Leseprobe Swiss J. of Psychology 2018 by Hogrefe

Volume 76 / Number 1 / 2017

Swiss Journal of Psychology

Editor-in-Chief GrĂŠgoire Zimmermann Associate Editors Thierry Lecerf Nicolas Rothen Christian StaerklĂŠ

Mit ACT zu einem besseren Leben gelangen

Matthias Wengenroth

Das Leben annehmen So hilft die Akzeptanz- und Commitment-Therapie (ACT) Mit einem Vorwort von Thomas Heidenreich. 3., unveränd. Aufl. 2016. 308 S., 11 Abb., 4 Tab., Gb € 24,95 / CHF 32.50 ISBN 978-3-456-85683-4 Auch als eBook erhältlich

Mit dem Ansatz der Akzeptanz- und Commitment-Therapie (ACT) gelangen Sie zu einem zufriedeneren und besseren Leben: Dieser Ratgeber zeigt Ihnen anschaulich und unterhaltsam, wie ACT wirkt. Kämpfen Sie mit Gefühlen und Gedanken, die Ihnen das Leben schwer machen? Haben Sie schon vieles ausprobiert, um dagegen anzugehen, und sind dabei kaum weitergekommen? Und haben Sie das Gefühl, dass währenddessen das Leben an Ihnen vorbeizieht? Die diesem Buch zugrunde liegende Akzeptanz- und Commitment-Therapie

www.hogrefe.com

(ACT), die hier in leicht verständlicher und unterhaltsamer Weise vorgestellt wird, bietet neuartige und sehr erhellende Einblicke in die Ursachen menschlichen Leidens. Zudem zeigt ACT auf, wie wir besser mit den schwierigen Aspekten des Menschseins umgehen und gleichzeitig die eigenen Fähigkeiten und Stärken weiterentwickeln können. Dieses Buch hilft Ihnen dabei! Es zeigt Ihnen, wie Sie mithilfe einfacher, aber nachweislich wirksamer Methoden selbst die Voraussetzungen dafür schaffen können, vom Leben das zu bekommen, wonach Sie sich tief in Ihrem Inneren sehnen.

Swiss Journal of Psychology Official Publication of the Swiss Psychological Society

Volume 76, Issue 1, January 2017 Editor-in-Chief GrĂŠgoire Zimmermann, Lausanne

Editor-in-Chief

Grégoire Zimmermann, Lausanne

Associate Editors

Thierry Lecerf, Geneva Nicolas Rothen, Bern Christian Staerklé, Lausanne

Assistants to the Editors

Tamara Herz, Zurich Grégory Mantzouranis, Lausanne, gregory.mantzouranis@chuv.ch

Advisory Board

Fabrizio Butera, Lausanne Franz Caspar, Bern Paolo Ghisletta, Geneva Alexander Grob, Basel Claude-Alain Hauert, Geneva Ulrich Hoffrage, Lausanne Lutz Jäncke, Zurich Martin Kleinmann, Zurich Petra Klumb, Fribourg Klaus Opwis, Basel Sonja Perren, Konstanz Diego Pizzagalli, Cambridge, MA Pasqualina Perrig-Chiello, Bern Jérôme Rossier, Lausanne Willibald Ruch, Zurich Klaus Scherer, Geneva Franziska Tschan, Neuchâtel

Publisher

Hogrefe AG, Postfach, Länggass-Strasse 76, CH-3000 Bern 9, Tel. +41 31 300 45 00, Fax +41 31 300 45 93 verlag@hogrefe.ch, www.hogrefe.com

Support

The Swiss Journal of Psychology is supported by a contribution from the Swiss Academy of Humanities and Social Sciences

Media Manager

Josef Nietlispach, Hogrefe AG, Postfach, Länggass-Strasse 76, CH-3000 Bern 9, Tel. +41 31 300 45 00, Fax +41 31 300 45 91, inserate@hogrefe.ch

Typesetting

Satzspiegel, Nörten-Hardenberg, Germany

Printing Office

AZ Druck und Datentechnik GmbH, Kempten, Germany

ISSN

ISSN-L 1421-0185, 1421-0185 (Print), 1662-0879 (Online)

Frequency

4 issues per year

Subscriptions

Annual subscription rates: Libraries / Institutions CHF 336.– / e 261.–. Individuals CHF 127.– / e 94.– plus postage and handling. For members of the Swiss Psychological Society the subscription rate is included in their annual membership.* Single issue CHF 53.50 / e 40.– plus postage and handling

Indexing

Swiss Journal of Psychology is abstracted/indexed in Social Sciences Citation Index (SCIE), Social Scisearch, Current Contents/Social and Behavioral Sciences, Journal Citation Reports/Social Sciences Edition, PSYCLIT (Psychological Abstracts), PSYNDEX, PsycINFO, Europ. Reference List for the Humanities (ERIH), IBZ, IBR, and Scopus. Impact Factor: 0.696 2015 Journal Citation Reports® Social Sciences Edition (Thomson Reuters, 2016)

Electronic Full Data

www.econtent.hogrefe.com/toc/sjp/current *To become a member of the Swiss Psychological Society, please apply to the President, Sabine Sczesny, SGP/SSP, Institute für Psychologie, Universität Bern, Muesmattstrasse 45, CH-3000 Bern 9.

Swiss Journal of Psychology (2017), 76 (1)

Contents Original Communications

Short Research Note

Stéphanie Baggio, Victorin Luisier, and Cristina Vladescu: Relationships Between Social Networks and Mental Health: An Exponential Random Graph Model Approach among Romanian Adolescents

Eleni-Marina Ashikali, Helga Dittmar, and Susan Ayers: The Impact of Cosmetic Surgery Advertising on Swiss Women’s Body Image and Attitudes Toward Cosmetic Surgery

Marine Beaudoin and Olivier Desrichard: Memory Self-Efficacy and Memory Performance in Older Adults: The Mediating Role of Task Persistence

Chee-Seng Tan and Timothy Teo: Psychometric Qualities of the Creative Process Engagement Scale in a Malaysian Undergraduate Sample

Philip C. Mefoh and Valentine C. Ezeh: Effect of Cognitive Style on Prospective-Retrospective Memory Slips: Unipolar Approach

Swiss Journal of Psychology (2017), 76 (1)

NAB Neuropsychological Assessment Battery Franz Petermann / Lutz Jäncke / Hans-Christian Waldmann Unter Mitarbeit von Mona Bornschlegl Deutschsprachige Adaptation der Neuropsychological Assessment Battery (NAB) von Robert A. Stern und Travis White Test komplett Bestehend aus: 6 Modulen und Video-Tutorial-DVD* Best.-Nr. 03 203 01 € 1890,00 / CHF 2305.00 Modul Screening Best.-Nr. 03 203 02 € 484,00 / CHF 590.00 Modul Aufmerksamkeit Best.-Nr. 03 203 03 € 342,00 / CHF 417.00

Die NAB ist eine umfangreiche Testbatterie für die neuropsychologische Diagnostik. Sie findet primär in der klinischen Neuropsychologie bei Erwachsenen Verwendung. In der Rehabilitation von Schlaganfall-Patienten ist die NAB erprobt für den Einsatz in den Reha-Phasen B, C und D und kann auch bei spezifischen Fragestellungen in der klinischen Psychologie und Psychiatrie eingesetzt werden. Ebenso eignet sich die NAB als Messinstrument in der Fahreignungsdiagnostik, der Forschung zur kognitiven Entwicklung im Erwachsenenalter sowie bei gerontopsychologischen Fragestellungen.

Alle Aufgaben der deutschsprachigen NAB wurden gemeinsam auf Basis einer Stichprobe von N = 880 Erwachsenen standardisiert.

Zwei parallele Testformen der NAB ermöglichen Verlaufskontrollen (Eingangs- und Abschluss-Diagnostik).

Jedes der sechs NAB-Module kann unabhängig von den anderen Modulen angewendet werden.

www.hogrefe.com

Die NAB besteht aus sechs Modulen. Das Modul Screening erlaubt dem Untersucher zunächst eine effiziente Planung der tiefergehenden Diagnostik mit den Hauptmodulen der NAB. Die fünf Hauptmodule decken unterschiedliche Bereiche der psychologischen Diagnostik ab: • Aufmerksamkeit • Sprache • Gedächtnis • Wahrnehmung • Exekutive Funktionen

Modul Sprache Best.-Nr. 03 203 04 € 326,00 / CHF 398.00 Modul Wahrnehmung Best.-Nr. 03 203 05 € 330,00 / CHF 403.00 Modul Gedächtnis Best.-Nr. 03 203 06 € 384,00 / CHF 469.00 Modul Exekutive Funktionen Best.-Nr. 03 203 07 € 238,00 / CHF 290.00 Elektronische Auswertung Best.-Nr. H5 283 70 € 198,00 / CHF 240.00 Elektronische Auswertung Screening Best.-Nr. H5 283 80 € 78,00 / CHF 95.00 * Einführungsangebot (gültig bis September 2017): Bei Kauf eines Test komplett erhalten Sie die elektronischen Auswertungsprogramme kostenlos.

