate and univariate regression results showed that ethnic dyadic dissimilarity was not significantly predictive of pre-training readiness for diversity training; thus, Hypothesis 2 was not supported. We found robust support for Hypothesis 3. The multivariate multiple regression results showed that the interaction between perceived ethnic discrimination and ethnic dyadic dissimilarity was significantly associated with the linear combination of the four dimensions of trainee readiness for diversity training (Wilks’ L = .92; Pillai’s Trace = .08; F(4, 143) = 3.03, p < .05, η2p = .08 for both). The univariate regression results showed that the perceived ethnic

Using MPlus with the XWITH, we performed a structural equation modeling (SEM) with latent variables as well. The results were consistent with our findings from multivariate and univariate multiple regressions. First, the four dimensions were significantly associated with readiness for diversity training (p < .01). Second, perceived ethnic discrimination and dyadic dissimilarity were not significantly associated with readiness for diversity training. Last, the interaction term was significantly associated with readiness for diversity training (p < .01), supporting Hypothesis 3.

Journal of Personnel Psychology (2017), 16(1), 25–35

Ó 2016 Hogrefe Publishing

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

Table 1. Means, standard deviations, and correlations M 1. Pre-training motivation

3.83

0.73

.90

2. Pre-training self-efficacy 3.92

0.61

.60**

.58

3. Pre-training intention to 4.03 use 4. Pre-training utility 3.59

0.72

.50**

.64

0.70

.68**

.61**

.56**

.85

5. Education level

1.23

.08

.00

.24**

.01

3.78

.63

6. Gender

0.63

0.49

.34**

.24**

.30**

.21**

.12

7. Diversity training experience 8. Industry

0.51

0.50

.04

.01

.03

.07

.12

.06

0.31

0.46

.07

.01

.04

.01

.12

.20*

9. Management

0.28

0.45

.07

.01

.04

.11

.10

.12

.05

.27**

10. Finance

0.18

0.39

.02

.07

.00

.03

.07

.06

.13

.03

.29**

11. Engineering

0.14

0.35

.11

.04

.02

.01

.13

.14

.01

.16*

.25**

.19*

12. Admin Support

0.17

0.38

.12

.07

.21**

.25**

.14

.06

.16*

.28**

.21**

13. Tenure

6.25

6.74

.01

.08

.10

.03

.09

.13

.07

.32**

.05

.20*

.08

14. Citizenship

0.89

0.32

.03

.16*

.00

.05

.26**

.17*

.09

.02

.05

.04

.15

.11

.12

15. Ethnic discrimination

2.78

1.16

.20*

.02

.04

.20*

.15

.11

.05

.03

.10

.08

.02

.10

.03

.06

16. Ethnic dyadic dissimilarity

0.28

0.45

.17*

.11

.07

.17*

.10

.11

.03

.08

.07

.09

.12

.22** .14

.04

.18*

Notes. n = 160. *p < .05. **p < .01. Two-tailed. Internal consistency reliabilities are bold. Age was not controlled for due to the high correlation between tenure and age (r = .88). Of the total respondents, 60.3% were female, 66.2% were white, 16.2% were Asian, 13.6% were black, and 4% were Hispanic. Respondents (31.3%) worked at a firm in manufacturing industries and 68.8% of the respondents worked at a firm in service industries. For job types, 28% hold a management job, 18% hold a business and financial operations, 13.8% hold a computer and engineering occupations, and 27% hold an office and administrative support job. Seventy-two percent (n = 115) were same-ethnicity dyads and 28% (n = 45) were different-ethnicity dyads.

Table 2. Summary of hierarchical regression results DV1: Pre-training motivation

DV2: Pre-training selfefficacy

DV3: Pre-training intention to DV4: Pre-training utility use

Education level

.02

.00

.01

.03

.06

.05

.04

.29** .30** .30** .28** .07

.10

.09

.07

Gender

.34** .32** .30** .28** .22** .22** .21* .20* .32** .32** .32** .29** .20* .18* .16* .14†

Controls

Diversity training experience .01

.03

.04

.01

.00

.01

.08

.06

.05

Industry

.04

.02

.01

.05

.04

.03

.04

.05

.03

.02

Management

.06

.03

.00

.01

.02

.14

.15

.18

.03

.01

.00

.02

Finance

.02

.00

.01

.08

.04

.05

.04

.05

.08

.09

.08

Engineering

.08

.07

.09

.03

.04

.06

.04

.05

.06

.04

Admin Support

.04

.03

.05

.02

.01

.02

.15

.17

.22* .22* .21* .23*

Tenure

.04

.05

.08

.07

.06

.03

.04

Citizenship

.05

.03

.00

.02

.14

.16

.17

.01

.02

.04

.01

.03

.06

.07

.15†

.14†

.13†

.00

.01

.02

.13

.11

.10

0.01

0.04

Hypothesized variables Ethnic discrimination (RD) Ethnic dyadic dissimilarity (EDD) RD EDD

.21**

ΔR2

0.02

ΔF

3.84†

.05

.17* .16* .15

.01

.03

.12

0.00

0.05

.11 0.00

.23**

0.01 0.01

0.00

.11 .20*

0.03 0.01 0.04

2.48

7.33**

0.00

1.60 1.83

0.47

0.01

9.54**

0.13

0.15

0.17

0.21

0.09

0.10 0.11 0.19

0.19

0.24

0.10 0.13 0.14 0.17

Adjusted R2

0.07

0.09

0.10

0.14

0.03

0.02

0.03 0.03 0.14

0.13

0.18

0.04 0.06 0.07 0.10

2.24* 2.42** 2.45** 2.92** 1.47

1.32

1.35 1.40 3.55** 3.25** 2.96** 3.63** 1.63 1.95* 1.98* 2.38**

4.70* 2.18 6.26*

Note. n = 160. *p < .05. **p < .01. †p < .10. Two-tailed.

Ó 2016 Hogrefe Publishing

Journal of Personnel Psychology (2017), 16(1), 25–35

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

(a)

(b)

(c)

Figure 2. (a) The moderation of ethnic dyadic dissimilarity on the relationship between perceived ethnic discrimination and pre-training motivation to learn. (b) The moderation of ethnic dyadic dissimilarity on the relationship between perceived ethnic discrimination and pre-training intention to use. (c) The moderation of ethnic dyadic dissimilarity on the relationship between perceived ethnic discrimination and pre-training perceived utility.

Journal of Personnel Psychology (2017), 16(1), 25–35

discrimination and ethnic dyadic dissimilarity interaction was significantly associated with three of the dimensions of trainee readiness for diversity training (pre-training motivation to learn, intention to use, and utility) (see Table 2). However, pre-training self-efficacy was not significantly affected by this interaction. Figures 2a through 2c depict the interactions. Specifically, the results of simple slope tests (Aiken & West, 1991) showed that perceived ethnic discrimination was positively associated with three dimensions of trainee readiness for diversity training (pre-training motivation to learn, intention to use, and utility) under ethnic dissimilarity (β = .34, p < .05; β = .27, p < .05; β = .35, p < .05, respectively), whereas perceived ethnic discrimination was not significantly associated with readiness for diversity training under ethnic similarity (p > .55). Although not hypothesized, we performed post hoc analyses to examine a three-way interaction among a subordinate’s ethnicity, ethnic discrimination, and dyadic dissimilarity. We examined whether ethnic minorities report different relationships between perceived ethnic discrimination and readiness for diversity training from white employees. We found that ethnic minorities who worked with white supervisors exhibited more positive relationships between perceived ethnic discrimination and pre-training self-efficacy and between perceived ethnic discrimination and pre-training intention to use diversity skills and knowledge than the white employees who worked with either minority or white supervisors. We did not find significant three-way interactions for pre-training motivation to learn and utility.

Discussion Our goals were to assess the extent to which perceived ethnic discrimination and ethnic dyadic dissimilarity between a supervisor and a subordinate jointly influence readiness for diversity training. Our findings indicate that perceptions of ethnic discrimination are positively related to pre-training motivation to learn diversity-related content and pre-training perceived utility of the diversity training. Ethnic dyadic dissimilarity was not predictive of pre-training readiness for diversity training. However, individuals had significantly higher pre-training motivation to learn, intention to use, and perceived utility of diversity training content when both perceived ethnic discrimination and ethnic dyadic dissimilarity were present. A post hoc analysis further suggested such effects are strongest among ethnic minority employees who have white supervisors with regard to pre-training self-efficacy and intention to use diversity training. These findings contribute to the existing literature on diversity training and relational demography and provide implications as follows.

Ó 2016 Hogrefe Publishing

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

Assessing pre-training conditions and needs is critical for the success of diversity training: however, this area is relatively neglected (Roberson et al., 2013). Previous research has often measured pre-training readiness for or interests in diversity training using a single proxy variable (e.g., voluntary participation) and found no effect on training effectiveness (Buzrukova et al., 2012). To extend previous research, we focused on four multifaceted readiness variables that directly relate to pre-training interests in and motivation to learn diversity training. Findings contribute to diversity training research and highlight the importance of needs assessment for diversity training. The findings also speak to how relational demography (Avery et al., 2012) may influence the receptiveness to and effectiveness of diversity training initiatives. Although many previous research studies have found that dissimilarity creates interpersonal problems, findings from previous research on the effect of dyadic dissimilarity were inconsistent. Some found that dyadic dissimilarity lowers trust and increases employee withdrawal behaviors (e.g., Avery et al., 2012) while others found that it has no influence on employee behaviors and attitudes (e.g., Green, Whitten, & Medlin, 2005; Lau, Lam, & Salamon, 2008). This may be because relational demography does not solely determine interpersonal relationships (Wilk & Makarius, 2015): the effects of relational demography may be augmented (or diminished) in conjunction with other diversity-related experiences (e.g., workplace discrimination). Because employees work in a larger organizational context, how employees perceive diversity-related social contexts may play a significant role in employee work outcomes and supervisor-subordinate relationships (Lau et al., 2008). Therefore, as we found, dyadic dissimilarity may not have direct impact on readiness for diversity training but instead moderates the relationships between perceived ethnic discrimination and readiness for diversity training. In addition, our results showed that pre-training selfefficacy was not affected by the two-way interaction between perceived ethnic discrimination and ethnic dyadic dissimilarity. However, the three-way post hoc analysis suggests self-efficacy may be differentially influenced by dissimilarity and perceived discrimination, depending on employee ethnicity. The exploratory post hoc analyses suggest that ethnic minorities find interest in and need for diversity training only if they encounter problems due to ethnic discrimination and dissimilarity. Therefore, this study suggests that a “one size fits all” approach to diversity training may not be the best route to success. To maximize the effectiveness of diversity programs, companies may need to assess organizational and personal needs for diversity programs as the first step in the instructional design process (Roberson, Kulik, & Pepper, 2003). For organizations, an important issue Ó 2016 Hogrefe Publishing

involves the determination of who is supposed to attend diversity training and who is more likely to learn and respond to the content of diversity training. Ethnic dyadic dissimilarity and perceived ethnic discrimination may serve as contextual and individual characteristics to provide constructive information for the organizational needs analysis (e.g., individuals who have different ethnicity from their supervisors/subordinates should be selected as a priority). In addition, specialized interventions to increase pre-training readiness for diversity training may be put in place for trainees who are more likely to be unready for diversity training. Although this study provides results that extend our knowledge in the areas of diversity, training, and pretraining readiness, there are some limitations. First, perceived ethnic discrimination and four dependent variables were from the same source. However, method bias cannot easily explain interactions because it is an artificial deflation or inflation across the covariance/ correlation matrix due to measurement, timing, or other unevaluated artifacts (Siemsen, Roth, & Oliveira, 2010). Second, snowball sampling may have selection bias because students do not randomly recruit survey participants (Wheeler, Shanine, Leon, & Whitman, 2014). However, meta-analytic research has found that snowball samples are not demographically different from other samples in management research and that research results using snowball samples are not substantially different from those using other samples (Wheeler et al., 2014). In addition, snowball sampling can be more appropriate when research is on a sensitive issue (Biernacki & Waldorf, 1981), which is our case. Thus, even if our sample was not randomly selected, we expect that it would not be significantly biased. Nonetheless, we call for future research to adopt random sampling which is less susceptible to selection bias. Third, despite the importance of perceived ethnic discrimination and dissimilarity in determining pre-training readiness for diversity training, it is critical to know how diversity-related individual and contextual variables influence post-training outcomes through various mediators (e.g., Mathieu et al., 1992). Hence, we call for research to continue to examine the connection between pre- and post-training reactions and attitudes. Fourth, we have chosen ethnicity as our focal demographic characteristic because it has been identified as one of the most significant and identifiable attributes that influence interpersonal differences (Wilk & Makarius, 2015). However, drawing on the same theories (e.g., expectancy theory, social identity theory, etc.) and conceptual arguments, perceived discrimination and dyadic dissimilarity based on other demographic attributes may positively predict readiness for diversity training. Thus, we call for future research that examines this area. Journal of Personnel Psychology (2017), 16(1), 25–35

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

Fifth, since most training programs in the US deal with several demographic attributes and ethnicity is a focal characteristic of the programs, respondents are likely to report their readiness considering ethnicity is a core characteristic in diversity training. Nevertheless, it is possible that respondents might have interpreted diversity training tied to specific other demographic attributes (e.g., gender). This limitation can be addressed by measuring readiness for diversity training for each demographic attribute separately. Finally, this study focuses on ethnic dyadic dissimilarity as a contextual variable. Employees spend a great deal of time with an immediate supervisor at work. Because of frequent interactions between a subordinate and a supervisor at a workplace, potential tensions, and conflicts caused by dyadic dissimilarity can become crucial issues and generate different levels of receptiveness toward diversity training initiatives. However, other variables (e.g., broader group and organizational diversity) related to diversity training content may also influence readiness for diversity training. Diversity climates (Chung et al., 2015) and diversity beliefs (Van Dick, van Knippenberg, Hagele, Guillaume, & Brodbeck, 2008) that mitigate social categorization may alter the influence of perceived discrimination and dissimilarity on readiness for diversity training. Therefore, we call for future research to investigate the influence of group and organizational diversity, or other diversity contextual variables on readiness for diversity training. To conclude, this study sheds light on pre-training mechanisms of diversity training by identifying four dimensions of employee readiness for diversity initiatives. We examine perceived ethnic discrimination and ethnic dissimilarity as individual and contextual determinants of readiness for diversity training. These findings provide implications for needs assessment and effectiveness of diversity training efforts.

References Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. doi: 10.1016/0749-5978(91)90020-T Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall. Alliger, G. M., Tannenbaum, S. I., Bennett, W. Jr., Traver, H., & Shotland, A. (1997). A meta-analysis of the relations among training criteria. Personnel Psychology, 502, 341–358. doi: 10.1111/j.1744-6570.1997.tb00911.x Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy of Management Review, 14, 20–39. doi: 10.2307/258189 Avery, D. R., McKay, P. F., & Wilson, D. C. (2008). What are the odds? How demographic similarity affects the prevalence of

Journal of Personnel Psychology (2017), 16(1), 25–35

perceived employment discrimination. Journal of Applied Psychology, 93, 235–249. doi: 10.1037/0021-9010.93.2.235 Avery, D. R., Volpone, S. D., Mckay, P. K., King, E. B., & Wilson, D. C. (2012). Is relational demography relative? How employment status influences effects of supervisor-subordinate demographic similarity. Journal of Business Psychology, 27, 83–98. doi: 10.1007/s10869-011-9230-9 Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Biernacki, P., & Waldorf, D. (1981). Snowball sampling: Problems and techniques of chain referral sampling. Sociological Methods & Research, 10, 141–163. doi: 10.1177/ 004912418101000205 Blanchard, P. N., & Thacker, J. W. (1998). Effective training: Systems, strategies, and practices. NJ: Prince Hall. Buzrukova, K., Jehn, K. A., & Spell, C. S. (2012). Reviewing diversity training: Where we have been and where we should go. Academy of Management Learning & Education, 11, 207–227. doi: 10.5465/amle.2008.0090 Byrne, D. (1971). The attraction paradigm. New York, NY: Academic Press. Chung, Y. (2013). Trainee readiness for diversity training. Journal of Diversity Management, 8, 77–84. doi: 10.19030/jdm.v8i2. 8234 Chung, Y., Liao, H., Jackson, S. E., Subramony, M. A., Colakoglu, S., & Jiang, Y. (2015). Cracking but not breaking: Joint effects of faultline strength and diversity climate on loyal behavior. Academy of Management Journal, 58, 1495–1515. doi: 10.5465/amj.2011.0829 Colquitt, J. A., LePine, J. A., & Noe, R. A. (2000). Toward an integrative theory of training motivation: A meta-analytic path analysis of 20 years of research. Journal of Applied Psychology, 855, 678–707. doi: 10.1037/0021-9010.85.5.678 Combs, G. M., & Luthans, F. (2007). Diversity training: Analysis of the impact of self-efficacy. Human Resource Development Quarterly, 18, 91–120. doi: 10.1002/hrdq.1193 Corobot-Mason, D., Konrad, A. M., & Linnehan, F. (2006)Measures for quantitative diversity scholarship. In A. M. Konrad P. Prasad, & J. K. Pringle (Eds.), Handbook of workplace diversity (pp. 237–270). Thousand Oaks, CA: Sage. Dattalo, P. (2013). Analysis of multiple dependent variables New York, NY, Oxford University Press. doi: 10.1093/acprof: oso/9780199773596.001.0001 Equal Employment Opportunity Commission (EEOC). (2015). Charge statistics FY 1997 through FY 2014. Retrieved from http://www1.eeoc.gov/eeoc/statistics/enforcement/charges. cfm Ford, J. K., & Noe, R. A. (1987). Self-assessed training needs: The effects of attitudes towards training, managerial level, and function. Personnel Psychology, 40, 39–53. doi: 10.1111/j.17446570.1987.tb02376.x Green, K. W., Whitten, D., & Medlin, B. (2005). Impact of relational differences on supervisor/subordinate dyad. Industrial Management & Data Systems, 105, 369–383. doi: 10.1108/ 02635570510590165 Goldman, B. M., Gutek, B. A., Stein, J. H., & Lewis, K. (2006). Employment discrimination in organizations: Antecedents and consequences. Journal of Management, 32, 786–830. doi: 10.1177/0149206306293544 Gully, S. M., & Chen, G. (2010). Individual differences, attributetreatment interactions, and training outcomes. In S. W. J. Kozlowski & E. Salas (Eds.), Learning, training, and development in organizations (pp. 3–64). San Francisco, CA: Jossey-Bass, SIOP Organizational Frontiers Series. Holladay, C. L., Knight, J. L., Paige, D. L., & Quiñones, M. A. (2003). The influence of framing on attitudes toward diversity training.

Ó 2016 Hogrefe Publishing

Y. Chung et al., Readiness for Diversity Training, Discrimination, and Dyadic Dissimilarity

Human Resource Development Quarterly, 14, 245–263. doi: 10.1002/hrdq.1065 James, K., Lavato, C., & Cropanzano, R. (1994). Correlation and known-group comparison validation of a workplace prejudice/ discrimination inventory. Journal of Applied Social Psychology, 24, 1573–1592. doi: 10.1111/j.1559-1816.1994.tb01563.x Kalinoski, Z. T., Steele-Johnson, D., Peyton, E. J., Leas, K. A., Steinke, J., & Bowling, N. A. (2013). A meta-analytic evaluation of diversity training outcomes. Journal of Organizational Behavior, 34, 1076–1104. doi: 10.1002/job.1839 King, E. B., Dawson, J. F., Kravitz, D. A., & Gulick, L. M. V. (2012). Multilevel relationships between diversity training, discrimination, and job satisfaction. Journal of Organizational Behavior, 33, 5–20. doi: 10.1002/job.728 Kulik, C. T., Pepper, M. B., Roberson, L., & Parker, S. K. (2007). The rich get richer: Predicting participation in voluntary diversity training. Journal of Organizational Behavior, 28, 753–769. doi: 10.1002/job.444 Lau, D. C., Lam, L. W., & Salamon, S. D. (2008). The impact of relational demographics on perceived managerial trustworthiness: Similarity or norms? Journal of Social Psychology, 148, 187–208. doi: 10.3200/socp.148.2.187-209 Mathieu, J. E., & Martineau, J. W. (1997). Individual and situational influences on training motivation. In J. K. Ford, S. W. J. Kozlowski, K. Kraiger, E. Salas, & M. S. Teachout (Eds.), Improving training effectiveness in work organizations (pp. 193–221). Mahwah, NJ: Erlbaum. Mathieu, J. E., Tannenbaum, S. I., & Salas, E. (1992). Influences of individual and situational characteristics on measures of training effectiveness. Academy of Management Journal, 35, 828–847. doi: 10.2307/256317 Mor Barak, M. E., Cherin, D. A., & Berkman, S. (1998). Organizational and personal dimensions in diversity climate: Ethnic and gender differences in employee perceptions. Journal of Applied Behavioral Science, 34, 82–104. doi: 10.1177/ 0021886398341006 Noe, R. A. (2010). Employee training and development. New York, NY: McGraw-Hill/Irwin. Noe, R. A., & Schmitt, N. (1986). The influence of trainee attitudes on training effectiveness: Test of a model. Personnel Psychology, 39, 497–523. doi: 10.1111/j.1744-6570.1986.tb00950.x Patrick, J., Smy, V., Tombs, M., & Shelton, K. (2012). Being in one’s chosen job determines pre-training attitudes and training outcomes. Journal of Occupational and Organizational Psychology, 85, 245–257. doi: 10.1111/j.2044-8325.2011.02027.x Phillips, J. M., & Gully, S. M. (1997). The role of goal orientation, ability, need for achievement, and locus of control in the self-efficacy and goal setting process. Journal of Applied Psychology, 82, 792–802. doi: 10.1037/0021-9010.82.5.792 Roberson, L., Kulik, C. T., & Pepper, M. B. (2001). Designing effective diversity training: Influence of group composition and trainee experience. Journal of Organizational Behavior, 22, 871–885. doi: 10.1002/job.117 Roberson, L., Kulik, C. T., & Pepper, M. B. (2003). Using needs assessment to resolve controversies in diversity training design. Group & Organization Management, 28, 148–174. doi: 10.1177/1059601102250028 Roberson, L., Kulik, C. T., & Pepper, M. B. (2009). Individual and environmental factors influencing the use of transfer strategies after diversity training. Group and Organization Management, 34, 67–89. doi: 10.1177/1059601108329732 Roberson, L., Kulik, C. T., & Tan, R. Y. (2013). Effective diversity climate. In Q. M. Roberson (Ed.), The Oxford handbook of diversity and work (pp. 341–365). New York, NY: Oxford University Press.

Ó 2016 Hogrefe Publishing

Siemsen, E., Roth, A., & Oliveira, P. (2010). Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods, 13, 456–476. doi: 10.1177/1094428109351241 Sitzmann, T., Brown, K. G., Casper, W. J., Ely, K., & Zimmerman, R. E. (2008). A review and meta-analysis of the nomological network of trainee reactions. Journal of Applied Psychology, 93, 280–295. doi: 10.1037/0021-9010.93.2.280 Society for Human Resource Management. (2010). SHRM research spotlight: Workplace diversity practices poll. Retrieved from http://www.shrm.org/Research/SurveyFindings/Articles/ Documents/DiversityFlier_FINAL_spotlight.pdf Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics (6th ed.). Upper Saddle River, NJ: Pearson. Tracey, J. B., Hinkin, T. R., Tannenbaum, S. I., & Mathieu, J. E. (2001). The influence of individual characteristics and the work environment on varying levels of training outcomes. Human Resource Development Quarterly, 12, 5–23. doi: 10.1002/15321096(200101/02)12:1%3C5::aid-hrdq2%3E3.0.co;2-j Tsui, A. S., & O’Reilly, C. A. III (1989). Beyond simple demographic effects: The importance of relational demography in superiorsubordinate dyads. Academy of Management Journal, 32, 402–423. doi: 10.2307/256368 Tsui, A. S., Porter, L. W., & Egan, T. D. (2002). When both similarities and dissimilarities matter: Extending the concept of relational demography. Human Relations, 55, 899–929. doi: 10.1177/0018726702055008176 Turner, J. C., & Haslam, S. A. (2001). Social Identity, organizations and leadership. In M. E. Turner (Ed.), Groups at work: Theory and research. Mahwah, NJ: Erlbaum. Van Dick, R., van Knippenberg, D., Hagele, S., Guillaume, Y., & Brodbeck, F. (2008). Group diversity and group identification: The moderating role of diversity beliefs. Human Relations, 61, 1463–1492. doi: 10.1177/0018726708095711 Wesolowski, M., & Mossholder, K. (1997). Relational demography in supervisor-subordinate dyads: Impact on job satisfaction, burnout, and perceived procedural justice. Journal of Organizational Behavior, 18, 351–362. doi: 10.1002/(sici)10991379(199707)18:4%3C351::aid-job802%3E3.0.co;2-# Wheeler, A. R., Shanine, K. K., Leon, M. R., & Whitman, M. W. (2014). Student-recruited samples in organizational research: A review, analysis, and guidelines for future research. Journal of Occupational and Organizational Psychology, 87, 1–26. doi: 10.1111/joop.12042 Wiethoff, C. (2004). Motivation to learn and diversity training: Application of the theory of planned behavior. Human Resource Development Quarterly, 15, 263–278. doi: 10.1002/ hrdq.1103 Wilk, S. L., & Makarius, E. E. (2015). Choosing the company you keep: Racial relational demography outside and inside of work. Organization Science, 26, 1316–1331. doi: 10.1287/orsc. 2015.0991 Received December 5, 2015 Revision received April 19, 2016 Accepted April 20, 2016 Published online September 22, 2016 Yunhyung Chung College of Business and Economics University of Idaho 875 Perimeter Drive MS 3161 Moscow, ID 83844-3161 yunchung@uidaho.edu

Journal of Personnel Psychology (2017), 16(1), 25–35

Managing and nurturing talent in the workplace

“Robert and Marion Edenborough bring research, a wide knowledge of various psychological theories, experience in the field, and wisdom to their discussion of talent... This is a ‘must read’ for anyone interested in hiring or developing staff!” Christy Hammer, PhD, Former Senior Analyst and Leader of International Development with Gallup

Robert Edenborough / Marion Edenborough

The Psychology of Talent Exploring and Exploding the Myths 2012, viii + 152 pages US $34.80 / € 24.95 ISBN 978-0-88937-396-9 Also available as eBook The core concepts in this book are the idea of talent, how it can be assessed, and how it can be nurtured and put to effective use in the workplace. Line managers, HR professionals, business or industrial/organizational psychologists, and consultants will find their understanding challenged and extended - and are shown how to improve their professional practices. The authors explore various psychological tools and approaches that can be pressed into service in connection with talent. Uniquely, they also set

www.hogrefe.com

the psychological assessment of talent in the context of attitudes to talent and various myths and misunderstandings about it. The positive psychology/strengths movement and the relation between psychology and talent management are also explored in a clear and objective manner. This easy-to-read volume will be of interest to anyone concerned with understanding how talent can be pressed into service to improve performance in the workplace.

Fredrike Bannink

Handbook of Positive Supervision for Supervisors, Facilitators, and Peer Groups 2015, xii + 204 pp. US $49.00 / € 34.95 ISBN 978-0-88937-465-2 Also available as eBook

Positive supervision focuses on what actually works instead of on problems and on supervisees’ strengths rather than on their deficits. The task of supervisors using this approach is – unlike the more traditional problemsolving – to create solutions with their supervisees and to teach them to apply the same approach when working with their own clients.

Essential reading for all supervisors, this book introduces a new form of supervision, based on positive psychology and solution-focused brief therapy, that is shorter, more positive and hopeful, and more cost-effective than traditional methods.

Fredrike Bannink

Handbook of Solution-Focused Conflict Management 2010, xii + 180 pp., hardcover US $49.00 / € 34.95 ISBN 978-0-88937-384-6 Also available as eBook

Benjamin Franklin once said: “Every problem is an opportunity in disguise.” In the new and highly successful approach of solution-focused conflict management described here, the focus is on discovering these opportunities to find the “win-win” scenario. The key lies in asking eliciting questions about goals, exceptions, and competencies and in motivating clients to change. Clients’ perspectives are considered primary, and they are empowered to formulate their own hopes for the future and to devise ways to make them happen. Focusing on the preferred future facilitates change in the desired direction.

www.hogrefe.com

This book is essential reading for all those who need to manage conflicts. It provides a detailed description of the highly successful solution-focused model, its theoretical background, and practical applications in conflict management practice: divorce, workplace, family, neighbors, personal injury cases, victim-offender conflicts. Kenneth Cloke, one of the most inspiring mediators in the world, wrote the Foreword and Epilogue.

Original Article

How Mindset Matters The Direct and Indirect Effects of Employees’ Mindsets on Job Performance Matt Zingoni1 and Christy M. Corey2 1

Department of Management & Marketing, University of New Orleans, LA, USA

College of Business Administration, University of New Orleans, LA, USA Abstract: Individuals vary in their mindsets – their implicit beliefs regarding the malleability of human attributes. Because mindset influences responses to achievement situations, we hypothesize that employees with a more incremental mindset (i.e., believing that human attributes can be changed through effort and hard work) will have higher job performance and better relationships with their manager. We found that employees with a more incremental mindset have higher job performance. Also, when their manager has a strong learning goal orientation, employees with a more incremental mindset have a higher quality relationship with their manager, which, in turn, positively relates to their job performance. Keywords: implicit person theory, job performance, individual differences

Individuals vary in their mindsets – that is, their implicit beliefs and assumptions regarding the malleability of human attributes (Dweck, 1999; Dweck & Leggett, 1988). At one end of the continuum are those with an entity mindset; they believe that human attributes are fixed and cannot be changed. At the other end of the continuum are those with an incremental mindset; they believe that human attributes can be changed through effort and hard work. Where individuals fall on this continuum has been found to have profound effects on their thoughts and behaviors (e.g., Dweck, 1999; Dweck, Chiu, & Hong, 1995). To date, with few exceptions (e.g., Heslin, Latham, & VandeWalle, 2005; Heslin & VandeWalle, 2011; Heslin, VandeWalle, & Latham, 2006), most research on mindset has been conducted in nonwork settings (e.g., Dweck, 1999; Ommundsen, 2001). Due to the fact that individuals’ mindset influences how they respond to achievement situations and how they relate to others (see Dweck, 1999 for a review), we contend that employees’ mindset is likely to influence important work-related outcomes such as their job performance and their relationships with others. Namely, we expect that employees with a more incremental mindset will have higher job performance and have a higher quality relationship with their manager. Managerial characteristics such as their learning goal orientation – the extent to which they focus on developing capabilities and increasing competencies in themselves and others may moderate the relationship between employees’

Journal of Personnel Psychology (2017), 16(1), 36–45 DOI: 10.1027/1866-5888/a000171

mindset and the quality of their relationship. Because managers with a strong learning goal orientation are more focused on developing and supporting their employees, they may be well matched with employees with a more incremental mindset because these employees desire and believe they are capable of further development (Hong, Chiu, Dweck, Lin, & Wan, 1999). The purpose of this study was to contribute to the potential theoretical and practical implications. First, we aim to further extend the literature on mindset in the work domain by examining whether employees’ mindset influences their job performance and the quality of their relationship with their manager. Second, this study is also likely to extend theory on mindset by incorporating situational factors that may moderate the mindset-performance relationship. In this study we use a sample from the medical sales industry. The sales industry is a strong domain to examine contextual factors on the influence of mindset and performance. In particular, mindset is likely to have a significant direct influence on sales performance as sales people frequently encounter setbacks and failures making the persistence and desire to learn reflected by those with an incremental mindset very valuable. On the other hand, the sales domain offers a conservative context to examine the indirect influence on job performance through relationship quality with managers, as managerial relationships are less important in a sales setting because sales employees, at certain stages of their career, require less guidance from their manager (Kohli, Shervani, & Challagalla, 1998). Ó 2016 Hogrefe Publishing

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

Lastly, the results of our study may have implications for selection and performance management. Specifically, our findings may speak to whether employee mindset could be used in selection decisions and whether training to increase employee’s incremental mindset should be conducted to increase employee performance.

Theoretical Background and Hypotheses In this study, we draw on Dweck and colleagues’ socialcognitive model of individual mindset (e.g., Dweck, 1999; Dweck & Leggett, 1988). In this model, individuals’ implicit beliefs about the fixedness or malleability of ability influence their motivational and behavioral responses, particularly when experiencing failure. Those with more of an incremental mindset, who believe that ability is largely a function of effort, tend to respond to failure with perseverance, adopting behavioral strategies that facilitate task mastery (e.g., Henderson & Dweck, 1990). Those with more of an entity mindset, who believe that ability is largely immutable, tend to respond to failure with a helplessness response. They experience negative affect and seek to avoid further related challenges because they believe failure and a need for effort indicates low ability (e.g., Henderson & Dweck, 1990). This present study extends research on individuals’ mindset that has been conducted in nonwork settings and work settings. Many studies in nonwork settings have examined the relationship between individuals’ mindset and their performance on tasks (e.g., Dweck, 1999) and their approach to relationships (Knee, Nanayakkara, Vietor, Neighbors, & Patrick, 2001). There are a few studies that have examined individuals’ mindset in work settings (e.g., Heslin et al., 2005, 2006). Of the studies in work settings we were able to locate, most have focused on managers’ mindset and their consequent treatment of employees (Heslin & VandeWalle, 2011; Heslin et al., 2005, 2006). Thus, the present study expands on these studies’ findings by considering how nonmanagerial employees’ mindset affects their performance and other outcomes. Figure 1 summarizes the proposed hypotheses examined in the present study.

Employee Mindset and Employee Job Performance Employees with more of an incremental mindset are likely to have better job performance for several reasons. First, they are likely to respond to work situations with a learning Ó 2016 Hogrefe Publishing

orientation. Past research has found that individuals with a more incremental mindset tend to view achievement situations as opportunities to learn and improve their skills and abilities (e.g., Mueller & Dweck, 1998). Additionally, some research suggests that they are more likely to adopt learning goals (e.g., Dweck, 1999; Dweck & Leggett, 1988). However, not all research has found support for the relationship between mindset and learning goal orientation (e.g., Cury et al., 2006; Tabernero & Wood, 1999; VandeWalle, 1997), suggesting that other mechanisms besides goal orientation – which I detail below – may link mindset to task performance. Second, those with a more incremental mindset tend to be more self-efficacious and persistent when they experience failure (e.g., Henderson & Dweck, 1990). Rather than giving up, they tend to perceive failure as an indication that they need to increase their effort (Henderson & Dweck, 1990). In contrast, individuals with more of an entity mindset (i.e., less of an incremental mindset) tend to view achievement situations as an opportunity to validate their perceived fixed level of ability (Dweck & Leggett, 1988). They tend to perceive failure as an indication that success is beyond their reach – causing them to lose confidence and interest, and reduce effort or give up (e.g., Cury et al., 2006; Tabernero & Wood, 1999). Third, employees with more of an incremental mindset are more likely to find work to be intrinsically motivating. Past research has found that, as compared to those with a more entity mindset, individuals with a more incremental mindset tend to be more intrinsically motivated (Dweck, 1986). Because they are more driven to learn and improve from achievement situations, they respond to challenges with feelings of interest – rather than with feelings of defeat. Therefore, because those with more of an incremental mindset focus on learning, are more efficacious and persistent in the face of setbacks, we propose the following hypothesis: Hypothesis 1: Employees who have a more incremental mindset have higher job performance.

Employees’ Mindset and Relationship Quality In addition to being related to job performance, employees’ mindset may also be related to the quality of their relationship with their manager. Namely, employees with more of an incremental mindset are likely to have a better quality relationship with their manager. That is, consistent with Janssen and Van Yperen’s (2004) definition of a high quality employee-manager relationship, the relationship with their manager is more likely to be characterized by a Journal of Personnel Psychology (2017), 16(1), 36–45

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

Figure 1. Proposed model of relationships.

Manager’s learning goal orientation Hypotheses 4 & 5 Hypothesis 3

Manager-employee relationship quality Hypothesis 2

Employee mindset

Employee job performance Hypothesis 1

high level of mutual trust, respect, and an open exchange of information. There are several reasons we expect this. First, employees with more of an incremental mindset are likely to perceive relationships as worthy of effort and to be willing to invest effort into their relationship with their manager. High quality relationships need time to mature and develop; they often require concerted effort (Graen & Uhl-Bien, 1995). Because individuals with more of an incremental mindset are more likely to persist at situations that are effortful (Mueller & Dweck, 1998), we expect that employees with a more incremental mindset are likely to invest their time and effort into a relationship. Additionally, because they are more likely to view themselves and others as capable of change (Heyman & Dweck, 1998), they are more likely to view relationships as having the potential to improve. Research has found that individuals who believe that their intimate relationships can change overtime invest more effort in strategies designed to maintain that relationship (Knee, 1998). Second, employees with more of an incremental mindset may be more likely to view their manager as a resource that can assist in their development. Individuals with more of an incremental mindset are more concerned with their development and are less likely to believe that effort indicates low ability (see Chiu, Hong, & Dweck, 1997 for a review). Accordingly, more incremental employees may be more likely to seek feedback from their manager and to respond favorably to the feedback, thus, leading to better informational exchanges between the employees and their manager. In contrast, individuals with a more entity mindset have been shown to be less interested in personal growth and more interested in gaining self-validation in their relationships with intimate partners (Knee et al., 2001). Accordingly, more entity employees may seek less feedback from their manager and respond less favorably to the feedback, thus, leading to fewer and less informational exchanges between more entity mindset employees and their manager. Third, employees with more of an incremental mindset may be more likely to elicit trust and respect from their manager, thus increasing the quality of their relationship. Since employees with an incremental mindset are expected Journal of Personnel Psychology (2017), 16(1), 36–45

to respond to setbacks with persistence or increased effort. This persistence and effort is likely to increase their manager’s trust and respect of them, thus, increasing the quality of their relationship. In contrast, employees with a more entity mindset tend to respond poorly to setbacks and have deteriorating performance after setbacks (Tabernero & Wood, 1999). This maladaptive pattern of behavior is likely to erode their manager’s trust and respect, thus, decreasing the quality of their relationship. In sum, because those with a more incremental mindset will put forth more effort into the relationship, view the relationship as a valued resource for development, and garner more respect and trust from their manager, we propose the following hypothesis: Hypothesis 2: Employees with a more incremental mindset have higher quality relationships with their manager.

Manager’s Goal Orientation as a Moderator Above we argued that employees with a more incremental mindset are likely to have higher quality relationships with their manager. It is possible that this relationship depends on the manager’s learning goal orientation – the extent to which the manager focuses on developing capabilities and increasing competencies – as this orientation helps to establish the manager’s expectations. Because the quality of a relationship depends on the expectations and behaviors of both employee and manager (Graen & Uhl-Bien, 1995), a manager’s learning goal orientation may interact with their employees’ mindset to determine the quality of their relationship. When their manager has a strong learning goal orientation, the relationship between employees’ mindset and the manager-employee relationship quality is expected to be more strongly positive than when their manager has a weak learning goal orientation. We expect this to be the case because, when their manager has a strong learning goal orientation, employees who have a more incremental Ó 2016 Hogrefe Publishing

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

mindset are likely to have a high quality relationship with their manager due to a match between their mindset and the manager’s goal orientation. Further, we expect that, when their manager has a strong learning goal orientation, employees who have a more entity mindset are likely to have a low quality relationship with their manager due to a mismatch between their mindset and the manager’s goal orientation. When individuals have a strong orientation toward learning goals, they are more focused on developing capabilities in themselves and others (Godshalk & Sosik, 2003). Employees with more of an incremental mindset are likely to be receptive to this orientation because they are also focused on developing their capabilities. In comparison, because employees with a less incremental mindset believe that their capabilities are fixed (Chiu et al., 1997), they are likely to be reluctant to invest time and effort in developing their capabilities, which would be at odds with their manager’s strong learning goal orientation and potentially harm their relationship with their manager. Further, managers with a strong learning goal orientation are likely to offer more information and feedback in order to support the learning process (Janssen & Van Yperen, 2004). Employees with more of an incremental mindset, who are likely to see their managers as a source of information and guidance that help cultivate their ability, are likely to respond favorably to this feedback and guidance. They are also likely to believe that their manager is responsive to their need for information and feedback. In contrast, for employees with a more entity mindset are likely to be unreceptive to the information and feedback offered by their manager. Conversely, when their manager has a weak learning goal orientation, we expect the relationship between employee mindset and the manager-employee relationship quality to be less positive. When their manager has a weak orientation toward learning goals, employees who have a more incremental mindset may view their manager as not offering the guidance and feedback that they need to improve their capabilities. Due to their tendency to persevere and to accept challenges, we expect that incremental employees may adapt to the situation. When employees have a less incremental mindset, their manager’s lack of focus on the employees’ learning is also unlikely to harm their relationship. For these reasons, we expect that, when their manager has a weak learning goal orientation, the relationship between employee mindset and the manager-employee relationship quality will be less positively related. Hypothesis 3: The relationship between employees’ mindset and the quality of their relationship with

Ó 2016 Hogrefe Publishing

their manager is moderated by the manager’s learning goal orientation, such that the relationship is more positive when a manager has a strong learning goal orientation than when their manager has a weak learning goal orientation.

The Mediating Role of Relationship Quality As outlined above, we expect employees’ mindset to have a direct effect on their performance. We also expect employees’ mindset to influence performance indirectly through manager-employee relationship quality. That is, manager-employee relationship quality will serve as a mediator between employees’ mindset on their job performance and, as suggested above, this mediated effect will be moderated by their manager’s learning goal orientation. First, we will outline why relationship quality will mediate the effect of employees’ mindset on their job performance. Then we will discuss how this mediation is dependent on their manager’s learning goal orientation. High quality relationships are marked by an open exchange of information between both parties in the relationship. In addition, these relationships characteristically have high levels of trust and loyalty resulting from the mutual investment of time and effort in the relationship (Graen & Uhl-Bien, 1995). When their manager is loyal to them, employees may reciprocate, demonstrating their loyalty to their manager by putting forth effort toward their job. Furthermore, high quality relationships are likely to foster the development of employees’ competencies, through increased exchange of resources, information, and support, which in turn may increase employees’ job performance. Indeed, prior research has found that high quality manager-employee relationships are positively related to employees’ job performance (e.g., Basu & Green, 1997; Janssen & Van Yperen, 2004). Thus, we propose the following hypothesis: Hypothesis 4: Manager-employee relationship quality mediates the relationship between employees’ mindset and their job performance. As we argued above, we expect manager’s learning goal orientation moderates the relationship between employees’ mindset and the quality of their relationship with their manager. Specifically, we proposed that employees’ with a more incremental mindset will have a higher quality relationship with their manager when their manager has a strong learning goal orientation than when a manager has a weak learning goal orientation. Therefore, we hypothesize

Journal of Personnel Psychology (2017), 16(1), 36–45

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

that the mediating effect of relationship quality will be dependent on the strength of the manager’s orientation toward learning goals. Thus we propose: Hypothesis 5: Manager learning goal orientation moderates the strength of the mediated relationship between employees’ mindset and their job performance via manager-employee relationship quality, such that the mediated relationship will be stronger when the manager has a strong learning goal orientation than when the manager has a weak learning goal orientation.

Method Sample Descriptions and Procedure Through the use of our business networks we collected data from two different sources, employees and their respective manager. First, we invited 129 medical sales employees in the United States with similarly sized territories and job scopes from the same company to complete a web-based survey including items regarding their mindset, their relationship with their manager, and other constructs. Then, we asked in person, each manager of each target employee to complete a survey including items on their learning goal orientation, the employee’s job performance, and other constructs. Employees were notified that, with their consent, we would asking their respective managers to complete a survey as well. Of the 129 employees asked to participate, 96 completed the survey and had a survey completed by their manager, yielding a 74% overall response rate. In terms of demographics, 68% of the sales employees in our sample were male. On average, they had 5.8 years of experience in the industry. Their managers were mostly male (62%) and had, on average, 8.1 years of experience in the industry. The average length of the manageremployee relationships was 2.4 years.