S. Baggio SwissJournal et al.: Social of Psychology Networks and (2017), © Mental 2017 76 (1), Hogrefe Health 5–11

Original Communication

Relationships Between Social Networks and Mental Health An Exponential Random Graph Model Approach among Romanian Adolescents Stéphanie Baggio1, Victorin Luisier2, and Cristina Vladescu3 1

Life Course and Social Inequality Research Centre, University of Lausanne, Switzerland Swiss National Centre of Competence in Research LIVES, University of Lausanne, Switzerland

2 3

Terre des Hommes, Bucharest, Romania

Abstract. Social networks have an important effect on health, and social network analysis has become essential for understanding human behavior and vulnerability. Using exponential random graph models (ERGM), this study explores the associations between mental health and network structure (or more specifically, mental health homophily) and the association between poor mental health and social isolation. Two classes of Romanian adolescents aged 12–14 years participated in the study (n = 26 in each class). We assessed school network, sociodemographic covariates, and mental health using the Strengths and Difficulties Questionnaire (SDQ). ERGM was first used to test the presence of sex and mental health homophily and then to test whether mental health was a predictor of social isolation. The results showed homophily patterns regarding sex and mental health. Moreover, participants with a higher SDQ score had a lower probability of a tie. Overall, this study showed how social networks are structured with different forms of homophily and that adolescents with poor mental health are more likely to be social isolates. Thus, prevention and interventions should focus on these vulnerable adolescents. Methodological advances like ERGM constitute a promising avenue for further research. Keywords: adolescents, ERGM, homophily, marginalization, mental health

Recent studies pointed out the important effect of social networks on health behaviors (Daw, Margolis, & Verdery, 2015), and health campaigns are increasingly using network interventions (Valente, 2012). Social-network analysis has thus become an essential tool for understanding social relationships and their association with physical and mental health (Greenblatt, Becerra, & Serafetinides, 1982; Schaefer & Simpkins, 2014). This study investigated how school-based networks of Romanian adolescents are associated with mental health and other sociodemographic characteristics, using an exponential random graph model (ERGM) approach. This was a first step in studying adolescent social networks with this recent and promising approach.

The Importance of Adolescents’ Social Networks Emancipation from one’s family is one of the biggest challenges of late childhood; adolescents spend less and less time with their parents and more and more time with their peers such as friends and classmates (Larson & Verma, 1999). In this context, adolescents have an increasing need for affiliation and social recogni© 2017 Hogrefe

tion outside the family (Gonet, 1994; Macdonald, 1989). Peer social networks constitute an important source of welfare (Prinstein & Dodge, 2008). Adolescents’ social networks are associated with health outcomes such as depression (Prinstein, 2007; Rubin, Bukowski, & Parker, 2006) and substance use (Daw et al., 2015; Hall & Valente, 2007; Jeon & Goodson, 2015) as well as with social outcomes (e.g., academic achievement, Lavy & Sand, 2012; aggressive behavior, Faris & Ennett, 2012). Because schools are the primary place of adolescents’ social interactions, the processes that occur in this context are likely to be generalizable to other social contexts (Haas, Schaefer, & Kornienko, 2010). More precisely, in this study, we used classbased networks, in which adolescents know each other and have the opportunity to interact with all of their classmates. Therefore, the network reveals affinities, and the absence of ties is not synonymous with a lack of opportunity to know each other.

Homophily in Social Networks An important pattern of relationships in social networks is homophily. It can be defined as a predominance of within-categoSwiss Journal of Psychology (2017), 76 (1), 5–11 DOI 10.1024/1421-0185/a000185

ry ties (McPherson, Smith-Lovin, & Cook, 2001). In other words, similar people are more likely to be friends than dissimilar people (McPherson et al., 2001). Homophily is a well-studied topic regarding sociodemographic covariates such as age, race, sex, and social status (Goodreau, Kitts, & Morris, 2009; McPherson et al., 2001). In addition, other studies reported that homophily is higher among numeric minorities, that is, when one group has a low prevalence rate compared to the other(s) (Goodreau et al., 2009). Studies that have investigated mental-health homophily are scarce. Regarding subjective well-being (SWB), one study found, for example, that users of online social networks with high SWB were more likely to be connected with users with a similar level of SWB (Bollen, Gonçalves, Ruan, & Mao, 2011). However, this study focused only on Twitter users, and the researchers coded SWB according to the valence (positive/negative) of the tweets posted on the social network, without directly measuring SWB. Another study found that depressed individuals were more likely to be friends with other marginalized individuals because of a withdrawal mechanism (Schaefer, Kornienko, & Fox, 2011). To our knowledge, no study has provided a more general picture of mental health using screening questionnaires to measure the children’s and adolescents’ overall mental health. This study overcame this gap by using a tool that is used worldwide, has robust psychometric properties, and was developed to measure mental health problems and the risks of mental health problems/psychiatric disorders among children and adolescents: the Strengths and Difficulties Questionnaire (SDQ; Goodman & Goodman, 2009; Goodman, Meltzer, & Bailey, 1998). The SDQ is an emotional and behavioral screening questionnaire used in child psychiatric research for screening, clinical assessment, and the evaluation of interventions. Therefore, it should be adapted to provide a more general overview of the relationship between social network and mental health.

Social Isolation and Mental Health Beyond homophily patterns, mental health is associated with social isolation. For example, losing or not having friends tends to increase the probability of depression (Rubin et al., 2006), and depressed people have more marginalized network positions (Schaefer et al., 2011). However, despite the important link between mental health and social network, this topic has only been investigated recently (Schaefer et al., 2011). Moreover, the most recent studies using social network analysis were conducted in the United States (Cornwell, 2009), using data from the National Longitudinal Study of Adolescent to Adult Health (“Add Health”; Goodreau et al., 2009; Schaefer et al., 2011). Studies conducted outside the United States, as well as studies using tools such as SDQ, are therefore needed. Swiss Journal of Psychology (2017), 76 (1), 5–11

S. Baggio et al.: Social Networks and Mental Health

The Exponential Random Graph Model Approach When it comes to the modeling of a social relationship between two people (tied, not tied) based on other variables and thirdparty relationships, traditional logistic regression does not apply. This is mainly due to the fact that the relationships in real networks are auto-correlated. These social relationship structures violate the assumption of independence of observations, and traditional analyses like logistic regression cannot be used. Exponential random graph model (ERGM) is a recent statistical framework developed to describe and understand the structure and features of complex social networks (Morris, Handcock, & Hunter, 2008). More precisely, ERGM allows statistical modeling, which can test the effect of individually measured variables on a specific social network of dichotomous ties, while also testing the effects of tie-interdependent structures that we consider important. Thus, the results are interpreted very similarly to those of logistic regression. Different dependencies related to the structure of the network are taken into account in ERGM. A first one is a dyadic dependence, that is, reciprocity: If Adolescent A names B as his friend, then Adolescent B is more likely to name A as his friend (I am friends with my friend). Another common network dependence is transitivity, which corresponds to the fact that friends of my friends are also my friends. Homophily is another form of dependence, whereby adolescents who share a common characteristic (e.g., sex, race) are more likely to name those sharing the same characteristic as friends. ERGM tests whether there is significantly less or more dependence according to these characteristics in the observed social network than expected by chance (Robins, Pattison, Kalish, & Lusher, 2007), or, in other words, whether the characteristics of the network members can predict the observed patterns of relationships (Harris, 2013). ERGM uses a particular kind of social network. First, the network needs to be clearly delimited inside an entity, such as a school. Therefore, ego-centered or snowball networks cannot be used with ERGM. Second, the ties are considered as random variables with a dichotomous outcome: the presence or absence of a tie for each possible dyad. The ERGM approach is relatively new in public health literature. It goes beyond the descriptive methods often used to examine social networks and is thus a very useful tool for social network analysis (Harris, 2013). Indeed, it allows one to represent social network structures and test associations with network members’ characteristics using a familiar technique (logistic regression form) to test micro-level processes. Other techniques such as multilevel modeling that take dependence into account do not allow us to test hypotheses related to the structure of the network. Thus, the goal of this study was to explore school-based net© 2017 Hogrefe

S. Baggio et al.: Social Networks and Mental Health

works of Romanian adolescents using the ERGM approach. We aimed to replicate previous results regarding sociodemographic homophily (i.e., gender homophily) and to explore mental health homophily. Moreover, we investigated whether adolescents with poor mental health were more likely to be social isolates by testing whether mental health was a predictor of the network structure.

Method Participants and Procedures The study took place in two Romanian classes (Class A, Class B) of the general school curriculum located near Bucharest in May-June 2011. It was part of a larger program called “Movement, Games, and Sports” (MGS) of the nongovernmental organization Terre des Hommes. This project provided training for teachers, and our entire study was designed to evaluate the program and the children’s welfare. A total of 52 adolescents aged 12 to 14 (M = 12.9 years) were interviewed, with n = 26 in each class. Participation in the study was voluntary and the parents gave consent for their child’s participation. Since it was the end of the school year, some of the adolescents were absent, n = 9 (35%) and n = 7 (27%), respectively. Adolescents who were absent were removed from the social networks. Analyses including missing adolescents were performed, without notable differences regarding the structure of social networks. There were no missing values for other variables. Participants completed the questionnaire during class, and there was a debriefing at the end of the study. Since the study was performed in collaboration with a nongovernmental organization, no ethics committee was asked to approve the study.

Measures Social Network Participants were asked to report the names of their friends in the class, with no maximum number. Thus, we focused on strong links (i.e., friendship).

Mental Health The Strengths and Difficulties Questionnaire (SQD; Goodman et al., 1998) was used to assess mental health. The SDQ is a brief emotional and behavioral screening questionnaire for child and adolescent mental health problems. It includes emotional symptoms (five items; e.g., “I am often unhappy, depressed or tearful”), conduct problems (five items; e.g., “I get very angry and often lose my temper”), hyperactivity/inattention (five items; e.g., © 2017 Hogrefe

“I am easily distracted, I find it difficult to concentrate”), and peer relationship problems (five items; e.g., “Other children or young people pick on me or bully me”). Altogether, the 20 items generate a total difficulties score ranging from 0 to 40. An English version, available at www.sdqinfo.com, was translated into Romanian, using back-translation and pretest. The SDQ was used as a continuous variable since it is a dimensional measure of child and adolescent mental health (Goodman & Goodman, 2009) and was dichotomized to create two groups, with lower and higher levels of mental health, respectively, in order to examine mental health homophily. To investigate patterns of homophily, distinct groups of participants are needed, and a continuous score did not allow for the separation of participants into subgroups. We used the mean score of the classes to define higher and lower levels of mental health. Indeed, there were not enough children with a score higher than the cut-off recommended for SDQ (= 20).

Sociodemographic Covariates Age and sex were recorded.

Statistical Analysis As a preliminary analysis, descriptive statistics for sociodemographic variables, mental health, and network information were computed separately for each class. Regarding network information, the number of edges (i.e., the number of ties within the social network), network density (i.e., the sum of the ties divided by the number of possible ties), reciprocity (i.e., the proportion of dyads that are symmetric), and transitivity (i.e., the proportion of triads) were computed. Next, a first set of ERGM was performed to test the presence of homophily. Two models were computed for each class including the following predictors: a) sex and b) SDQ dichotomized. These models also included reciprocity and transitivity between ties (i.e., the geometrically weighted edgewise shared partner [GWEPS] distribution parameter) which counts triangles. Two parameters were computed for each variable, for example, one parameter for girls, one for boys. This procedure allowed us to test different patterns of homophily among each subgroup to see whether homophily was higher for one group than the other. A parameter significantly different from zero means that the corresponding configuration occurs at a greater level (for positive parameters) than expected by chance. Parameters are computed using Markov chain Monte Carlo maximum likelihood estimation (MCMCMLE; Robins et al., 2007). Finally, a second set of ERGM tested whether mental health was a predictor of ties in the social network, which was a test of whether participants with poor mental health were more likely to be social isolates. One model was computed for each class, including SDQ on a continuous scale as a predictor Swiss Journal of Psychology (2017), 76 (1), 5–11

S. Baggio et al.: Social Networks and Mental Health

(MCMCMLE). This model also included reciprocity and transitivity. For all models, goodness-of-fit diagnostics were tested using AIC and BIC criteria, as compared to the AIC and BIC of the null models including only edges. The SDQ has a subscale measuring peer-related problems, which is related to social isolation. To confirm that there was no confounding effect, we ran a sensitivity analysis excluding this subscale from all models tested with the SDQ. The results were very similar (estimates, p-values, goodness-of-fit), so we kept the entire SDQ scale in the analyses. All analyses were carried out with R, version 3.2.0, using the package “statnet,” version 2015.11.0 (Handcock, Hunter, Butts, Goodreau, & Morris, 2008).

Results Preliminary Results Girls were the numeric majority in Class B (72%) and boys were the numeric majority in Class A (56%). Regarding mental health, the classes showed different patterns, with participants in Class A having a poorer level of mental health (e.g., MSDQ = 17.92) than those in Class B (e.g., MSDQ = 9.12). The results of the two samples were significantly different (t(50) = 6.26, p < .001).

Table 1. Descriptive statistics for sociodemographic variables, mental health, and social network Class A

Age1

12.56 (0.65)

13.20 (0.50)

Boys

56 (14)

28 (7)

Girls

44 (11)

72 (18)

Gender2

Social network information also differed by class, with Class B having a higher connectivity (number of ties = 173, density = 28.8) than Class A (number of ties = 82, density = 13.7). Both classes had high levels of reciprocity (67.0% and 82.7%) and moderate levels of transitivity (39.9% and 30.9%).

Patterns of Homophily The first panel of Table 2 presents the results regarding homophily patterns among the two social networks. Gender homophily (i.e., significant positive estimates) was found in both classes, with higher rates of homophily across the numeric minority. In Class A, girls were more likely to have ties with other girls (estimate = 2.24, p < .001) while boys’ homophily was significant but lower (estimate = 1.62, p < .001). In Class B, boys were more likely to have ties with other boys (estimate = 0.94, p < .001), while girls were not significantly homophilic (estimate = 0.26, p = .26). Figure 1 displays the social network showing the minority homophily in Class A. Within-category preferences are clearly visible. One-sided arrows mean that a participant named the other participant as a friend, but the other did not reciprocate (i.e., the second participant did not mention the first as a friend). Two-sided arrows mean that both participants mentioned the other as a friend, meaning that the friendship was reciprocal. Considering mental health, mental health homophily appeared to be found in both classes. Participants with better mental health (i.e., groups with lower SDQ scores) were more likely to have ties with similar participants (lower SDQ score in Class A: estimate = 1.26, p < .001; lower SDQ score in Class B: estimate = 0.81). Participants with poorer mental health (i.e., groups with higher SDQ scores) were also more likely to have ties with similar participants (higher SDQ score in Class A: estimate = 0.76, p < .001). Results were nonsignificant in Class B. Figure

SDQ (0–40) Mean score1

17.92 (5.05)

9.12 (5.08)

44 (11)

52 (13)

≥ mean score2

56 (14)

48 (12)

< mean score

Social network No. of edges

173

Density2

13.7

28.8

Reciprocity2

82.7

67.0

30.9

39.9

Transitivity

Note. SDQ = Strengths and Difficulties Questionnaire, whereby a higher score indicated a higher number of mental health problems; No. of edges = total number of ties within the network (number of children mentioned as friends by each participant); density = sum of the ties divided by the number of possible ties; reciprocity = proportion of dyads that are symmetric; transitivity = proportion of triads. 1Means and standard errors are given; 2Percentages and n are given.

Swiss Journal of Psychology (2017), 76 (1), 5–11

Figure 1. Gender homophily in the social network of Class A.

S. Baggio et al.: Social Networks and Mental Health

Table 2. Model coefficients for homophily ERGM and social isolation ERGM Class A Model 1

Class B Model 2

Model 1

Model 2

–2.07***

–2.04***

Homophily Edge

–3.74***

–3.29***

Sex Boys

1.62***

–

0.94**

–

Girls

2.24***

–

0.26

–

SDQ < mean score

–

1.26***

–

0.81***

≥ mean score

–

0.76***

–

0.04

Reciprocity

0.76*

1.05*

0.81**

0.75**

Transitivity

0.45

0.64**

0.51

0.49

706.1

696.8

728.1

718.8

AIC

400.2

430.6

BIC

422.1

452.6 Social isolation

Edge SDQ

1.84*** –0.03*

–3.35***

–1.14*

–2.51

–

–0.04**

–

Reciprocity

1.31***

1.34***

0.82**

0.87***

Transitivity

0.67***

0.72***

0.47

0.51

AIC

450.2

454.1

704.1

712.2

BIC

467.8

471.7

721.7

729.8

Note. SDQ = Strengths and Difficulties Questionnaire. Null models (with only edges): Class A: AIC = 480.6, BIC = 485.0; Class B: AIC = 722.8, BIC = 727.2. *p < .05, **p < .01, ***p < .001.