Measures Employee Mindset We used the 8-item general kind of person scale to measure employee mindset (Levy, Stroessner, & Dweck, 1998). We used this measure instead of a domain-specific measure because we were concerned with employees’ implicit beliefs across various domains. Additionally, we believe that a general measure is more likely relevant to employees’ work relationships and job performance. Research supports the construct validity of this measure. For example, Journal of Personnel Psychology (2017), 16(1), 36–45

its discriminant validity is suggested by low correlations with academic aptitude, optimism, self-esteem, political beliefs, and locus of control (see Dweck, 1999 for review). The scale consists of eight items, including “Everyone, no matter who they are, can significantly change their basic characteristics” and “The kind of person someone is, is something very basic about them, and it can’t be changed very much” (reverse-scored). Employees responded to each item using a 6-point Likert scale. We averaged the items, such that a higher score indicates a more incremental mindset (α = .93). Manager-Employee Relationship Quality We used seven items based on the LMX-7 (LeadershipMember Exchange) scale developed by Graen, Novak, and Sommerkamp (1982) to measure the quality of the manager-employee relationship from the employee’s perspective. Sample items are “My working relationship with my supervisor is effective,” and “Usually know where I stand with my manager.” Employees responded to each item using a 5-point Likert scale. We averaged the items, such that a higher score indicates a higher quality relationship (α = .84). Manager Learning Goal Orientation Because the sales managers were also involved in sales, we measured the learning goal orientation of the managers using a 6-item measure of learning goal orientation designed for a sales context (Sujan, Weitz, & Kumar, 1994). Sample items are “It is important for me to learn from each selling experience I have,” and “An important part of being a good salesperson is continually improving your sales skills.” Managers responded to each item using a 5-point Likert scale. We averaged the items, such that a higher score indicates a stronger preference for a learning goal orientation (α = .88). Job Performance We measured employee job performance by asking managers to respond to three items from the job performance scale developed by Williams and Anderson (1991). Sample items are “This individual meets performance expectations,” and “This individual fulfills the responsibilities specified in his or her job description.” Manager responded to each item using a 5-point Likert scale. We averaged the items, such that a higher score indicates better job performance (α = .90). Control Variables We controlled for the sex of the employee because research has shown that there may be sex differences in performance ratings (Atwater, Brett, Waldman, DiMare, & Hayden, 2004) and perceived relationship quality (Bauer & Green, 1996). In addition, we controlled for Ó 2016 Hogrefe Publishing

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

Hypothesis 2 predicted that employees with a more incremental mindset will have higher quality relationships with their manager. The significant regression coefficient associated with employee mindset when entered in the regression model predicting the quality of their relationship with their manager. As shown in Table 2, employees’ mindset is not significantly related to the quality of their relationship with their manager (β = .02, ns). Therefore, the results did not support Hypothesis 2. Hypothesis 3 stated that a manager’s orientation toward learning goals will moderate the relationship between employees’ mindset and the quality of their relationship with their manager, such that the relationship is more positive when a manager has a strong learning goal orientation than when the manager’s orientation is weak. As shown in Table 2, the significant regression coefficient associated with the interaction term when entered in the regression model predicting the quality of their relationship with their manager (β = .34, p < .01). Next, we calculated the predicted values for relationship quality for managers had a strong orientation toward learning goals (1 standard deviation above the mean) and for managers had a weak orientation toward learning goals (1 standard deviation below the mean). As shown in Figure 2, employees’ mindset is more positively related to relationship quality when their manager has a strong orientation toward learning goals than when their manager has a weak orientation toward learning goals. In fact, we found that employees’ mindset is negatively related to the manager-employee relationship quality when the manager has a weak learning goal orientation. Thus, we found support for Hypothesis 3. Hypothesis 4 predicted that manager-employee relationship quality mediates the relationship between employees’ mindset and their job performance. To test Hypothesis 4, we conducted a Sobel test, which examines the joint significance of the two effects comprising the mediational effect. The results of this test fail to support Hypothesis 4 as the indirect effect of employee mindset on job performance was not significantly different from zero (z = 0.17, ns).

manager-employee relationship length, the number of years that each target employee has worked for his or her current manager, because the length of a relationship has been shown to influence relationship quality (Maslyn & Uhl-Bien, 2001) and because congruence in goal orientation may occur over time (Kristof-Brown & Stevens, 2001). Lastly, we controlled for the employee’s learning goal orientation because we were interested in isolating the effect of employee mindset. We measured employees’ learning goal orientation by asking each sales employee to respond to Sujan et al.’s (1994) 6-item measure of learning goal orientation (α = .85).

Results Table 1 includes descriptive statistics for and correlations between all study variables. Table 2 includes the standardized coefficients from the regression models used to test our study’s hypotheses. Before proceeding, we tested and found that we had not violated the assumptions of regression, such as normality and the absence of multicollinearity. To test our hypotheses, we used the moderated path analysis framework developed by Edwards and Lambert (2007). First, we regressed the control variables, employees’ mindset, their manager’s learning goal orientation, and the interaction term on the manager-employee relationship quality. Second, we regressed the control variables, employees’ mindset, and the manager-employee relationship quality on employees’ job performance. Hypothesis 1 predicted that employees who have a more incremental mindset will have higher job performance. In regard to Hypothesis 1, the significant regression coefficient associated with employee mindset when entered in the regression model predicting their job performance. As shown in Table 2, employees with a more incremental mindset have significantly higher job performance (β = .45, p < .001). The results supported Hypothesis 1. Table 1. Means, standard deviations, and correlations among variables Variable

1. Employee job performance

3.58

0.90

2. Relationship quality

3.45

0.96

3. Employee sex

1.31

0.47

0.07

4. Relationship length

2.12

1.51

0.14

0.04

5. Employee learning goal orientation

4.14

0.63

0.08

0.26*

0.00

0.06

6. Employee mindset

3.38

1.10

0.42***

0.04

0.13

0.10

0.19

7. Manager learning goal orientation

3.77

0.87

0.13

0.09

0.03

0.00

0.09

0.17 0.07 0.02

0.01

Notes. N = 96. Employee sex is male = 1 and female = 2; relationship length is how many years the employee has worked for the manager. *p < .05, ***p < .001.

Ó 2016 Hogrefe Publishing

Journal of Personnel Psychology (2017), 16(1), 36–45

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

Table 2. Regression analyses predicting relationship quality and job performance Relationship quality β

Variable

Job performance t

Step 1 Employee sex

0.07

0.65

0.06

0.61

Relationship length

0.03

0.27

0.14

1.36

Employee learning goal orientation

0.26

2.56*

0.09

0.87

ΔR

0.07

0.03

Step 2 Employee sex

0.07

0.65

0.12

1.28

Relationship length

0.03

0.29

0.18

1.93

Employee learning goal orientation

0.25

2.45*

0.01

0.05

Employee mindset

0.02

0.17

0.45

4.76***

Manager learning goal orientation

0.07

0.67

ΔR2

0.01

0.19***

Step 3 Employee sex

0.09

0.91

0.13

1.43

Relationship length

0.00

0.01

0.19

2.02*

Employee learning goal orientation

0.23

2.33*

0.04

0.44

Employee mindset

0.09

0.92

0.46

4.87***

Manager learning goal orientation

0.11

1.12 0.18

1.91

0.34

3.41**

Relationship quality Employee mindset Manager learning goal orientation ΔR2

0.12** 2

Notes. N = 96. Total R = .18 and R

adj

0.03 2

= .13 for relationship quality; total R = .25 and R

Hypothesis 5 predicted a moderated mediation effect with a first stage moderation. Specifically, we predicted that the learning goal orientation of the manager moderates the mediated effect (via manager-employee relationship quality) of employees’ mindset on their job performance. Using the moderated path analysis framework developed by Edwards and Lambert (2007), we applied bootstrapping procedures to generate 1,000 random samples with replacement from the full sample and used the bootstrapped estimates to constructed bias-corrected confidence intervals. Support would be indicated by a significant difference between the indirect effect when a manager has a weak orientation toward learning goals and the indirect effect when a manager has a strong orientation. As shown in Figure 3, when a manager had a weak learning goal orientation, the indirect effect was .07 and the 95% confidence interval excluded zero [ .16, .01], thus supporting an indirect effect. When a manager had a strong learning goal orientation, the indirect effect was .04 but the 95% confidence interval included zero [.00, .11], thus failing to find support for an indirect effect. Finally, the difference in indirect effect size between strong and weak levels of manager’s learning goal orientation was .10 and the 95% confidence interval also excluded zero [.01, .24] showing the indirect effect of employees’ mindset on job Journal of Personnel Psychology (2017), 16(1), 36–45

adj

= .21 for job performance. *p < .05, **p < .01, ***p < .001.

performance varies significantly with the learning goal orientation of the manager. Thus, we found support for Hypothesis 5.

Discussion The primary aim of this study was to examine whether and how employees’ mindset is related to their job performance and their relationship with their manager. We found that employees’ mindset had a direct relationship with job performance – employees with a more incremental mindset had higher job performance. Although employees’ mindset was not directly related to the quality of their relationship with their manager, we found that employees’ mindset interacted with their manager’s learning goal orientation to determine the manager-employee relationship quality. Consistent with our predictions, employees with more of an incremental mindset had a higher quality relationship with their manager when their manager had a strong learning goal orientation. Conversely, employees with a more incremental mindset had a lower quality relationship quality with their manager when their manager had a weak learning goal orientation. Finally, we found that, depending on their managers’ learning goal orientation, the manager-employee relationship quality mediated the Ó 2016 Hogrefe Publishing

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

Relationship quality

High Manager Learning Goal Orientation 4 Low Manager Learning Goal Orientation

mechanism by which managers form stronger LMX relationships with employees: managers may form stronger relationships with employees who have a mindset that is complementary to their own goal orientation. Thus, our results extend LMX theory by suggesting a possible situational factor that determines the quality of manageremployee linkages.

Practical Implications 2 1

Employees' Mindset Figure 2. Relationship quality and employees’ mindset for managers with strong and weak orientation toward learning goals.

relationship between employees’ mindset and their job performance. Namely, the mediated relationship was stronger when managers’ learning goal orientation was strong.

Theoretical Implications The results of this study have several theoretical implications. First, our findings provide further support and suggest refinements of Dweck and colleagues’ social-cognitive model of individual mindset (e.g., Dweck, 1999; Dweck & Leggett, 1988). We found that employee mindset does matter, as it directly and indirectly predicted job performance, thus we extended findings regarding individuals’ mindset on performance to work settings. Possibly due to their higher level of persistence and focus on learning, employees with a more incremental mindset performed higher than those employees with a more entity mindset. Second, by examining the role of manager’s learning goal orientation, we contribute to the person-supervisor fit literature. Specifically, we identified a managerial characteristic, their learning goal orientation, which, combined with employees’ mindset, appears to results in fit between the employee and his or her manager that influences their job performance. Future researchers could further explore the influence of fit between employee mindset and manager’s goal orientation by incorporating performance goal orientation as well as learning goal orientation. It is possible there could be some benefit to relationship quality when an employee has an entity mindset and their manager has a performance goal orientation. Lastly, the results of this study also have theoretical implications for research on relationships at work, more broadly, and on leader-member exchange relationships, more specifically. Namely, our results suggest a possible Ó 2016 Hogrefe Publishing

The results of this paper also have several practical implications. First, employees with a more incremental mindset had higher job performance, companies’ human resource department and hiring managers may want to assess the mindset of job candidates as part of the selection process. At least for sales positions, it appears that employees who have a more incremental mindset may be better equipped to deal with the inevitable setbacks and rejection that sales employees experience. Furthermore, human resource departments and managers may want to adjust their performance management process to recognize improvement overtime to help cultivate a learning culture. Second, because other research suggests that individuals’ mindset can be changed through interventions (e.g., Heslin et al., 2005), organizations may want to use such interventions as a way to increase employee performance. Third, our findings supporting the importance of employees’ mindset “fitting” with their manager’s goal orientation suggests that companies may wish to use this information when assigning employees to managers or other work roles (e.g., project teams).

Limitations and Future Research Directions This study does have some limitations. First, we collected data from sales employees; therefore, we do not know the extent to which our findings generalize to employees in other types of jobs. However, we contend that the sales setting of this study offered a relevant and conservative test of our hypothesis, as managerial relationships are less important in a sales setting because sales employees, at certain stages of their career, require less guidance from their manager (Kohli et al., 1998). Future research should replicate this study with nonsales employees to examine the generalizability of these results. Second, we measured manager’s goal orientation using a self-reported measure. It is possible that employees may not always perceive their manager’s goal orientation consistent with how the manager perceives his or her own goal orientation. Future research should examine how managers communicate their goal orientation and how consistently it is perceived by employees. Journal of Personnel Psychology (2017), 16(1), 36–45

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

Weak learning goal orientation of manager Employee mindset

-.40*

Manager-employee relationship quality

Indirect effect .17*

Employee job performance

-.07*

Employee job performance

.04

Figure 3. Analysis of the indirect effects for weak and strong levels of manager’s learning goal orientation (N = 96. *p < .05).

Strong learning goal orientation of manager Employee mindset

.22*

Manager-employee relationship quality

.17*

Difference in indirect effect = .10*

Third, our study examined one situational factor – manager’s learning goal orientation – that moderates the relationship between employees’ mindset and their job performance. Other situational factors (including other managerial characteristics) may also moderate this relationship. More empirical research is needed on whether other managerial characteristics combine with employees’ mindset to shape the quality of the manager-employee relationship. For example, managers’ leadership style, such as the extent to which they engage in transformational leadership behaviors, may be similarly beneficial when managing employees with more incremental mindsets. This may be due to transformational leaders trying to motivate their employees to exceed expectations (Judge & Piccolo, 2004), a situation employees with a more entity mindset might find threatening.

Conclusion The present study extends a vast literature examining mindset in nonwork settings and a budding literature examining mindset in work settings. Our findings show the value of examining employees’ mindset – and demonstrate the value of examining situational factors that serve as determinants of its effect. We hope this study encourages more research on how employees’ mindset relates to other organizationally relevant outcomes.

References Atwater, L., Brett, J., Waldman, D., DiMare, L., & Hayden, M. (2004). Men’s and women’s perceptions of the gender typing of management subroles. Sex Roles, 3, 191–199. doi: 10.1023/B: SERS.0000015551.78544.35 Basu, R., & Green, S. (1997). Leader-member exchange and transformational leadership: An empirical examination of innovative behaviors in leader-member dyads. Journal of Applied Social Psychology, 6, 477–499. doi: 10.1111/j.15591816 Bauer, T., & Green, S. (1996). Development of leader-member exchange: A longitudinal test. Academy of Management Journal, 6, 1538–1567. doi: 10.2307/257068

Journal of Personnel Psychology (2017), 16(1), 36–45

Chiu, C., Hong, Y., & Dweck, C. S. (1997). Lay dispositionism and implicit theories of personality. Journal of Personality and Social Psychology, 73, 19–30. doi: 10.1037/0022-3514.73.1.19 Cury, F., Elliot, A. J., Da Fonseca, D., & Moller, A. C. (2006). The social-cognitive model of achievement motivation and the 2 2 achievement goal framework. Journal of Personality and social Psychology, 90, 666–679. doi: 10.1037/0022-3514.90.4.666 Dweck, C. S. (1986). Motivational processes affecting learning. The American Psychologist, 10, 1040–1048. doi: 10.1037/0003066X.41.10.1040 Dweck, C. S. (1999). Self-theories: Their role in motivation, personality, and development. Philadelphia, PA: Psychology Press. Dweck, C. S., Chiu, C., & Hong, Y. (1995). Implicit theories and their role in judgments and reactions: A word from two perspectives. Psychological Inquiry, 4, 267–285. doi: 10.1207/ s15327965pli0604_1 Dweck, C. S., & Leggett, E. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 2, 256–273. doi: 10.1037/0033-295X.95.2.256 Edwards, J., & Lambert, L. (2007). Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis. Psychological Methods, 1, 1–22. doi: 10.1037/1082-989X.12.1.1 Godshalk, V., & Sosik, J. (2003). Aiming for career success: The role of learning goal orientation in mentoring relationships. Journal of Vocational Behavior, 3, 417–437. doi: 10.1016/ S0001-8791(02)00038-6 Graen, G., Novak, M. A., & Sommerkamp, P. (1982). The effects of leader-member exchange and job design on productivity and satisfaction: Testing a dual attachment model. Organizational Behavior and Human Performance, 30, 109–131. doi: 10.1016/ 0030-5073(82)90236-7 Graen, G., & Uhl-Bien, M. (1995). Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. Leadership Quarterly, 2, 219–247. doi: 10.1016/1048-9843(95)90036-5 Henderson, V., & Dweck, C. S. (1990). Motivation and achievement. In S. Feldman & G. Elliot (Eds.), At the threshold: The developing adolescent (pp. 309–329). Cambridge, MA: Harvard University Press. Heslin, P. A., Latham, G., & VandeWalle, D. (2005). The effect of implicit person theory on performance appraisals. Journal of Applied Psychology, 5, 842–856. doi: 10.1037/0021-9010. 90.5.842 Heslin, P. A., & VandeWalle, D. (2011). Performance appraisal procedural justice: The role of a manager’s implicit person theory. Journal of Management, 6, 1694–1718. doi: 10.1177/ 0149206309342895 Heslin, P. A., VandeWalle, D., & Latham, G. (2006). Keen to help? Managers’ implicit person theories and their subsequent