Figure 2. Mental health homophily (SDQ scale) in the social network of Class B. SDQ = Strengths and Difficulties Questionnaire.

Figure 3. SDQ scores in the social network of Class B. SDQ = Strengths and Difficulties Questionnaire. Larger circles indicate higher SDQ scores.

2 summarizes the results of the SDQ scores in Class B, showing that participants were homophilic regarding mental health. Overall, both classes showed reciprocity and transitivity, with a tendency for transitivity (.05 < ps < .10). Fit indices showed that most models were better than the null model in each class.

Mental Health and Social Isolation

The results of the ERGM investigation of social isolation using continuous scores of SDQ are presented in the second panel of Table 2. In both classes, participants with a higher SDQ score had Swiss Journal of Psychology (2017), 76 (1), 5–11

a lower probability of ties (Class A: estimate = –0.03, p < .05; Class B: estimate = –0.04, p < .01). Figure 3 represents the SDQ scores in the social network of Class B: The larger the circle representing a participant, the higher his or her SDQ score was. Overall, the figure shows that participants with low SDQ scores were more central in the social network: They were more likely to have ties with other participants. Both classes showed reciprocity (p < .01), and only Class A showed significant transitivity (p < .001) whereas transitivity was tendentially significant in Class B (.05 < p < .10). Fit indices showed that the models were better than the null model in each class, except for BIC of Model 2 in Class B, which was similar to the BIC of the null model.

Discussion This study was a first step in investigating school-based networks of Romanian adolescents using the ERGM approach, focusing on friendship ties, aiming to explore homophily patterns and social isolation in relation to mental health. Regarding patterns of homophily, the study appeared to replicate previous well-known findings of sociodemographic homophily (Goodreau et al., 2009; McPherson et al., 2001), such as gender homophily. Indeed, boys were more likely to be friends with others boys in both classes, whereas girls were more likely to be friends with other girls in Class B. Therefore, gender homophily for the numeric minority in the social network seemed to be supported, as reported by Goodreau et al. (2009). We also found mental health homophily in both social networks. In both classes, participants with better mental health were more likely to be friends with similar participants. In Class A, participants with poorer mental health were more likely to be friends with similar participants. This finding of mental health homophily was in line with the findings of previous studies that reported homophily with respect to mental health behaviors (Bollen et al., 2011; Schaefer et al., 2011). Thus, the predominance of within-category ties also seemed to apply to mental health characteristics. Regarding social isolation, the second set of ERGM showed that mental health was a predictor of network structure. Adolescents with poorer mental health were less likely to have friends within the social network. Adolescents with poor mental health appear to be more marginalized than participants with good mental health, as reported in a previous U. S. study (Schaefer & Simpkins, 2014). Thus, mental health and social network appear to be interconnected (Prinstein, 2007). Because of its cross-sectional design, this study did not reveal causal relationships between the two, but the results highlighted the importance of taking social network into account when studying health behaviors (Daw et al., 2015; Valente, 2012). Social isolation, marginalization, and social networks are probably linked in a reciprocal relationship (Haas et al., 2010), each influencing the other. Accordingly, adolescents Swiss Journal of Psychology (2017), 76 (1), 5–11

S. Baggio et al.: Social Networks and Mental Health

who are isolated and have poor mental health may become more and more isolated and have increasingly poor mental health. They should be a focus on prevention and early intervention because they are vulnerable adolescents. The relationship between social network and health is complex and dynamic (Haas et al., 2010). However, studies using larger sample sizes and more extended social networks are needed to confirm these preliminary results. Overall, the ERGM approach appears to be a useful tool for the study of social network structure and its relationships with both categorical and continuous individual measures. Most of the previous studies focused on categorical predictors, such as sociodemographic characteristics, and this study extended social network analysis to continuous predictors. It moved beyond network description and allowed for the study of microlevel processes through simple models using a logistic regression form (Goodreau et al., 2009; Harris, 2013) while taking the dependence between individuals into account. Indeed, ERGM took reciprocity and transitivity between adolescents into account, and, in most cases, both were significant, except for transitivity in Class B (tendency). This means that the ties in the social networks were structured with a dyadic dependence (reciprocity: I am friends with my friends) and triangles (transitivity: friends of my friends are my friends). This illustrates the principle of linked lives described by Settersten (2015). Indeed, people do not exist in isolation from others, and relationships themselves are interdependent. We should focus on the interdependence of lives when we study social and psychological processes. In addition, this confirms that analyses conducted on network structures should take their interdependent structure into account, which has long been a methodological challenge (Scott, 2012; Wasserman & Faust, 1994). This study had some limitations. A first one was its cross-sectional design, which made it impossible to identify causal paths between mental health and social network. Even if cross-sectional data had permitted us to study processes, more longitudinal network data would be needed. Second, the study involved a small sample of Romanian adolescents, in only two classes. Thus, the results of the study should be replicated in larger samples. Regarding modeling concerns, the small sample size did not appear to be an issue. Indeed, the sample size in ERGM models is not unambiguous as in conventional statistics because the dependent variable is the number of ties (edges), which is usually larger than the number of participants (n), which was the case in this study (Krivitsky & Kolaczyk, 2015). Moreover, studies in other contexts are needed because even though school is a primary place for adolescents’ social interactions (Haas et al., 2010), others environments are also meaningful. Third, school-based networks were not complete because of the missing participants (35% and 27%) and this may have resulted in a higher number of social isolates (Valente, 2012). Even if missing participants were not frequently mentioned, the social networks within the classes were not complete, and this may have affected the density of social networks. Finally, we used an arbitrary cut-off to create subgroups of better/poorer mental health, and other cut-offs should be tested, such as clinical cut-offs to define poor mental health. © 2017 Hogrefe

S. Baggio et al.: Social Networks and Mental Health

In conclusion, this preliminary study showed that the social networks seemed to be structured with different forms of homophily, including mental health homophily. Adolescents with poor mental health seemed more likely to be marginalized and social isolates, and prevention and interventions should focus on these vulnerable individuals. Methodological advances like ERGM permitted the exploration and understanding of such complex social processes, and they constitute a promising avenue for further research.

Acknowledgments We thank the NGO Terre des Hommes for providing the sample and support.

References Bollen, J., Gonçalves, B., Ruan, G., & Mao, H. (2011). Happiness is assortative in online social networks. Artificial Life, 17, 237–251. doi 10.1162/artl_a_00034 Cornwell, B. (2009). Good health and the bridging of structural roles. Social Networks, 31, 92–103. doi 10.1016/j.socnet.2008.10.005 Daw, J., Margolis, R., & Verdery, A. M. (2015). Siblings, friends, course-mates, club-mates: How adolescent health behavior homophily varies by race, class, gender, and health status. Social Science & Medicine, 125, 32–39. doi 10.1016/j.socscimed.2014. 02.047 Faris, R., & Ennett, S. (2012). Adolescent aggression: The role of peer group status motives, peer aggression, and group characteristics. Social Networks, 34, 371–378. doi 10.1016/j.socnet.2010.06.003 Gonet, M. M. (1994). Counseling the adolescent substance abuser: School-based intervention and prevention. Thousand Oaks, CA: Sage. Goodman, A., & Goodman, R. (2009). Strengths and Difficulties Questionnaire as a dimensional measure of child mental health. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 400–403. doi 10.1097/CHI.0b013e3181985068 Goodman, R., Meltzer, H., & Bailey, V. (1998). The Strengths and Difficulties Questionnaire: A pilot study on the validity of the self-report version. European Child & Adolescent Psychiatry, 7, 125–130. doi 10.1007/s007870050057 Goodreau, S. M., Kitts, J. A., & Morris, M. (2009). Birds of a feather, or friend of a friend? Using exponential random graph models to investigate adolescent social networks. Demography, 46, 103–125. doi 10.1353/dem.0.0045 Greenblatt, M., Becerra, R. M., & Serafetinides, E. A. (1982). Social networks and mental health: An overview. The American Journal of Psychiatry, 139, 977–984. doi 10.1176/ajp.139.8.977 Haas, S. A., Schaefer, D. R., & Kornienko, O. (2010). Health and the structure of adolescent social networks. Journal of Health and Social Behavior, 51, 424–439. doi 10.1177/0022146510386791 Hall, J. A., & Valente, T. W. (2007). Adolescent smoking networks: The effects of influence and selection on future smoking. Addictive Behaviors, 32, 3054–3059. doi 10.1016/j.addbeh.2007.04.008 Handcock, M. S., Hunter, D. R., Butts, C. T., Goodreau, S. M., & Morris, M. (2008). statnet: Software tools for the representation, visualization, analysis and simulation of network data. Journal of Statistical Software, 24(1), doi 10.18637/jss.v024.i01

Harris, J. K. (2013). An introduction to exponential random graph modeling. Thousand Oaks, CA: Sage. Jeon, K. C., & Goodson, P. (2015). U. S. adolescents’ friendship networks and health risk behaviors: A systematic review of studies using social network analysis and Add Health data. PeerJ, 3, e1052. doi 10.7717/peerj.1052 Krivitsky, P. N., & Kolaczyk, E. D. (2015). On the question of effective sample size in network modeling: An asymptotic inquiry. Statistical Science, 30, 184–198. doi 10.1214/14-STS502 Larson, R. W., & Verma, S. (1999). How children and adolescents spend time across the world: Work, play, and developmental opportunities. Psychological Bulletin, 125, 701–736. doi 10.1037/0033-2909.125.6.701 Lavy, V., & Sand, E. (2012). The friends factor: How students’ social networks affect their academic achievement and well-being? [NBER Working Paper No. 18430]. Retrieved from http://www.nber.org/papers/w18430 Macdonald, D. I. (1989). Drugs, drinking, and adolescents. Chicago, IL: Year Book Medical Publishers. McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27, 415–444. doi 10.1146/annurev.soc.27.1.415 Morris, M., Handcock, M. S., & Hunter, D. R. (2008). Specification of exponential-family random graph models: Terms and computational aspects. Journal of Statistical Software, 24, 1548–7660. doi 10.18637/jss.v024.i04 Prinstein, M. J. (2007). Moderators of peer contagion: A longitudinal examination of depression socialization between adolescents and their best friends. Journal of Clinical Child and Adolescent Psychology, 36, 159–170. doi 10.1080/15374410701274934 Prinstein, M. J., & Dodge, K. A. (2008). Understanding peer influence in children and adolescents. New York: Guilford. Robins, G., Pattison, P., Kalish, Y., & Lusher, D. (2007). An introduction to exponential random graph (p*) models for social networks. Social Networks, 29, 173–191. doi 10.1016/j.socnet.2006.08.002 Rubin, K. H., Bukowski, W. M., & Parker, J. G. (2006). Peer interactions, relationships, and groups. In N. Eisenberg, W. Damon, & R. M. Lerner (Eds.), Handbook of child psychology: Vol. 3. Social, emotional and personality development (6th ed., pp. 571–645). Hoboken, NJ: Wiley. Schaefer, D. R., Kornienko, O., & Fox, A. M. (2011). Misery does not love company: Network selection mechanisms and depression homophily. American Sociological Review, 76, 764–785. doi 10.1177/0003122411420813 Schaefer, D. R., & Simpkins, S. D. (2014). Using social network analysis to clarify the role of obesity in selection of adolescent friends. American Journal of Public Health, 104, 1223–1229. doi 10.2105/AJPH.2013.301768 Scott, J. (2012). Social network analysis (3rd ed.). London, UK: Sage. Settersten, R. A. (2015). Relationships in time and the life course: The significance of linked lives. Research in Human Development, 12, 217–223. doi 10.1080/15427609.2015.1071944 Valente, T. W. (2012). Network interventions. Science, 337, 49–53. doi 10.1126/science.1217330 Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications. Cambridge, UK: Cambridge University Press. Stéphanie Baggio Life Course and Social Inequality Research Centre University of Lausanne Geopolis Building 5623 1015 Lausanne Switzerland stephanie.baggio@unil.ch

Swiss Journal of Psychology (2017), 76 (1), 5–11

Verständlich und mit zahlreichen Fallbeispielen

Michael Rufer / Susanne Fricke

Der Zwang in meiner Nähe Rat und Hilfe für Angehörige von zwangskranken Menschen 2., akt. u. überarb. Aufl. 2016. 192 S., 3 Tab., Kt € 19,95 / CHF 26.90 ISBN 978-3-456-85556-1 Auch als eBook erhältlich

Verständlich und mit zahlreichen Fallbeispielen geschrieben, leistet dieser praxisnahe Ratgeber einen wichtigen Beitrag zum Umgang mit dem Zwangserkrankten. Zwangserkrankungen wirken sich erheblich auf das Umfeld der Betroffenen aus. Familienangehörige, Partner, Freunde, Arbeitskollegen und andere Menschen, die einem Zwangskranken nahestehen, sind oft alleine gelassen mit ihren Fragen: Wie kann ich dem Betroffenen am besten helfen? Wie kann ich seine zwanghaften Verhaltensweisen verstehen? Wie kann ich mich abgrenzen, und was mache ich dann mit meinem schlechten Gewissen? Wann ist eine professionelle Therapie notwendig? Soll ich einmal mit zu seinem Therapeuten gehen?

www.hogrefe.com

Diese und viele andere häufig gestellte Fragen werden von den Autoren, die beide jahrelange Erfahrung in der Beratung von Angehörigen und in der Therapie von Zwangskranken haben, aufgegriffen und mit Beispielen aus der Praxis und konkreten Tipps verständlich beantwortet. Die 2. Auflage wurde aktualisiert und um ein Kapitel über die Besonderheiten der Situation von Eltern zwangskranker Kinder ergänzt. Darüber hinaus wurden Informationen zu neuen Therapieansätzen eingearbeitet.

E.-M. Ashikali et al.: Co smetic Surgery Sw issJournal Advertising, of Psychology Bod y Image, (2017), ©and 2017 76 (1), Attitudes Hogrefe 13–21

Original Communication

The Impact of Cosmetic Surgery Advertising on Swiss Women’s Body Image and Attitudes Toward Cosmetic Surgery Eleni-Marina Ashikali1, Helga Dittmar1, and Susan Ayers2 1

School of Psychology, University of Sussex, Brighton and Hove, UK

School of Health Sciences, City University of London, UK

Abstract. International concern has been expressed about advertising for cosmetic surgery (British Association of Aesthetic Plastic Surgeons [BAAPS], 2005, 2008). A recent study showed that exposure to such advertising resulted in a more negative body image and attitudes toward surgery among women living in the UK (Ashikali, Dittmar, & Ayers, 2015). This study investigates the impact of cosmetic surgery advertising on women living in Switzerland, a country with relatively little advertising for cosmetic surgery. A group of 145 women (mean age 23.07) were exposed to advertising for cosmetic surgery containing either discount incentives, risk information, no additional information, or to the control condition. Exposure to advertising for cosmetic surgery resulted in increased dissatisfaction with both bodyweight and appearance. Highly materialistic women perceived such surgery as being less beneficial to their image when exposed to advertising for cosmetic surgery as well as when exposed to risk information rather than discount incentives. Moreover, appearance-dissatisfied women considered surgery to a lesser extent when exposed to risk information compared to discount incentives. Our findings highlight the need for research examining the impact of cosmetic surgery media, the content of advertising for cosmetic surgery as well as cultural variability. Keywords: advertising, cosmetic surgery, body image, attitudes

The number of people undergoing cosmetic surgery increased substantially in recent years (American Society of Plastic Surgeons [ASPS], 2016; British Association of Aesthetic Plastic Surgeons [BAAPS], 2016). A report of the top 25 countries for the number of procedures carried out in 2010 showed worldwide interest in cosmetic surgery (International Society of Aesthetic Plastic Surgeons [ISAPS], 2011). The report included countries with a variety of socioeconomic statuses and cultural backgrounds, such as the UK, Brazil, China, India, Germany, and Saudi Arabia. This suggests that interest in cosmetic surgery is not limited to the Western world, but rather is widespread and multinational. A number of reasons could be behind this trend. First, medical and technological advances have allowed for less complicated and less risky procedures, while also reducing recovery time. Second, undergoing a cosmetic procedure has become more affordable for the average person, especially given that many clinics around the world which now offer financial plans, loans, and promotions. Third, and the focus of this research, the increased interest in cosmetic surgery may be related to the role played by the media in normalizing and promoting cosmetic surgery. The present study examines the impact of advertising for © 2017 Hogrefe

cosmetic surgery and of different types of information provided in such advertising on women’s body image and attitudes toward cosmetic surgery in Switzerland. It is a replication of research carried out in the UK, where advertising for cosmetic surgery is more common and rates of cosmetic surgery uptake higher (ISAPS, 2011).