Ó 2016 Hogrefe Publishing

M. Zingoni & C. M. Corey, Effects of Employees’ Mindsets on Job Performance

employee coaching. Personnel Psychology, 4, 871–902. doi: 10.1111/j.1744-6570.2006.00057.x Heyman, G., & Dweck, C. S. (1998). Children’s thinking about traits: Implications for judgments of the self and others. Child Development, 2, 391–403. doi: 10.1111/j.1467-8624.1998. tb06197.x Hong, Y. Y., Chiu, C. Y., Dweck, C. S., Lin, D. M. S., & Wan, W. (1999). Implicit theories, attributions, and coping: A meaning system approach. Journal of Personality and Social psychology, 3, 588–599. doi: 10.1037/0022-3514.77.3.588 Janssen, O., & Van Yperen, N. (2004). Employees’ goal orientations, the quality of leader-member exchange, and the outcomes of job performance and job satisfaction. The Academy of Management Journal, 3, 368–384. doi: 10.2307/ 20159587 Judge, T., & Piccolo, R. (2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 5, 755–768. Knee, C. R. (1998). Implicit theories of relationships: Assessment and prediction of romantic relationship initiation, coping, and longevity. Journal of Personality and Social Psychology, 2, 360– 370. doi: 10.1037/0022-3514.74.2.360 Knee, C. R., Nanayakkara, A., Vietor, N. A., Neighbors, C., & Patrick, H. (2001). Implicit theories of relationships: Who cares if romantic partners are less than ideal? Personality and Social Psychology Bulletin, 7, 808–819. doi: 10.1177/ 0146167201277004 Kohli, A., Shervani, T., & Challagalla, G. (1998). Learning and performance orientation of salespeople: The role of supervisors. Journal of Marketing Research, 2, 263–274. doi: 10.2307/ 3151853 Kristof-Brown, A., & Stevens, C. (2001). Goal congruence in project teams: Does the fit between members’ personal mastery and performance goals matter? Journal of Applied Psychology, 6, 1083–1095. doi: 10.1037/0021-9010.86.6.1083 Levy, S., Stroessner, S., & Dweck, C. S. (1998). Stereotype formation and endorsement: The role of implicit theories. Journal of Personality and Social Psychology, 6, 1421–1436. doi: 10.1037/0022-3514.74.6.1421 Maslyn, J., & Uhl-Bien, M. (2001). Leader-member exchange and its dimensions: Effects of self-effort and other’s effort on

Ó 2016 Hogrefe Publishing

relationship quality. Journal of Applied Psychology, 4, 697–708. doi: 10.1037/0021-9010.86.4.697 Mueller, C., & Dweck, C. S. (1998). Praise for intelligence can undermine children’s motivation and performance. Journal of Personality and Social Psychology, 1, 33–52. doi: 10.1037/00223514.75.1.33 Ommundsen, Y. (2001). Self-handicapping strategies in physical education classes: The influence of implicit theories of the nature of ability and achievement goal orientations. Psychology of Sport and Exercise, 2, 139–156. doi: 10.1016/S1469-0292 (00)00019-4 Sujan, H., Weitz, B., & Kumar, N. (1994). Learning orientation, working smart, and effective selling. The Journal of Marketing, 3, 39–52. doi: 10.2307/1252309 Tabernero, C., & Wood, R. E. (1999). Implicit theories versus the social construal of ability in self-regulation and performance on a complex task. Organizational Behavior and Human Decision Processes, 2, 104–127. doi: 10.1006/obhd.1999.2829 VandeWalle, D. (1997). Development and validation of a work domain goal orientation instrument. Educational and Psychological Measurement, 57, 995–1015. doi: 10.1177/ 0013164497057006009 Williams, L., & Anderson, S. (1991). Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 3, 601–617. doi: 10.1177/014920639101700305 Received September 28, 2015 Revision received May 11, 2016 Accepted May 19, 2016 Published online September 22, 2016 Matt Zingoni Department of Management & Marketing University of New Orleans Kirschman Hall, Room 351 2000 Lakeshore Drive New Orleans, LA 70148 USA mzingoni@uno.edu

Journal of Personnel Psychology (2017), 16(1), 36–45

Original Article

Work–Home Interface and Well-Being A Cross-Lagged Analysis Audrey Babic1, Florence Stinglhamber2, Françoise Bertrand3, and Isabelle Hansez1 1

Human Resources Development Unit, Work Psychology Department, University of Liège, Belgium

Psychological Sciences Research Institute (IPSY), Université Catholique de Louvain, Louvain-la-Neuve, Belgium Belgian Defense, DGHR, Department of Recruitment and Selection, Research & Development, Brussels, Belgium

Abstract: Much effort has been expended in the past decade to examine the causal relationship between work–family conflict (WFC) and negative indicators of well-being. Comparatively little is known about the effects of work–family enrichment (WFE) on well-being. Even more importantly, very few studies have examined the concomitant effects of both WFC and WFE in terms of well-being. This study aims to fill these gaps by investigating the directionality of the causal relationships between WFC, WFE, and two well-being variables (i.e., job strain and job engagement). We examined these relationships using a two-wave cross-lagged panel design. Our sample was composed of 978 workers from a Belgian Federal Public Service. Reciprocal relationships were found between WFC–job strain, WFC–job engagement, and WFE–job engagement. Keywords: work–family conflict, work–family enrichment, job strain, job engagement, cross-lagged

Today, it is crucial for employees to balance the demands of their professional and private lives. Work–family issues are important research targets for understanding employees’ well-being. Indeed, many empirical studies have shown that employees perceiving an important work–family conflict (WFC) are more likely to experience lower wellbeing at work (Amstad, Meier, Fasel, Elfering, & Semmer, 2011). While most research on WFC has been based until recently on cross-sectional studies leaving the direction of causality uncertain (Casper, Eby, Bordeaux, Lockwood, & Lambert, 2007), much effort has been made in the past decade to examine the causal relationship between WFC and well-being. Using panel design, Matthews, Wayne, and Ford (2014) showed that, whereas WFC predicts employees’ subjective well-being, the reverse causation must also be taken into consideration. In the same vein, Nohe, Meier, Sonntag, and Michel (2015) conducted a meta-analysis on 32 studies using panel designs which shows reciprocal effects between WFC and work-specific strain. While the impact of WFC on negative indicators of wellbeing is thus well documented, little is known about the effects of another aspect of work–family interface whereby experience or participation in one role increases the individual’s performance or functioning in another, that is, work– family enrichment (WFE; Peeters, ten Brummelhuis, & van Steenbergen, 2013). Even more importantly, very few Journal of Personnel Psychology (2017), 16(1), 46–55 DOI: 10.1027/1866-5888/a000172

studies have examined the concomitant effects of both WFC and WFE in terms of well-being (Peeters et al., 2013). Conflict and enrichment are not opposites of each other but coexist (Grzywacz & Butler, 2005). One may experience more conflict or enrichment at certain times in life, but both are always present (e.g., Rantanen, Kinnunen, Mauno, & Tement, 2013). Therefore, focusing only on one of the two sides of the work–family interface may limit the understanding of the processes influencing work–family issues and their consequences (Boz, Martínez-Corts, & Munduate, 2016), notably in terms of employees’ well-being at work. Further, as suggested by Peeters et al. (2013), a deeper analysis of the impact of WFC/WFE on well-being, or vice versa, is needed to enable organizations to design appropriate interventions. Just as very little attention has been paid to the effects of WFE (in conjunction or not with WFC) on well-being, little is also known about the effects of WFC/WFE on positive indicators of well-being (Peeters et al., 2013). Yet, for organizations, it is important to know whether the levers of negative versus positive indicators of well-being are the same regarding different aspects of work–family interface. The present research was designed to fill these gaps in the literature. The aim of our study is to assess the directionality of the causal relationships between work–family interface (i.e., conflict and enrichment) and two well-being Ó 2016 Hogrefe Publishing

A. Babic et al., Work–Home Interface and Well-Being

variables (i.e., job strain and job engagement). In line with the recommendation of Tziner and Sharoni (2014), we investigated these relationships through a cross-lagged panel design (i.e., full panel design), which provides the strongest evidence of causal direction in field studies as compared to designs with simultaneous measurement of the variables (e.g., Little, 2013).

Theoretical Background and Hypotheses According to the role conflict theory (Goode, 1960), individuals are unable to satisfy the demands of all their social roles because of their limited time and energy resources. Conflict is expected to occur when too many demands are placed on an individual’s limited time and energy (Sieber, 1974). People may then perceive work– family conflict, generally defined as “a form of inter-role conflict in which the role pressures from the work and family domains are mutually incompatible in some respect” (Greenhaus & Beutell, 1985, p. 77). Being unable to manage work and family demands can result in two kinds of conflict. The first occurs when work-role demands impede the performance of family responsibilities (work-to-family conflict – WFC; Greenhaus & Beutell, 1985). The second occurs when family-role demands hinder work performance (family-to-work conflict; Greenhaus & Beutell, 1985). Work and family roles can also have positive effects on each other (Greenhaus & Parasuraman, 1999). According to the expansionist theory (Marks, 1977), having multiple roles results in greater access to resources which in turn can foster gains that facilitate performance or generate satisfaction in another role. Participation in several roles could thus benefit the individual by providing access to resources and experiences that contribute to individual fulfillment (Grzywacz & Butler, 2005). Work–family enrichment occurs when, thanks to participation in one role, an individual’s performance or functioning in the other role is enhanced (Greenhaus & Powell, 2006). Work–family enrichment is bi-directional, meaning that either work can provide gains that enhance the functioning of the family domain (work-to-family enrichment – WFE) or that family can provide gains that enhance the functioning of the work domain (family-to-work enrichment). Previous research suggests that conflict and enrichment may coexist and are not bipolar opposites but are rather conceptually distinct and orthogonal constructs (Grzywacz & Butler, 2005) having common and distinct determinants and consequences (e.g., Edwards & Rothbard, 2000; Grzywacz & Marks, 2000). While we recognize that work–family conflict/enrichment are bi-directional, the present research focuses only on the work-to-family direction in examining conflict and enrichment (i.e., WFC and Ó 2016 Hogrefe Publishing

WFE) because this direction is more likely to be influenced by an organization’s practices and policies (Friedman & Greenhaus, 2000).

Work–Family Conflict and Well-Being Work–family literature has consistently shown that high levels of WFC are associated with negative indicators of well-being such as more job strain (i.e., psychological hardship resulting from the job, when a worker considers he/she has not the necessary resources to face the demands he/she is confronted with; Hansez, 2008). According to the Conservation of Resources theory (COR theory; Hobfoll, 1989), people are motivated to acquire, preserve, protect, and expand their resources. When they perceive a (threat of) loss of resources, and when no action is taken (no coping behaviors used), a loss spiral appears, in which more and more resources are depleted and lost. Due to this loss or consumption of resources, resulting from the process of juggling work and family roles, high levels of WFC may lead workers to report more job strain. While WFC may thus predict employees’ job strain, the reverse causation can also be hypothesized. Being exposed to strain in a given domain (e.g., work) may lead to tension, irritability, fatigue, or preoccupation with problems (Greenhaus & Beutell, 1985). This negative state, by spilling over (Pleck, 1977), may affect an individual’s ability to perform in another domain (e.g., family), leading him/her to perceive more WFC. In their recent meta-analysis focusing on 32 studies based on cross-lagged panel designs, Nohe et al. (2015) provided empirical support for these positive reciprocal effects between WFC and work-specific strain (e.g., disengagement, emotional exhaustion, irritation, need for recovery, and personal accomplishment). Traditional work–family predictions suggest that, above and beyond its relationships with negative indicators of well-being, WFC is also associated with lower levels on positive indicators of well-being like job engagement (i.e., a state of positive emotional attachment and motivation toward one’s work; Hallberg & Schaufeli, 2006). In order to protect their remaining resources in situations of resources’ depletion (Hobfoll, 1989), workers perceiving WFC are more likely to reduce their level of job engagement. According to the source attribution perspective of WFC (Shockley & Singla, 2011), workers perceiving WFC psychologically attribute blame to the domain that was the source of the conflict (i.e., work) and are dissatisfied with this domain because of its responsibility in the conflict’s emergence. Faced with such situations of dissatisfaction, workers react/cope by adjusting their attitudes (e.g., reducing their engagement toward their job). Supporting this view, Opie and Henn’s cross-sectional study on 267 South African working mothers from several organizations Journal of Personnel Psychology (2017), 16(1), 46–55

(Opie & Henn, 2013) showed that, by experiencing conflict as a result of incompatible demands in their work and family lives, employees are less engaged in their work. In a cross-sectional study on a sample of 98 nurses from southern Poland, Wilczek-Ruzyczka, Basinska, and Dåderman (2012) reached the same conclusion. Conversely, however, Matthews et al. (2014) found, using a full panel design with three measurement times, that over time, WFC was associated with higher levels of subjective well-being (as a general concept). They relied on the adaptation theory (Diener, Lucas, & Scollon, 2006) to explain their unexpected results. According to this theory, employees who have to face a stressor (i.e., WFC) are likely to cognitively, emotionally, and behaviorally adapt to this situation over time, leading them to experience a corresponding increase in well-being. The reverse causation (i.e., the fact that job engagement predicts WFC) can also be considered. By being engaged in their job, workers have fewer resources available to dedicate to other domains (Greenhaus & Beutell, 1985), thus reducing their chance of meeting all roles’ expectations and leading them to perceive WFC. Accordingly, Halbesleben, Harvey, and Bolino (2009) found that job engagement was positively associated with WFC. However, even if they collected their data at multiple measurement times and from diverse samples, they did not assess all their variables at each time. Therefore, this design did not allow them to assess the causal relationship between variables. Furthermore, it is also possible that, by working enthusiastically and energetically (i.e., job engagement), workers gain new resources (gain spiral of resource – Hobfoll, 2002), reducing their perception of WFC. Hakanen, Perhoniemi, and Rodríguez-Sánchez (2012) found, in a cross-sectional study among Finnish judges, that job engagement was negatively related to WFC. In the same vein, Matthews et al. (2014) found that greater subjective well-being was associated with reduced WFC over time. By perceiving greater subjective well-being, workers enrich their package of resources, leading them to better cope with/react to a stressful situation (Hobfoll, 2002), and therefore reducing their perception of WFC. As we can see, results concerning the valence of the WFC-job engagement relationships are ambivalent. However, considering that the COR theory (Hobfoll, 1989) also explains the relationships between WFC and job strain (i.e., loss spiral of resources; see above), we coherently adopted the perspective that workers perceiving WFC experience job strain but also reduce their job engagement in order to preserve their remaining resources (Hobfoll, 1989). In the present research, we thus postulated a negative relationship between WFC and job engagement. Similarly, considering that the gain spiral of resources (Hobfoll, 2002) also explains the relationship between job Journal of Personnel Psychology (2017), 16(1), 46–55

A. Babic et al., Work–Home Interface and Well-Being

engagement and WFE (i.e., by being engaged in their work, a gain spiral of resources appears in which employees acquire more and more resources, thereby enriching their family domain; see below), we coherently adopted the perspective that job engagement reduces the perception of WFC and postulated a negative relationship between job engagement and WFC in this research. Therefore, we hypothesized that: Hypothesis 1: WFC and job strain have a reciprocal relationship such that, WFC at T1 will positively predict job strain at T2, and job strain at T1 will positively predict WFC at T2.