Cosmetic Surgery in the Media In some countries, the media have embraced the cosmetic surgery industry, making it a feature in reality TV shows and documentaries, magazine and internet articles, and as part of radio programs. Another major and widespread promotion of cosmetic surgery is advertising, which is also present in all types of media as well as on billboards in highly crowded public spaces. Advertising for cosmetic surgery typically features images of beautiful women who have allegedly undergone procedures, emphasizing the benefits of surgery. A common feature in such advertising – and the cause for debate among cosmetic surgery associations – is discount incentives appearing in a range of formats. Swiss Journal of Psychology (2017), 76 (1), 13–21 DOI 10.1024/1421-0185/a000187

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

In the UK, for example, such incentives include time-limited direct monetary discounts, promotions like “two areas for one” liposuction or loyalty cards that encourage multiple procedures (BAAPS, 2014, 2015, 2016). Moreover, the use of such discount incentives is increasing. An analysis of cosmetic surgery advertisements placed in popular American magazines between 1985 and 2004 found an increase in promotional sales and offers, whereas information on risks associated with surgery was present in less than 10% of advertisements and did not differ over time (Hennink-Kaminski, Reid, & King, 2010). Both British and American associations for cosmetic surgery have expressed concern about the standard and style of advertising for cosmetic surgery and marketing strategies, saying that it trivializes surgery and does not adequately represent either the severity of procedures or the risks associated with it (BAAPS, 2008, 2014). Advertising guidelines in the UK are similar to those in Switzerland, where clinics and surgeons are able to advertise their practices as long as they present a truthful image of their expertise and of cosmetic surgery itself, without attempting to encourage clients to undergo surgery (Swiss Medical Association, 2006). However, there appears to be a different media environment involving cosmetic surgery in these two countries. Cosmetic surgery reality television, for example, was quite successful in the UK, with American shows having been aired, while UK versions or new programs were also created. In Switzerland, to our knowledge, no such programs have aired, nor were Swiss versions of existing shows created (M. Sorbera, personal communication, September 27, 2010). Overall, it appears that cosmetic surgery is present to a lesser extent in Swiss media than in the UK and the United States. Examining responses of Swiss women may, therefore, provide interesting insights into the impact of such media on women’s body image and on attitudes toward cosmetic surgery when they have not been overly exposed to such media. Two questions related to the media arise: First, what is the actual impact of advertising for cosmetic surgery on women and can different portrayals of surgery have a distinct impact? Second, are there differences in how women who live in different media environments respond to advertising for cosmetic surgery? This study examines these research questions in relation to Swiss women’s attitudes toward cosmetic surgery and also in terms of their body image.

Psychological Research on Cosmetic Surgery Media Correlational research has found a positive relationship between cosmetic surgery media consumption and body dissatisfaction (Henderson-King & Henderson-King, 2005; Markey & Markey, 2009; Sarwer et al., 2005; Sperry, Thompson, Sarwer, & Cash, Swiss Journal of Psychology (2017), 76 (1), 13–21

2009) as well as more favorable attitudes toward surgery and an increased willingness to undergo cosmetic procedures (Crockett, Pruzinsky, & Persing, 2007; Delinsky, 2005; Nabi, 2009; Sperry et al., 2009). Little experimental research on cosmetic surgery has been carried out. However, three studies that did use reality TV as their experimental material showed that exposure to the cosmetic surgery show resulted in an increase in body dissatisfaction and a desire for cosmetic surgery (Markey & Markey, 2010, 2012), greater perceived pressure from the media to be thin as well as increased endorsement of their ability to control their appearance (Mazzeo, Trace, Mitchell, & Gow, 2007). A recent UK study on advertising for cosmetic surgery compared the impact of discount incentives and risk information in such advertising, while also taking into consideration potential moderation variables to these effects (Ashikali, Dittmar, & Ayers, 2015). It showed a negative effect on weight and appearance satisfaction following exposure to advertising for surgery, irrespective of the type of information provided. Moreover, weight satisfaction was moderated by women’s materialistic values. In terms of attitudes toward surgery, exposure to advertising for cosmetic surgery led to less perceived benefits of surgery, whereas exposure to risk information in comparison to discount incentives increased consideration of surgery. These findings suggest that advertising for cosmetic surgery has a negative impact on young women’s body image, and that the impact of such advertising on attitudes toward surgery can vary depending on the type of information provided. Therefore, the content of cosmetic surgery media and the way in which they portray surgery may play a role in how surgery is perceived by the public and the extent to which an individual would consider undergoing a procedure. The UK is representative of the majority of research on cosmetic surgery, which has generally been carried out in the United States, the UK, and Australia. However, to our knowledge, no research has been carried out in a country that is relatively low on cosmetic procedures and where cosmetic surgery is not a commonplace feature in the media. Examining these effects among a sample of women who are likely to have had less daily exposure to cosmetic surgery media is important in gaining an understanding of how such women respond to this type of media. Moreover, the investigation of potential moderating roles will provide insight into which women are most affected by cosmetic surgery media. Finally, discussing findings from the UK and Swiss samples in conjunction can shed light on the impact of different media environments on women’s attitudes toward surgery.

The Present Research This study investigate the effects of advertising for cosmetic surgery in Switzerland, a country low on cosmetic surgery advertisements in the media. Specifically, we investigate whether advertising with discount incentives has a different effect on body image and attitudes toward surgery than advertising with © 2017 Hogrefe

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

risk information does. We also examine a number of potential moderators, most related to body image, but also materialistic values, which may make some women more vulnerable than others in terms of negative responses to advertising for cosmetic surgery (Ashikali & Dittmar, 2012). Our specific hypotheses were: – H1: Exposure to advertising for cosmetic surgery will result in greater body dissatisfaction. – H2: Discount incentives will elicit more positive attitudes toward surgery than risk information will, on the premise that such discounts make surgery more affordable for the general public. Risk information, on the other hand, will highlight the dangers of undergoing surgery, which will discourage women from considering it. – H3: The impact of exposure to advertising for cosmetic surgery on body image and/or cosmetic surgery attitudes will be moderated by one or more of the proposed variables: thin-ideal internalization, restrained eating, trait body dissatisfaction, or materialistic values.

Method Design This experimental study contained four conditions in which women were exposed to advertisements about cosmetic surgery containing no additional information (Condition 1), discount incentives (Condition 2), risk information (Condition 3), or to the control condition, in which advertisements for flower delivery providers were presented. The effect of exposure to these on women’s body image and attitudes toward cosmetic surgery was examined. Potential moderators of thin-ideal internalization, restrained eating, trait body dissatisfaction, and materialistic values were measured at a one-week follow-up session.

Participants Participants were recruited from the University of Neuchâtel in Switzerland. A group of 145 women took part in the study, with 36 exposed to cosmetic surgery advertisements containing no additional information, 39 to advertisements containing discount incentives, 37 to advertisements containing risk information, and 33 to the control condition. 76% of respondents (total N = 97; Condition 1: n = 23; Condition 2: n = 30; Condition 3: n = 20; control condition: n = 24) also participated in the second part of the experiment. Participation was on a voluntary basis, and participants were entered into a prize drawing for two gift vouchers of EUR 25. The overall mean age was 23.07 years (SD = 4.66, range = 18–44) and 79.3% were Caucasian. The overall mean body mass index (BMI) was 21.00 (SD = 2.59, range = 14.7–33.31), with 9% of the participants classified as © 2017 Hogrefe

underweight, 85.5% as normal weight, and 5.5% as overweight according to population-relevant guidelines (Zaninotto, Wardle, Stamatakis, Mindell, & Head, 2006).

Materials Advertisements Each of the four conditions contained two sets of two advertisements presented side-by-side. Participants in each condition, therefore, saw a total of four advertisements. The cosmetic surgery advertisements were identical in all conditions, containing images of idealized media models. However, the text in each condition differed, such that Condition 1 had no additional information; Condition 2 included discounts in the form of direct monetary reductions (e.g., “700 Swiss francs off your first procedure”); and Condition 3 contained information about the risks associated with cosmetic surgery (e.g., “Cosmetic surgery involves risks such as procedural and anesthetic complications, bleeding, scarring, and infection. Our surgeons will provide further information at your first consultation”). We made every effort to ensure that the advertisements we created resembled real-life advertising for cosmetic surgery in terms of the slogans and information provided, and that they were as professional-looking as possible. All advertisements were for all types of procedures (i.e., facial and body). They also included some information about doctors/clinics, sometimes mentioning expertise and years of experience and sometimes in the form of slogans aimed at inspiring trust in prospective clients. Moreover, they emphasized the benefits of undergoing surgery, with mentions of fulfilling a dream and the joys of a “new self.” All of the above are elements that characterize the style of real-life advertising for cosmetic surgery today. In the control condition, we presented advertising for flower delivery services. We wanted the control condition to be a form of advertising that did not reflect any appearance-related matters, so the advertisements did not contain any images of models and the slogans used were not related to appearance. All advertisements were in French.

Questionnaire Measures Given that the study took place in the French-speaking part of Switzerland, all scales were translated and presented in French.