Hypothesis 2: WFC and job engagement have a reciprocal relationship such that, WFC at T1 will negatively predict job engagement at T2, and job engagement at T1 will negatively predict WFC at T2.

Work–Family Enrichment and Well-Being As mentioned earlier, little is known about the effects of WFE (Peeters et al., 2013). Regarding its potential predictive power on positive indicators of well-being, it is reasonable to expect that WFE is positively related to job engagement. According to the expansionist theory (Marks, 1977), occupying multiple roles has beneficial effects on physical and mental well-being due to greater access to resources. Indeed, WFE is the process whereby resources from the work domain lead individuals to develop their personal resources in the family domain, subsequently facilitating their performance in the latter (Greenhaus & Powell, 2006). The motivational process (Job DemandsResource model; Schaufeli & Bakker, 2004) posits that the presence of suitable job resources increases employees’ motivation, leading them to be engaged in their job. Individuals make cognitive attributions regarding the source of enrichment and realize that better functioning in their family is achieved due to psychological, intellectual, emotional, or other gains stemming from work (Sieber, 1974). Based on the social exchange theory (Blau, 1964) and the norm of reciprocity (Gouldner, 1960), when employees perceive that their organization provides something beneficial to them or their families, they are more likely to have positive affects and tend to reciprocate by demonstrating attitudes and behaviors toward work that are consistent with the perceived benefits they receive. Thus, it is reasonable to expect that when employees view their work as providing important benefits to them in their family role (i.e., WFE), they are more engaged in their job and make more effort. In line with this view, Wayne, Musisca, and Fleeson (2004) found that when people Ó 2016 Hogrefe Publishing

A. Babic et al., Work–Home Interface and Well-Being

experience WFE, they report putting greater efforts into their work and being more engaged in their job. However, we can also consider that perceiving job engagement increases the perception of WFE. By being engaged in their work, a gain spiral of resources may appear in which employees acquire more and more resources (i.e., skills, positive emotions, improved self-esteem; Hobfoll, 2002) therefore enriching their family domain. According to Edwards and Rothbard (2000), job engagement impacts the family domain through positive affective spillover. By being engaged in or feeling more absorbed by their job, individuals are likely to have more positive affects that then spill over into their family life, influencing affects at home and facilitating positive or beneficial interactions (i.e., WFE). In their two-wave study where variables were not measured at each measurement time, Siu et al. (2010) found that work engagement was positively related to WFE. In the same vein, Culbertson, Mills, and Fullagar (2012) collected repeated daily measures over a period of 10 working days and found that higher levels of job engagement were associated with greater WFE. Literature on the relationships between WFE and negative indicators of well-being is also scarce (Peeters et al., 2013). However, it is reasonable to assume that WFE is negatively related to job strain. According to the COR theory (Hobfoll, 2002), people with resources are less likely to see their well-being negatively influenced by stressful circumstances leading them to be less stressed by their job. Therefore, when they face a situation of stress, individuals with greater resources (i.e., perceiving WFE) are less likely to be affected by the drain of resources naturally resulting from stressful situations. Supporting this view, Kallaith (2014) found that WFE was negatively related with psychological strain in a sample of social workers. We might also argue that perceiving job strain decreases the perception of WFE. The negative emotions/feelings/ states generated by job strain (Greenhaus & Beutell, 1985) may spill over into the family domain (Pleck, 1977), leading to a negative state in this domain and therefore reducing the perception of WFE. But, to the best of our knowledge, we are not aware of any study that has considered job strain as a predictor of WFE. Therefore, we hypothesized that: Hypothesis 3: WFE and job engagement have a reciprocal relationship such that, WFE at T1 will positively predict job engagement at T2, and job engagement at T1 will positively predict WFE at T2. Hypothesis 4: WFE and job strain have a reciprocal relationship such that, WFE at T1 will negatively predict job strain at T2, and job strain at T1 will negatively predict WFE at T2. Ó 2016 Hogrefe Publishing

Method Sample and Procedure In order to test our hypotheses, a self-reported questionnaire was administered to employees from a Belgian Federal Public Service active in the field of security. The questionnaire was accompanied by an explanatory letter. Participation was voluntary and respondents completed the survey in group sessions. Our design comprised two measurement times with a 6-month interval between the assessments. This time lag of six months was chosen in agreement with the partner organization in order to be in accordance with their HR process requirements (i.e., a biannual evaluation). Moreover, this time lag is considered appropriate for longitudinal studies in work–family studies (Matthews et al., 2014). Indeed, WFC has previously been found to be related to well-being outcomes when the time lag is six months or less (e.g., Demerouti, Bakker, & Bulters, 2004). Paper-and-pencil questionnaires were filled in by 1,319 respondents at Time 1 (T1; response rate = 88.22%) and 1,002 at Time 2 (T2; response rate = 75.96%). In all, 978 respondents were matched between T1 and T2. This sample consisted of 886 men (90.59%) and 92 women (9.41%). Four hundred eighty-six (49.69%) were employees, 264 (27%) team leaders, and 228 (23.31%) executives. The totality of our sample was employed permanently by their organization and worked full-time. On average, participants were 20.92 years old (SD = 3.58). The majority of respondents had a high school diploma or less (87.66%), were single (91.21%), and had no children (94.79%). It should be noted that our sample was not representative of the general Belgian workforce in terms of age, sex, and marital status. For items with missing values, a Full Information Maximum Likelihood approach was used. We conducted logistic regressions in order to see if dropping out occurred according to socio-demographic variables (i.e., occupational and marital status, having children, gender, age, and level of education) or to the study variables (i.e., WFC, WFE, job strain, and job engagement). We did not find any significant selective dropout that would have biased our results.

Measures Work-to-family conflict and enrichment (WFC-WFE) were measured using the validated French version of the two ad hoc subscales of the Survey Work-Home Interaction – Nijmegen (Hansez, Etienne, & Geurts, 2006). The WFC subscale contains eight items (e.g., “I’m irritable at home because my work is demanding”). The WFE subscale contains five items (e.g., “I come home cheerfully after Journal of Personnel Psychology (2017), 16(1), 46–55

a successful day at work, positively affecting the atmosphere at home”). Respondents replied on a 4-point Likert-type scale (0 = never to 3 = always). Job strain and job engagement were measured, respectively, with the Negative Occupational State Inventory subscale (NOSI) and the Positive Occupational State Inventory subscale (POSI) developed by Barbier, Monseur, Bertrand, and Hansez (2012). The NOSI subscale comprises 11 items (e.g., “I feel demoralized by my work”). The POSI subscale comprises eight items (e.g., “I’m full of energy at work”). For each subscale, respondents replied on a 4-point Likert-type scale (1 = never to 4 = always).

Covariates Based on the semi-partial method (Little, 2013), we accounted for the influence of covariates by specifying paths from all covariates to all endogenous variables. After removing the nonsignificant effects, we pointed out that gender was significantly related to T2 WFE; and age to both T2 WFE and T2 job strain. The analyses were repeated without controlling for age and gender (Becker et al., 2016), and the results were essentially identical. In other words, including or excluding age and gender does not affect findings. However, as recommended by Becker et al. (2016), the inclusion of control variables has to be based on “relevant theory, or at least on sound reasoning and empirical evidence” (p. 162). Women were found to report a higher positive emotional reaction from the work-to-family direction than men, leading them to perceive more WFE (e.g., Grzywacz & Marks, 2000). Younger men were found to experience less WFE than older men, and younger women were found to experience more WFE than older women (Grzywacz & Marks, 2000). Finally, younger workers were found to experience more job strain than older ones (e.g., Tandon, Mahaur, & Gupta, 2014). Considering the significant relationships between these two socio-demographic variables, WFE and job strain, and the previous empirical evidence, we included age and gender as covariates in our analyses.

Data Analyses First, we performed a confirmatory factor analysis (CFA) with maximum likelihood estimation (Mplus 6, Muthén & Muthén, 1998–2010) to evaluate the fit of the hypothesized measurement model ensuring that the measures (i.e., WFE, WFC, job engagement, and job strain) were distinct latent constructs (Bentler & Bonett, 1980). Additionally, we also tested the measurement invariance (Little, 2013). The configural (equivalence between the pattern of fixed and free parameters), weak (equivalence between corresponding Journal of Personnel Psychology (2017), 16(1), 46–55

A. Babic et al., Work–Home Interface and Well-Being

factor loadings), and strong (equivalence between corresponding indicator means) invariance of our measurement model were tested over time. Second, we ran structural equation modeling (SEM) in order to test our hypotheses. Four different cross-lagged models were estimated: (1) a stability model where only the auto-regressions of WFE, WFC, job strain, and job engagement were estimated (Model 1); (2) a causal model where, in addition to the auto-regressions, four paths were added between WFE/WFC at T1 and job strain/ engagement at T2 (Model 2); (3) a reversed causal model where, in addition to the auto-regressions, four paths were specified between job strain/engagement at T1 and WFE/ WFC at T2 (Model 3); and finally (4) a reciprocal model where all paths from the two previous models were specified (Model 4). In these models, variances between constructs at T2 were allowed to covary and the error covariances of identical items over time were also allowed to correlate (Finkel, 1995). We reduced to three the number of indicators for each latent variable using a parceling strategy (balancing technique – Little, Cunningham, Shahar, & Widaman, 2002). We used parcels to limit the number of parameters to be estimated, to maintain the robustness of the analysis, and to preserve common construct variance while minimizing unrelated specific variance (Little et al., 2002).

Results In Table 1, descriptive statistics, reliabilities, and intercorrelations among all study variables are presented. As can be seen from the table, the reliabilities were acceptable. The internal consistencies of all constructs at each measurement time were satisfactory (α .77). The hypothesized measurement model showed a good fit with the data, w2(226) = 827.61, RMSEA = .05, CFI = .96, SRMR = .05. All items loaded reliably on their predicted factors, with standardized loadings ranging from .61 to .95. Concerning the measurement invariance (Table 2), the CFI differences tests between the three types of invariance were less than .01 (Little, 2013). The measurement of each scale was thus invariant over time. Results of the SEM analyses revealed that, in comparison with the stability model (Model 1), Models 2, 3, and 4 show a significant decrease in chi-square, indicating a better fit (Table 3). However, Model 4 displays the largest decrease in chi-square, Δw2(8) = 47.39, p < .01. In order to test further which of these three models represents the best depiction of the data, they were then compared to each other. Model 4, in comparison with both Model 2, Δw2(4) = 13.76, p < .01, and Model 3, Δw2(4) = 25.89, p < .001, showed a significant Ó 2016 Hogrefe Publishing

A. Babic et al., Work–Home Interface and Well-Being

Table 1. Descriptive statistics and intercorrelations among variables Variables

–

20.92

3.58

.04

–

T1 WFC

0.80

0.50

.04

.01

(.82)

T1 WFE

1.43

0.64

.06

.15***

.13***

T1 job strain

1.52

0.36

.06

.50***

.09

T1 JE

3.02

0.51

.03

.06

.23***

.37***

.23***

T2 WFC

0.81

0.52

.06

.00

.55***

.14***

.33***

.22***

(.84)

T2 WFE

1.42

0.64

.10**

.17***

.06

.49***

.02

.26***

.15***

Gender

Age

(.80) (.77) (.82) (.82)

T2 job strain

1.52

0.36

.06

.09**

.36***

.07*

.52***

.17***

.55***

.10

(.78)

T2 JE

2.88

0.54

.05

.06

.18***

.30***

.11***

.59***

.32***

.35***

.30***

(.85)

Notes. N = 978. Correlations among variables are provided below the diagonal and Cronbach’s alphas are provided on the diagonal. WFC = work-to-family conflict; WFE = work-to-family enrichment; JE = job engagement. Gender was coded 1 for male and 2 for female. *p < .05, **p < .01, ***p < .001.

Table 2. Measurement invariance Model

RMSEA

SRMR

CFI

ΔCFI

Comparison

Configural invariance

212

606.17

.04

.97

–

Weak invariance

220

614.36

.04

.97

1 vs. 2

.000

Strong invariance

228

649.49

.04

.97

1 vs. 3

.002

Notes. N = 978. The root-mean-square error of approximation (RMSEA), the standardized root-mean-square residual (SRMR), and the comparative fit index (CFI) were used in addition to the minimum fit function chi-square (w2) and the degrees of freedom (df). For the evaluation of the model fit, the following cutoff points were used: for the RMSEA, values of .06 or lower; for the SRMR, values of .08 or lower; and for the CFI, values .95 or higher (Hu & Bentler, 1999).

Table 3. Results of cross-lagged structural models Model

Model description

RMSEA

SRMR

CFI

AIC

Δw2(Δdf)

Comparison

Mstabil.

Stability model

269

782.42

.04

.05

.96

27,301.90

–

Mcausal

265

748.79

.04

.96

27,276.27

1 vs. 2

33.63(4)***

Mrevers.

265

760.92

.04

.05

.97

27,288.40

1 vs. 3

21.50(4)***

Mrecipr.

Causal model (Mstabil. + WFCT1/WFET1 ? job strain T2/job engagement T2) Reversed causation model (Mstabil. + job strain T1/job engagement T1 ? WFCT2/WFET2) Reciprocal model (Mcausal + Mrevers.)

261

735.03

.05

.96

27,267.07

1 vs. 4

47.39(8)***

2 vs. 4

13.76(4)**

3 vs. 4

25.89(4)***

Notes. N = 978. WFC = work-to-family conflict; WFE = work-to-family enrichment. The root-mean-square error of approximation (RMSEA), the standardized root-mean-square residual (SRMR), the Comparative Fit Index (CFI), and the Akaike Information Criterion (AIC) were used in addition to the Minimum Fit Function Chi-Square (w2) and the degrees of freedom (df). For the evaluation of the model fit, the following cut-off points were used: for the RMSEA, values of .06 or lower; for the SRMR, values of .08 or lower; and for the CFI, values .95 or higher (Hu & Bentler, 1999). **p < .01, ***p < .001.

decrease in chi-square. Consequently, Model 4 (i.e., the reciprocal model) was retained as the best fitting model. All lagged effects of Model 4 are presented in Table 4. In this reciprocal model (Model 4 – Figure 1), WFC T1 was positively related to job strain T2 and job strain T1 was positively related to WFC T2. These results support our Hypothesis 1. Results also show that WFC T1 was negatively related to job engagement T2 and job engagement T1 was negatively related to WFC T2; supporting our Hypothesis 2. WFE T1 was positively related to job engagement T2 and job engagement T1 was positively

Ó 2016 Hogrefe Publishing

Table 4. Lagged effects of the reciprocal model (Model 4) Lagged relationship WFC T1 ? job strain T2

Coefficient .130

.046

.005

Job strain T1 ? WFC T2

.098

.044

.026

WFC T1 ? job engagement T2

.084

.036

.014

Job engagement T1 ? WFC T2

.081

.031

.008

WFE T1 ? job strain T2

.010

.033

.765

Job strain T1 ? WFE T2

.026

.036

.471

WFE T1 ? job engagement T2

.172

.038

.000

Job engagement T1 ? WFE T2

.101

.038

.007

Note. N = 978. SE = standard error; WFC = work-to-family conflict; WFE = work-to-family enrichment.

Journal of Personnel Psychology (2017), 16(1), 46–55

A. Babic et al., Work–Home Interface and Well-Being

WFE T1

.53***

WFE T2 (R² = .35)

.17*** .10**

Job engagement T1

.51***

Job engagement T2 (R² = .11)

-.08** -.08*

WFC T1

.13**

WFC T2 (R² = .42)

.57***

.10*

Job strain T1

Figure 1. Reciprocal model (Model 4): standardized coefficients. WFC = workto-family conflict; WFE = work-tofamily enrichment; R2 = squared multiple correlation. For the sake of clarity, only significant relationships are shown. The reciprocal relationships between WFE and job strain being no significant, they were not shown. *p < .05, **p < .01, ***p < .001.

Job strain T2 (R² = .33)

.49***

Table 5. Comparison of strength of cross-lagged relationships Model description

RMSEA

SRMR

CFI

AIC

Comparison

Δw2(Δdf)

Reciprocal model

261

735.03

.05

.96

27,267.07

–

Constrained model (WFC and job strain)

262

733.81

.05

.96

27,267.29

1 vs. 2

1.22(1)

Constrained model (WFC and job engagement)

262

733.83

.05

.96

27,267.30

1 vs. 3

1.20(1)

Constrained model (WFE and job engagement)

262

734.75

.05

.96

27,267.23

1 vs. 4

0.28(1)

Notes. N = 978. WFC = work-to-family conflict; WFE = work-to-family enrichment. The root-mean-square error of approximation (RMSEA), the standardized root-mean-square residual (SRMR), the comparative fit index (CFI), and the akaike information criterion (AIC) were used in addition to the minimum fit function chi-square (w2) and the degrees of freedom (df). For the evaluation of the model fit, the following cut-off points were used: for the RMSEA, values of .06 or lower; for the SRMR, values of .08 or lower; and for the CFI, values .95 or higher (Hu & Bentler, 1999).

related to WFE T2; supporting our Hypothesis 3. Finally, we found no significant cross-lagged relationship between WFE and job strain. Therefore, our Hypothesis 4 was not supported. As shown in Table 5, by constraining the corresponding cross-lagged paths to be equal, the chi-square difference tests indicated that constrained models did not differ from the freely estimated model. These results indicated that any given causal relationship has the same strength as its reversed causal relationship.

Discussion Using a two-wave cross-lagged design, this research aimed to examine the relationships between WFC, WFE, job strain, and job engagement. The test of the direct, reversed, and reciprocal impacts of work–family interface on wellbeing filled the gap of knowledge due to the relative dearth of research in this area (Peeters et al., 2013). Concerning conflict, our study provides support for reciprocal effects between WFC-job strain (Hypothesis 1) and WFC-job engagement (Hypothesis 2). These results are in line with traditional work–family predictions. The amount of an individual’s resources is limited (Marks, 1977; Sieber, 1974). Perceiving conflict between work and family leads workers to lose/consume their finite Journal of Personnel Psychology (2017), 16(1), 46–55

resources in the process of juggling work and family roles. Over time, more and more resources are depleted/lost (Hobfoll, 1989), leading workers to perceive job strain. Conversely, when strained by their job, workers feel negative emotions, moods, or states which spill over and affect their performance and abilities with their families (Pleck, 1977), leading workers to perceive WFC. By perceiving WFC, workers engage less in their work in order to protect their limited personal resources (Hobfoll, 1989). They also psychologically attribute blame to the source of the conflict (i.e., work) and are dissatisfied with this domain (Shockley & Singla, 2011). To react to or cope with this situation of dissatisfaction, workers adapt their attitudes by reducing their engagement toward their job. Conversely, by being engaged in their work, workers gain more and more resources (Hobfoll, 2002), leading them to perceive less WFC. Concerning enrichment, our study provides support for reciprocal effects between WFE-job engagement (Hypothesis 3). By perceiving that work provides something beneficial for them (i.e., WFE), workers engage more in their job in order to reciprocate the received benefits (Blau, 1964; Gouldner, 1960). Conversely, the positive emotions and states resulting from job engagement spill over and positively impact the family domain (Edwards & Rothbard, 2000), leading workers to perceive WFE. Being engaged in work also leads employees to gain Ó 2016 Hogrefe Publishing

A. Babic et al., Work–Home Interface and Well-Being

resources (Hobfoll, 2002), enriching their family life and leading them to perceive WFE. While several studies have shown that WFE is related to positive indicators of well-being, empirical evidence has shown that WFE can also be related to negative indicators of well-being, such as depression or burnout (e.g., Innstranda, Langballe, Espnes, Falkum, & Aasland, 2008). According to COR theory, when not currently confronted with stressors, people strive to develop resource surpluses in order to offset the possibility of future loss; in other words, they enrich their resources pool by investing other resources (gain spiral of resources; Hobfoll, 2002). WFE can be considered as a resource surplus that might make the individual less vulnerable to resource loss. Indeed, this gain of resources might compensate for the burden of diverse or discrepant obligations (Sieber, 1974), reducing the perception of negative indicators of wellbeing. Accordingly, Innstranda et al. (2008) found negative reciprocal relationships between WFE and burnout. Concerning strain, Kallaith (2014) found that WFE was negatively related to psychological strain. More precisely, results showed that only two of the three dimensions of WFE (i.e., WFE-affect and WFE-capital) were significantly associated with reduced psychological strain. Therefore, the nonsignificant reciprocal effects found between WFE and job strain in the present study (Hypothesis 4) are perhaps due to the fact that we consider WFE as a global construct rather than a three-dimensional concept (i.e., development, affect, capital; see Carlson, Kacmar, Wayne, & Grzywacz, 2006). This issue should be investigated in future studies.

Limitations and Directions for Future Research The present study is not without limitations. First, although we included gender and age as control variables, a number of other factors could have influenced the investigated association, making it impossible to guarantee that the relationships were isolated from spurious influences (Bollen, 1989). For example, the presence of young children seems to influence the perception of work–family interference (e.g., Kossek & Ozeki, 1998). Second, the fact that our sample mainly comprised childless young men and was thus not representative of the general Belgian workforce may limit the generalization of findings. Thirdly, the use of self-reported data may lead to common-method bias (Podsakoff, MacKenzie, & Podsakoff, 2012). However, it is noteworthy that we used a longitudinal design with two measurement times to reduce the likelihood of this (Podsakoff et al., 2012). Finally, it should also be noted that levels of WFC and WFE at both times are rather low, which may have affected our results. Ó 2016 Hogrefe Publishing

As explained earlier, resources, and affects in particular, seem to play an important role in the relationship between work–family interface and well-being, as evidenced by the different theories detailed throughout this research (e.g., the conservation of resources theory with the gain/loss spiral of resources, Hobfoll, 1989, 2002; the spillover theory arguing an influence of work and family on each other in terms of moods, attitudes, emotions, feelings, etc., Pleck, 1977; positive affective spillover, Edwards & Rothbard, 2000). Therefore, to understand the underlying mechanism more fully, future research should investigate (in a longitudinal study) the role of resources or affects as mediators in the relationship between work– family interface and well-being.

Practical Implications Over time, WFE boosts workers’ job engagement; whereas WFC decreases their level of job engagement and increases their level of job strain. It is thus important that organizations allow employees to balance their work and family lives in order to decrease job strain and enhance job engagement. Organizations can achieve this through several measures supporting work and family. Employer work–family supports traditionally include three workplace characteristics that influence work–family relationships: (a) job conditions and structure of work, such as working hours and job designs that give workers control over when, where, or how they do their job; (b) organizational culture and norms about the hegemony of work and nonwork relationships; and (c) human resource policies such as “family-friendly policies” that support the juggling of work and family roles (e.g., flextime, telework; Kossek, 2006). Organizations can also raise workers’ awareness of how to better manage their work and family lives by allowing them to develop self-management skills. Indeed, these skills, important for developing work–family strategies, include, for example, establishing priorities, balancing work and family relationships, and maximizing the limited time available for outside work activities (Christensen, 1999). In the long term, being engaged in work boosts WFE and decreases WFC; whereas perceiving job strain increases WFC. This research thus highlights the importance of acting directly on job engagement. Improving various job resources (e.g., autonomy) and personal resources (e.g., self-efficacy) may be the most promising way to boost engagement (Bakker & Demerouti, 2008). More recently, Schaufeli and Salanova (2010) identify two levels of interventions to enhance job engagement: (1) individualbased interventions which refer to strategies focused on changing the individual’s behavior, beliefs, or goals and motives (e.g., identifying and developing one’s unique personal strengths, trying to achieve personal meaningful Journal of Personnel Psychology (2017), 16(1), 46–55

goals, increasing resilience); (2) organization-based interventions which refer to strategies focused on assessing and evaluating employees, designing and changing workplaces, work training, and career management (e.g., establishing and monitoring a fair psychological contract, (re)designing jobs, coaching and providing feedback, and social support). This research also highlights the importance of acting directly on job strain. One way that employers could reduce job strain is by implementing primary types of preventive actions, which attempt to reduce job strain by changing elements in the way work is organized and managed. These primary actions are recognized as the most relevant preventive actions to reduce sources of stress in the workplace (Bond, Flaxman, & Loivette, 2006). They include, for example, defining workers’ roles and responsibilities clearly, giving workers opportunities to participate in decisions/actions affecting their jobs or providing more support and opportunities for social interactions among workers.

References Amstad, F. T., Meier, L. L., Fasel, U., Elfering, A., & Semmer, N. K. (2011). A meta-analysis of work–family conflict and various outcomes with a special emphasis on cross-domain versus matching-domain relations. Journal of Occupational Health Psychology, 16, 151–169. doi: 10.1037/a0022170 Bakker, A. B., & Demerouti, E. (2008). Towards a model of work engagement. Career Development International, 13, 209–223. doi: 10.1108/13620430810870476 Barbier, M., Monseur, C., Bertrand, F., & Hansez, I. (2012). Measuring positive and negative occupational states at work: A structural and differential item functioning analysis. Psychologica Belgica, 52, 3–17. doi: 10.5334/pb-52-1-3 Becker, T. E., Atinc, G., Breaugh, J. A., Carlson, K. D., Edwards, J. R., & Spector, P. E. (2016). Statistical control in correlational studies: 10 essential recommendations for organizational researchers. Journal of Organizational Behavior, 37, 157–167. doi: 10.1002/job.2053 Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588–606. doi: 10.1037/00332909.88.3.588 Blau, P. M. (1964). Exchange and power in social life. New York, NY: Wiley. Bollen, K. A. (1989). Structural equations with latent variables. New York, NY: Wiley. Bond, F. W., Flaxman, P. E., & Loivette, S. (2006). A business case for the Management standards for stress. Norwich, UK: Health & Safety Executive. Boz, M., Martínez-Corts, I., & Munduate, L. (2016). Types of combined family-to-work conflict and enrichment and subjective health in Spain: A gender perspective. Sex Roles, 74, 136–153. doi: 10.1007/s11199-015-0461-5 Carlson, D. S., Kacmar, M. K., Wayne, J. H., & Grzywacz, J. G. (2006). Measuring the positive side of the work–family interface: Development and validation of a work–family enrichment scale. Journal of Vocational Behavior, 68, 131–164. doi: 10.1016/j.jvb.2005.02.002 Casper, W. J., Eby, L. T., Bordeaux, C., Lockwood, A., & Lambert, D. (2007). A review of research methods in IO/OB work–family

Journal of Personnel Psychology (2017), 16(1), 46–55

A. Babic et al., Work–Home Interface and Well-Being

research. Journal of Applied Psychology, 92, 28–43. doi: 10.1037/0021-9010.92.1.28 Christensen, P. M. (1999). Towards a comprehensive work/life strategy. In S. J. Parasuraman & J. H. Greenhaus (Eds.), Integrating work and family: Challenges and choices for a changing world. Westport, CT: Praeger. Culbertson, S. S., Mills, M. J., & Fullagar, C. J. (2012). Work engagement and work–family facilitation: Making homes happier through positive affective spillover. Human Relations, 65, 1155–1177. doi: 10.1177/0018726712440295 Demerouti, E., Bakker, A. B., & Bulters, A. (2004). The loss spiral of work pressure, work–home interference and exhaustion: Reciprocal relationships in a three-wave study. Journal of Vocational Behavior, 64, 131–149. doi: 10.1016/S0001-8791(03)00030-7 Diener, E., Lucas, R. E., & Scollon, C. N. (2006). Beyond the hedonic treadmill: Revising the adaptation theory of well-being. American Psychologist, 61, 305–314. doi: 10.1037/0003-066X.61.4.305 Edwards, J. R., & Rothbard, N. P. (2000). Mechanisms linking work and family: Clarifying the relationship between work and family constructs. Academy of Management Review, 25, 178–199. doi: 10.5465/AMR.2000.2791609 Finkel, S. E. (1995). Causal analysis with panel data. Thousand Oaks, CA: Sage. Friedman, S., & Greenhaus, J. (2000). Allies or enemies? What happens when business professionals confront life choices. New York, NY: Oxford University Press. Goode, W. J. (1960). A theory of role strain. American Sociological Review, 25, 483–496. doi: 10.2307/2092933 Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25, 161–178. doi: 10.2307/2092623 Greenhaus, J. H., & Beutell, N. J. (1985). Sources of conflict between work and family roles. Academy of Management Review, 10, 76–88. doi: 10.5465/AMR.1985.4277352 Greenhaus, J. H., & Parasuraman, S. (1999). Research on work, family, and gender: Current status and future directions. In G. Powell (Ed.), Handbook of gender and work. Thousand Oaks, CA: Sage. Greenhaus, J. H., & Powell, G. N. (2006). When work and family are allies: A theory of work–family enrichment. Academy of Management Review, 31, 72–92. doi: 10.5465/AMR.2006.19379625 Grzywacz, J. G., & Butler, A. B. (2005). The impact of job characteristics on work-to-family facilitation: Testing a theory and distinguishing a construct. Journal of Occupational Health Psychology, 10, 97–109. doi: 10.1037/1076-8998.10.2.97 Grzywacz, J. G., & Marks, N. F. (2000). Reconceptualizing the work–family interface: An ecological perspective on the correlates of positive and negative spillover between work and family. Journal of Occupational Health Psychology, 5, 111–126. doi: 10.1037/1076-8998.5.1.111 Hakanen, J. J., Perhoniemi, R., & Rodríguez-Sánchez, A. M. (2012). Demasiado Bueno para ser cierto? Similitudes y diferencias entre el engagement y la adicción al trabajo en jueces Finlandeses [Too good to be true? Similarities and differences between engagement and workaholism among Finnish judges] [Special issue on work engagement]. Ciencia y Trabajo, 14, 72–80. Retrieved from http://issuu.com/cienciaytrabajo/ docs/work_engagement/75 Halbesleben, J. R. B., Harvey, J., & Bolino, C. (2009). Too engaged? A conservation of resources view of the relationship between work engagement and work interference with family. Journal of Applied Psychology, 94, 1452–65. doi: 10.1037/a0017595 Hallberg, U. E., & Schaufeli, W. B. (2006). Same, same, but different? Can work engagement be empirically separated from job involvement and organizational commitment? European Psychologist, 11, 119–127. doi: 10.1027/1016-9040.11.2.119