Body-Related Self-Discrepancies The Self-Discrepancy Index (SDI; Halliwell & Dittmar, 2006) was used to assess the activation of women’s self-discrepancies. Participants completed three sentences in the format “I . . . but I would like . . .” allowing them to freely describe in their own words aspects of their life that they would like to Swiss Journal of Psychology (2017), 76 (1), 13–21

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

change. After each sentence, they rated how different they were from their ideal (magnitude) and how concerned they were about this difference (salience) on a 6-point Likert scale, ranging from 1 (a little) to 6 (extremely). Statements were then coded into three categories: appearance-related (e.g., “I have a big nose, but I would like a smaller one”); weight-related (e.g., “I am overweight, but I would like to be thinner”); and unrelated to appearance or weight. Normally, magnitude and salience scores in the appearance and weight categories are multiplied and then added together (Dittmar, 2009; Halliwell & Dittmar, 2006) to produce a unique index for each category. However, given that we wanted to conduct parametric analyses and these indices do not have a normal distribution, we turned weight- and appearance-related self-discrepancies into binary variables (presence or absence of discrepancies) and analyzed them using binary logistic regression. Weight-related self-discrepancies were reported by 38.6% of women, and 57.9% reported appearance-related self-discrepancies. The SDI was also administered in Part 2 of the study to measure women’s trait body dissatisfaction. Given that normal distribution issues in regression only apply to outcome variables, we used the normal procedure for this scale, multiplying and then adding together magnitude and salience scores. Weight-related self-discrepancies ranged from 0 to 36 (M = 4.31, SD = 6.85); appearance-related selfdiscrepancies ranged from 0 to 72 (M = 7.11, SD = 10.78).

Cosmetic Surgery Attitudes Women’s attitudes toward cosmetic surgery were assessed using the Acceptance of Cosmetic Surgery Scale (ACSS; Henderson-King & Henderson-King, 2005). Two of the three 5-item subscales were included in this study: intrapersonal, measuring the extent to which one believes that cosmetic surgery can be beneficial to one’s image (e.g., “Cosmetic surgery is a good thing because it can help people feel better about themselves”); and consider, measuring the extent to which one would consider undergoing surgery (e.g., “I have sometimes thought about having cosmetic surgery”). Reliabilities were good for all subscales: intrapersonal α = .88; consider α = .91. The ACSS was also administered in Part 2 of the study as a control variable to ensure that any significant findings from the exposure experiment were in fact due to the exposure itself rather than preexisting group differences (intrapersonal α = .92; consider α = .92).

Moderator Variables Thin-ideal internalization was measured using a scale developed by Dittmar (unpublished), which comprised 9 items measuring the desire to look like female models/actresses and meet the thin ideal (e.g., “I want to look like media models; I would like to have Swiss Journal of Psychology (2017), 76 (1), 13–21

a thin body”) and 4 items measuring identification with media models (e.g., “I identify with media models”). Reliability was good α = .87. Restrained eating behavior was measured using the relevant subscale from the Dutch Eating Behavior Questionnaire (DEBQ; Van Strien, Frijters, Bergers, & Defares, 1986) with items like “Do you deliberately eat less in order to avoid becoming heavier?” Participants responded on a 5-point Likert scale ranging from 1 (never) to 5 (very often). Reliability for this scale was high, α = .91. Materialistic values were measured using the Aspirations Index (Kasser & Ryan, 1996). This is a measure of the importance of extrinsic compared to intrinsic goals, which not only takes into account a focus on financial success, but also image and popularity as part of its conceptualization of materialism. Six subscales were included: three extrinsic (money, image, popularity) and three intrinsic (affiliation, community, self-acceptance). The degree of importance placed on extrinsic compared to intrinsic values was calculated by subtracting the mean of all subscales from the mean of the three extrinsic subscales combined (Sheldon & Kasser, 2008). Cronbach’s α for the extrinsic goals was α = .89.

Manipulation Check At the end of the study, participants in the experimental conditions were asked whether the advertisements they saw contained (1) no additional information, (2) discount incentives, or (3) risk information. A high percentage of participants in Conditions 1 (75.7%) and 2 (88.9%) noted seeing no additional information or discount incentives, respectively. The risk information in Condition 3, however, was consciously noted by a significantly lower percentage of women (45.2%). This was, to a certain extent, expected as the risk information was included in a more discreet and subtle manner than the discount incentives. Any findings relating to this condition may, therefore, imply that risk information can have an impact on women’s responses even if it is not consciously recognized.

Procedure and Ethical Issues The study was approved by the research ethics committee at the University of Sussex, UK. Participants received an email from their professor with an invitation to participate in the study. The study website randomly allocated participants to one of four conditions. The first page introduced the study, and participants were informed of their right to withdraw and assured confidentiality and anonymity. They were then asked to provide some personal information to create a unique identifier code for each of them so that their answers to Part 1 could be matched up to Part 2 of the study without compromising anonymity. © 2017 Hogrefe

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

The first part of the study was presented as a consumer decision-making task, in which participants were presented with and asked to rate in terms of preference two sets of advertisements allegedly from UK companies that wanted to expand their business into Switzerland. After viewing the advertisements, participants’ self-discrepancies were assessed. Participants then completed the ACSS, in which filler items about flower delivery were embedded in order to maintain the cover story. The study ended with a section on demographic information and a manipulation check. Part 2 of the study was emailed to participants a week after completion of Part 1 in order to minimize the likelihood that responses would be influenced by the experimental manipulation. On the introductory page, participants were asked to use a 6-month time-frame for their responses as this section was focused on their more general and enduring attitudes. Following the consent form and unique identifier code, participants completed the SDI, the Aspirations Index and thin-ideal internalization, followed by the DEBQ, and the ACSS. Participants then completed the manipulation check, were fully debriefed about the true purposes of the study, and were given the opportunity to withdraw or submit their responses.

Results Group Differences in Trait Measures, BMI, and Age Analyses to check whether groups exposed to different conditions differed on trait measures found no differences in BMI, thin-ideal internalization, restrained eating, trait body dissatisfaction, materialistic values, perceived intrapersonal benefits of surgery to their image. However, the groups did differ with respect to whether they would consider surgery, F(3, 93) = 2.74, p = .05, and small differences in age were observed, although they were not significant, F(3, 141) = 2.43, p = .07. We, therefore, controlled for these variables in subsequent regression analyses. We controlled for age in all analyses and we controlled for whether women would consider surgery when looking at a consideration of surgery as an outcome measure. Thus, regressions were structured as follows: age (Step 1); three exposure contrasts: control versus experimental (coded control: –.99; experimental: .33), no information versus information, and discounts versus risks (coded discounts: .33; risks: –.33; Step 2); and significant moderators (Step 3).

groups: Women exposed to advertising for cosmetic surgery reported more weight-related (43.8%) and appearance-related (67.9%) self-discrepancies than those in the control condition (21.2% and 24.2%, respectively). Table 1. Binary logistic regression for weight-related self-discrepancies 95% CI for odds ratio Predictor

Age

–.07

Control vs. experimental No Information vs. information

.84*

Lower OR

Upper

.04

.86

.93

1.01

.36

1.15

2.31

4.65

.10

.42

.48

1.10

2.52

Discounts vs. risks

1.15

.73

.08

.32

1.33

Constant

1.12

.97

3.05

Note. Model χ²(4) = 10.86, p = .03. Hosmer and Lemeshow χ²(7) = 5.18, p = .64. Cox and Snell R2 = .07. Nagelkerke R2 = .10. *p < .05.

Table 2. Binary Logistic Regression for Appearance-Related Self-Discrepancies 95% CI for odds ratio Predictor

Lower OR

Age

–.03

.04

.90

.97

1.05

Control vs. experimental

1.45*** .35

2.16

4.25

8.37

Upper

No Information vs. information –.04

.45

.40

.96

2.31

Discounts vs. risks

.22

.75

.29

1.24

5.40

1.01

.91

Constant

2.74

Note. Model χ²(4) = 21.03, p < .001. Hosmer and Lemeshow χ²(7) = 6.76, p = .45. Cox and Snell R2 = .14. Nagelkerke R2 = .18. ***p < .001.

Effects of Cosmetic Surgery Advertising on Women’s Attitudes toward Surgery Exposure to advertising for cosmetic surgery did not influence women’s perceived benefits of surgery (F(3, 92) = .88, p = .48) or the extent to which they would consider undergoing surgery (F(3, 93) = .34, p = .79). However, several moderating variables emerged as significant for both attitude measures. Perceived intrapersonal benefits of surgery were moderated by both trait weight and appearance self-discrepancies as well as by materialistic values. Consideration of surgery was also moderated by weight-related and appearance-related self-discrepancies and by restrained eating. In Step 3, in order to gain an understanding of which of these variables played a stronger moderating role, we entered all of them together in two separate multiple regressions – one for each measure of attitude. This showed that materialistic values were the strongest moderator of intrapersonal benefits, whereas appearance-related self-discrepancies were strongest for consideration of surgery. In the following, we report these findings in more detail. In terms of intrapersonal benefits, two interactions with materialistic values emerged as significant: the control versus experimental (β = –1.90, p = .007) and the discount versus risk Swiss Journal of Psychology (2017), 76 (1), 13–21

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

Figure 1. Perceived intrapersonal benefits of surgery at different levels of materialism for women exposed to advertising for cosmetic surgery or not.