Ó 2016 Hogrefe Publishing

A. Babic et al., Work–Home Interface and Well-Being

Hansez, I. (2008). The working conditions and control questionnaire: Towards a structural model of psychological stress. European Review of Applied Psychology, 58, 253–262. Hansez, I., Etienne, A. M., & Geurts, S. (2006). Le questionnaire d’Interaction Travail-Famille de Nijmegen : Résultats préliminaires et intérêt pour la clinique [The Nijmegen work-home interaction questionnaire: Preliminary results and clinical implications]. Revue Francophone de Clinique Comportementale et Cognitive, 11, 1–13. Retrieved from http://hdl.handle.net/ 2268/11292 Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44, 513–524. doi: 10.1037/0003-066X.44.3.513 Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of General Psychology, 6, 307–324. doi: 10.1037/1089-2680.6.4.307 Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. doi: 10.1080/10705519909540118 Innstranda, S. T., Langballe, E. M., Espnes, G. A., Falkum, E., & Aasland, O. G. (2008). Positive and negative work–family interaction and burnout: A longitudinal study of reciprocal relations. Work & Stress, 22, 1–15. doi: 10.1080/02678370801975842 Kallaith, P. (2014). Is work–family enrichment an antidote to experiences of psychological strain among Australian social workers? An empirical study. Australian Social Work, 67, 332–347. doi: 10.1080/0312407X.2013.825302 Kossek, E. E. (2006). Work and family in America: Growing tensions between employment policy and a changing workforce. A thirty-year perspective. In E. Lawler & J. O’Toole (Eds.), America at work: Choices and challenges. New York, NY: Palgrave MacMillan. Kossek, E. E., & Ozeki, C. (1998). Work–family conflict, policies, and the job-life satisfaction relationship: A review and directions for organizational behavior–human resources research. Journal of Applied Psychology, 83, 139–149. doi: 10.1037/0021-9010.83.2.139 Little, T. D. (2013). Longitudinal structural equation modeling. New York, NY: Guilford. Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling: A Multidisciplinary Journal, 9, 151–173. doi: 10.1207/ S15328007SEM0902_1 Marks, S. R. (1977). Multiple roles and role strain: Some notes on human energy, time and commitment. American Sociological Review, 42, 921–936. doi: 10.2307/2094577 Matthews, R. A., Wayne, J. H., & Ford, M. T. (2014). A work–family conflict/subjective well-being process model: A test of competing theories of longitudinal effects. Journal of Applied Psychology, 99, 1173–1187. doi: 10.1037/a0036674 Muthén, L. K., & Muthén, B. O. (1998–2010). Mplus User’s Guide (6th ed.). Los Angeles, CA: Muthén & Muthén. Nohe, C., Meier, L. L., Sonntag, K., & Michel, A. (2015). The chicken or the egg? A meta-analysis of panel studies of the relationship between work–family conflict and strain. Journal of Applied Psychology, 100, 522–536. doi: 10.1037/a0038012 Opie, T., & Henn, C. M. (2013). Work–family conflict and work engagement among mothers: Conscientiousness and neuroticism as moderators. SA Journal of Industrial Psychology, 39, 1–12. doi: 10.4102/sajip.v39i1.1082 Peeters, M., ten Brummelhuis, L. L., & van Steenbergen, E. F. (2013). Consequences of combining work and family roles: A closer look at cross-domain versus within-domain relations. In J. G. Grzywacz & E. Demerouti (Eds.), New frontiers in work

Ó 2016 Hogrefe Publishing

and family research (pp. 93–109). Hove, UK/New York, NY: Psychology Press. Pleck, J. H. (1977). The work–family role system. Social Problems, 24, 417–427. doi: 10.2307/800135 Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology, 63, 539–569. doi: 10.1146/annurev-psych-120710100452 Rantanen, J., Kinnunen, U., Mauno, S., & Tement, S. (2013). Patterns of conflict and enrichment in work–family balance: A three-dimensional typology. Work & Stress, 27, 141–163. doi: 10.1080/02678373.2013.791074 Schaufeli, W. B., & Bakker, A. B. (2004). Job demands, job resources, and their relationship with burnout and engagement: A multi-sample study. Journal of Organizational Behavior, 25, 293–315. doi: 10.1002/job.248 Schaufeli, W. B., & Salanova, M. (2010). How to improve work engagement? In S. Albrecht (Ed.), The handbook of employee engagement: Perspectives, issues, research and practice (pp. 399–415). Northampton, MA: Edwin Elgar. Shockley, K. M., & Singla, N. (2011). Reconsidering work–family interactions and satisfaction: A meta-analysis. Journal of Management, 37, 861–886. doi: 10.1177/0149206310394864 Sieber, S. D. (1974). Toward a theory of role accumulation. American Sociological Review, 39, 567–578. doi: 10.2307/ 2094422 Siu, O., Lu, J. F., Brough, P., Lu, C., Bakker, A. B., Kalliath, T., . . . Shi, K. (2010). Role resources and work–family enrichment: The role of work engagement. Journal of Vocational Behavior, 77, 470–480. doi: 10.1016/j.jvb.2010.06.007 Tandon, J. K., Mahaur, C., & Gupta, A. (2014). Effect of age and gender on occupational stress: A study on teaching fraternity. International Journal of Engineering Technology Management and Applied Sciences, 2, 41–46. Tziner, A., & Sharoni, G. (2014). Organizational citizenship behavior, organizational justice, job stress, and work–family conflict: Examination of their interrelationships with respondents from a non-Western culture. Journal of Work and Organizational Psychology, 30, 35–42. doi: 10.5093/tr2014a5 Wayne, J. H., Musisca, N., & Fleeson, W. (2004). Considering the role of personality in the work–family experience: Relationships of the big five to work–family conflict and facilitation. Journal of Vocational Behavior, 64, 108–130. doi: 10.1016/S0001-8791 (03)00035-6 Wilczek-Ruzyczka, E., Basinska, B. A., & Dåderman, A. (2012, April). How I manage home and work together: Occupational demands, engagement, and work–family conflict among nurses. Paper presented at the 10th European Academy of Occupational Health Psychology Conference, Zurich, Switzerland. Received June 7, 2015 Revision received April 22, 2016 Accepted May 18, 2016 Published online September 22, 2016 Audrey Babic Human Resources Development Unit Work Psychology Department University of Liège 2, Quartier Agora Place des Orateurs (Bât.32) 4000 Sart Tilman – Liège Belgium audrey.babic@ulg.ac.be

Journal of Personnel Psychology (2017), 16(1), 46–55

News and Announcements Changes Among Associate Editors Much to our regret, Despoina Xanthopolou has decided to step down from her role as Associate Editor at the Journal of Personnel Psycology in order take on different responsibilities. This definitely will be a great loss for us. Despoina served the journal for many years (since 2008 as a member of the Editorial Board, and since 2014 as Associate Editor) with her outstanding expertise, her engagement, and her warmth in interactions with authors, reviewers, and the rest of our team. She recently also served on our newly established award committee (see below). A big Thank You, Despoina, we will all miss you! At the same time, there is also good news on the editorial front, as Tanja Bipp and Jonas Lang have joined our team of Associate Editors. Both Tanja and Jonas were

previously members of our Editorial Board and have tremendously strong records in fields covered by the Journal of Personnel Psycology. Tanja Bipp is a Full Professor of Work and Organizational Psychology at the University of Würzburg, Germany. Her research interests and publications cover a broad range of topics, from personnel selection to work motivation and job design, to name only a few. Jonas Lang is an Associate Professor at the Department of Personnel Management, Work, and Organizational Psychology at Ghent University, Belgium. His research interests focus, among other things, on topics related to the study and assessment of abilities, personality, and emotions at work. We are very happy to have Tanja and Jonas in our editorial team, welcome on board!

Awards for Outstanding Achievements as Authors and Reviewers 2016 Starting with the 2016 volume, the Journal of Personnel Psychology has established two annual awards for outstanding achievements in different roles: the Best Paper Award and the Best Reviewers Award. We are delighted to announce the first award winners, chosen by a committee composed of members of the editorial team.

Best Paper Award 2016 The 2016 Best Paper Award goes to: Bilinska, P., Wegge, J., & Kliegel, M. (2016). Caring for the Elderly But Not for Ones Own Old Employees? Journal of Personnel Psychology, 15, 95 105. doi: 10.1027/18665888/a000144.

Journal of Personnel Psychology (2017), 16(1), 56 DOI: 10.1027/1866-5888/a000179

Best Reviewer Awards The Journal of Personnel Psychology distinguishes outstandingly helpful reviewers with its annual Best Reviewer Award. For 2016, these awards go to Heiko Breitsohl (University of Wuppertal, Germany) and Christian Vandenberghe (HEC Montréal, Canada). Congratulations to the award winners!

Ó 2017 Hogrefe Publishing

Instructions to Authors Aims and Scope of Journal of Personnel Psychology: The journal welcomes excellent empirical and theoretical contributions to basic and applied research in personnel psychology and related methodology. Reviews are also welcome, as are replications of previous research. Articles deal with all fields in personnel psychology, such as personnel selection, performance measurement, motivation, leadership, organizational commitment, personnel development and training, new test developments, and job analysis. As many topics in personnel psychology are closely related to issues in other branches of psychology or, more generally, the social sciences and human resource management, the journal is open to contributions of an interdisciplinary nature. Journal of Personnel Psychology publishes the following types of articles: Original Articles, Review Articles, Research Notes, Registered Reports, and Hybrid Registered Reports. Manuscript submisson: All manuscripts should be submitted electronically at http://www.editorialmanager.com/jppsy Detailed instructions to authors are provided at http://www. hogrefe.com/j/jpp Copyright Agreement: By submitting an article, the author confirms and guarantees on behalf of him-/herself and any coauthors that the manuscript has not been submitted or published elsewhere, and that he or she holds all copyright in and titles to the submitted contribution, including any figures, photographs, line drawings, plans, maps, sketches, and tables, and that the article and its contents do not infringe in any way on the rights of third parties. The author indemnifies and holds harmless the publisher from any third-party claims. The author agrees, upon acceptance of the article for publication, to transfer to the publisher the exclusive right to reproduce and distribute the article and its contents, both physically and in nonphysical, electronic, or other form, in the journal to which it has been submitted and in other independent publications, with no limitations on the number of copies or on the form or the extent of distribution. These rights are transferred for the duration of

copyright as defined by international law. Furthermore, the author transfers to the publisher the following exclusive rights to the article and its contents: 1. The rights to produce advance copies, reprints, or offprints of the article, in full or in part, to undertake or allow translations into other languages, to distribute other forms or modified versions of the article, and to produce and distribute summaries or abstracts. 2. The rights to microfilm and microfiche editions or similar, to the use of the article and its contents in videotext, teletext, and similar systems, to recordings or reproduction using other media, digital or analog, including electronic, magnetic, and optical media, and in multimedia form, as well as for public broadcasting in radio, television, or other forms of broadcast. 3. The rights to store the article and its content in machinereadable or electronic form on all media (such as computer disks, compact disks, magnetic tape), to store the article and its contents in online databases belonging to the publisher or third parties for viewing or downloading by third parties, and to present or reproduce the article or its contents on visual display screens, monitors, and similar devices, either directly or via data transmission. 4. The rights to reproduce and distribute the article and its contents by all other means, including photomechanical and similar processes (such as photocopying or facsimile), and as part of so-called document delivery services. 5. The right to transfer any or all rights mentioned in this agreement, as well as rights retained by the relevant copyright clearing centers, including royalty rights to third parties. Online Rights for Journal Articles: Guidelines on authors’ rights to archive electronic versions of their manuscripts online are given in the document ‘‘Guidelines on sharing and use of articles in Hogrefe journals’’ on the journal’s web page at www.hogrefe.com/ j/jpp

September 2016

25-26 April 2017 Hamburg Messe PERSONAL

09-10 May 2017 Messe Stuttgart

Exhibitions for human resource management

SAVE THE DATE

Leading exhibitions for human resource management with more than 8,800 visitors. www.personal-messe.de Get connected:

Main media partners

TOP Dark Triad of Personality at Work Dominik Schwarzinger / Heinz Schuler Einsatzbereich Erwachsene und Jugendliche ab 16 Jahren im erweiterten beruflichen Kontext (z. B. Bewerber, Berufstätige). Die TOP kann in Praxis und Forschung zur Personalauswahl und -entwicklung / Coaching sowie in Studien zu berufsbezogenen Effekten der „Dunklen Triade der Persönlichkeit“ eingesetzt werden.

Test komplett Bestehend aus: Manual, 10 Fragebogen Standardform, 10 Fragebogen Kurzform, 10 Auswertungs- und Profilbogen Standardform, 10 Auswertungs- und Profilbogen Kurzform, Schablonensatz und Box. Best.-Nr. 03 223 01 € 159,00 / CHF 195.00

www.hogrefe.com

Verfahren Die TOP erfasst für das Berufsleben relevante Aspekte der „Dunklen Triade der Persönlichkeit“ (Narzissmus, Machiavellismus und subklinische Psychopathie) auf drei Hauptfaktoren und elf Subskalen. Der Faktor Narzisstische Arbeitshaltung besteht aus den folgenden fünf Subskalen: Führungsanspruch, Überzeugungsglaube, Autoritätsbedürfnis, Risikofreude und Überlegenheitsgefühl. Der Faktor Machiavellistische Arbeitseinstellung setzt sich aus den drei Subskalen Unsentimentalität, Skepsis und Durchsetzungsglaube zusammen, der Faktor Psychopathischer Arbeitsstil aus den drei Subskalen Flexibilität, Impulsivität und Beschönigung. Die Items der TOP sind berufsbezogen formuliert, alle Entwicklungs- und Validierungsstudien wurden an Personen mit Berufserfahrung durchgeführt. Die TOP kann auch elektronisch im Hogrefe Testsystem 5 (HTS 5) durchgeführt und ausgewertet werden.

Zuverlässigkeit Die internen Konsistenzen der Faktoren liegen bei .81 ≤ α ≤ .94, die der Subskalen bei .65 ≤ α ≤ .93. Test-Retest-Reliabilitäten über ein halbes Jahr betragen für die Faktoren .69 ≤ rtt ≤ .76 bzw. .41 ≤ rtt ≤ .77 für die Subskalen. Gültigkeit Die Faktoren der TOP weisen hohe konvergente Zusammenhänge mit klassischen Standardverfahren zur Messung der Konstrukte Narzissmus, Machiavellismus und subklinische Psychopathie auf. Mit diversen Außenkriterien (z. B. NEO-FFI, HEXACO, RIASEC, kognitive Merkmale) zeigen sich hypothesenkonforme differenzielle Zusammenhänge. Befunde zu selbst- und fremdeingeschätztem Berufserfolg und zu objektiven Leistungs- und Erfolgsmaßen unterstützen die kriterienbezogene Validität. Normen Die Normstichprobe umfasst 1.298 Personen aus unterschiedlichen Berufsgruppen, Branchen und Arbeitszeitmodellen. Bearbeitungsdauer Standardform mit 60 Items: ca. 10 Minuten; Kurzform mit 9 Items: ca. 5 Minuten.

How to meet people’s increasing need to develop and manage their own lives and careers “A must read for all career development professionals and students alike. The Handbook of the Life Design is a new and essential resource for those working to improve career services in line with today’s challenges and conditions.” Sara Santilli, PhD, writing in Career Convergence, February 2015

Laura Nota / Jérôme Rossier (Editors)

Handbook of Life Design

From Practice to Theory and from Theory to Practice 2015, vi + 298 pp., hardcover US $54.00 / € 38.95 ISBN 978-0-88937-447-8 Also available as eBook Our lives and careers are becoming ever more unpredictable. The “lifedesign paradigm” described in detail in this ground-breaking handbook helps counselors and others meet people’s increasing need to develop and manage their own lives and careers. Life-design interventions, suited to a wide variety of cultural settings, help

www.hogrefe.com

individuals become actors in their own lives and careers by activating, stimulating, and developing their personal resources. This handbook first addresses life-design theory, then shows how to apply life designing to different age groups and with more at-risk people, and looks at how to train life-design counselors.