(β = 3.48, p = .002) exposure contrasts. Simple slopes analyses were carried out for both interactions. Figure 1 shows the interaction between materialistic values and the control versus experimental exposure contrast. When women are exposed to advertising for cosmetic surgery, there is very little difference between women who are low and high on materialistic values (.34 scale points). However, when exposed to neutral advertising, highly materialistic women perceive surgery to be more beneficial to their image by 1.32 scale points than women who are not focused on materialistic values. In terms of the discounts versus risks interaction, there is virtually no difference between women who are low and high on materialistic values when exposed to risk information (Figure 2). Differences in the perception of how beneficial surgery is to image were observed in the discounts condition: Materialistic women perceive surgery to be more beneficial by 1.21 scale points than women who are low on materialistic values. For consideration of surgery, the significant interaction term with trait appearance-related self-discrepancies was in the discount versus risk exposure contrast. As shown in Figure 3, there is little difference between women exposed to discount incentives (.26 scale points) and a larger difference in women exposed to risk information. Here, women who are low on appearance-related self-discrepancies reported considering surgery to a larger extent, scoring .61 scale points higher than women who are high on appearance-related self-discrepancies. In sum, and contrary to our expectations, exposure to advertising for cosmetic surgery did not have an impact on the extent to which Swiss women reported considering surgery or on the extent to which they believed surgery was beneficial to their image. However, some interesting interactions with materialistic values suggest that materialistic women perceive surgery as being more beneficial to their image when they are not exposed to advertising for cosmetic surgery and, in line with our predictions, when they are exposed to discount incentives rather than risk information. Furthermore, consideration of Swiss Journal of Psychology (2017), 76 (1), 13–21

Figure 2. Perceived intrapersonal benefits of surgery at different levels of materialism for women exposed to advertising for cosmetic surgery containing discount incentives or risk information.

Figure 3. Consideration of surgery at different levels of appearance-related self-discrepancies for women exposed to advertising for cosmetic surgery containing discount incentives or risk information.

surgery varied according to women’s preexisting dissatisfaction with their appearance, suggesting that women low on this trait consider surgery more when they are informed of the risks associated with surgery.

Discussion The present study addressed concerns expressed about the nature and style of advertising for cosmetic surgery by investigating its impact on Swiss women, a population with relatively little exposure to such advertising compared to other Western countries, such as the UK or the United States. The main findings for body image were an increased more general dissatisfaction with both weight and appearance following exposure to advertising for cosmetic surgery. These findings are consistent with previous correlational research that showed a link between cosmetic surgery media and body dissatisfaction (e.g., © 2017 Hogrefe

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

Markey & Markey, 2009) and provide experimental evidence of this relationship. Furthermore, they replicate findings with a UK sample (Ashikali et al., 2015) that showed a deleterious effect of advertising for cosmetic surgery on weight and appearance satisfaction. The consistency of these findings across two samples of women, who are living in different cosmetic surgery media environments, provides more robust evidence of the negative effect of advertising for cosmetic surgery on women’s body image. The main findings for attitudes toward cosmetic surgery were that perceived intrapersonal benefits of surgery were moderated by materialism and that considering undergoing a procedure was moderated by trait self-discrepancies in appearance. This suggests that, in this sample of women, only the responses of women who possess these trait variables are affected by advertising for cosmetic surgery. In contrast, in the UK sample, exposure to advertising for cosmetic surgery resulted in a decreased perception of intrapersonal benefits, and exposure to risk information resulted in an increased consideration of surgery. However, some of the findings are similar. Swiss materialistic women and British women who were exposed to advertising for cosmetic surgery believed that surgery was less beneficial to their image compared to those exposed to the control advertisements. This suggests that both samples of women had a negative response toward advertising for cosmetic surgery and rejected the benefits of surgery. A further interaction emerged in the Swiss sample, with materialistic women exposed to discount incentives perceiving surgery as being more beneficial to their image than those exposed to risk information. This could be related to materialistic women’s sensitivity to financial information and the offer of a discount making surgery more appealing. Consideration of surgery was influenced by women’s preexisting appearance-related self-discrepancies, which is consistent with the proposal that individuals with a low evaluation of their appearance are likely to consider surgery to a greater extent than those who are relatively satisfied with their looks (Sarwer, Wadden, Pertschuk, & Whitaker, 1998). Interestingly, when exposed to risks, women high on appearance-related self-discrepancies reported considering surgery to a lesser extent than those low on appearance-related self-discrepancies. However, when they were exposed to discount incentives, the opposite pattern emerged. Apparently, for women who are already dissatisfied with their appearance, the provision of risk information deterred them from cosmetic surgery, perhaps by posing a threat to their appearance by reminding them that surgery will leave them scarred and can involve other complications. These findings suggest that the contents of advertising for cosmetic surgery have an impact on some women’s attitudes toward it, and future research should take the contents of cosmetic surgery media into account. One possible reason for different attitudes toward cosmetic surgery in the UK and Swiss samples is the media environment in each country. For example, UK women are much more fre© 2017 Hogrefe

quently exposed to advertising for cosmetic surgery so they may be more responsive to the way in which it is advertised. Alternatively, these differences may be due to methodological issues such as reporting biases in self-reports of attitudes to cosmetic surgery. Maybe the lower exposure to cosmetic surgery in Switzerland results in more stigma being attached to cosmetic surgery and Swiss women, therefore, being less likely to report having a positive attitude toward it. These issues need to be examined by future research and the variability between different cultures illustrates the value of comparative research in cosmetic surgery. Overall, advertising for cosmetic surgery seems to have an impact on body image and, to a certain extent, on attitudes toward cosmetic surgery. It is important to consider the deleterious effect of advertising for cosmetic surgery on body image when formulating advertising guidelines, irrespective of whether the heightened body dissatisfaction following exposure to such advertising translates into the desire for a procedure. This is because a number of unhealthy behaviors and psychological issues can arise in response to such dissatisfaction (e.g., Stice & Shaw, 2002). The impact on attitudes toward surgery is more difficult to disentangle, and consequently, suggestions for advertising guidelines are also more difficult to make. However, findings from the present study suggest that women who possess certain individual characteristics are more likely to be affected by advertising for cosmetic surgery. Although further research is needed to determine which traits may act as vulnerability factors for negative responses to such advertising as well as the potential impact of different contents of such advertising, our findings suggest that this type of advertising does indeed affect the public. Considering that this type of advertising is for a medical procedure that carries risks to health, the fact that this advertising may encourage people to undergo procedures presents an ethical dilemma for cosmetic surgeons and providers in general. Some countries have responded to this by banning advertising for cosmetic surgery altogether (e.g., France and Italy), whereas others are still working on regulating the industry. Despite the novel aspects of the present research, there are limitations which must be noted. First, the sample was homogeneous in terms of age, ethnicity, and educational background. It is possible that people with different backgrounds or of different ages respond differently to this type of advertising. It would be particularly interesting to investigate the responses of older women, who also account for the greatest proportion of procedures each year (ASPS, 2016). Second, although experimental research is of great value in allowing for cause and effect inferences, this research cannot inform us as to whether the effects found are transient or long-lasting. Longitudinal research would, therefore, be particularly useful in increasing our understanding of the impact of advertising for cosmetic surgery. Third, all of the experimental materials in this study contained images of idealized media models. This makes it difficult to determine whether the adverse effects on body image following exposure to advertising for cosmetic surSwiss Journal of Psychology (2017), 76 (1), 13–21

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

gery were due to the images of the models or to the services being advertised. Therefore, it would be interesting to replicate and extend the present research with advertising for cosmetic surgery which uses a range of different images (for a UK study on this, see Ashikali et al., 2015). Finally, there are a number of individual difference variables that could play a role in how women respond to advertising for cosmetic surgery and how they perceive surgery in general. These include psychopathological symptoms, such as depression and anxiety (e.g., Meningaud et al., 2001), and body dysmorphic disorder (e.g., Glaser & Kaminer, 2005). It is important that future research takes such variables into account, particularly with regard to determining appropriate patient selection. Overall, this study provides further evidence that advertising for cosmetic surgery has a negative impact on women’s body image. However, the contents of such advertising have different and complex effects in different cultures. In Switzerland, the cosmetic surgery attitudes were only influenced in women who were highly materialistic and dissatisfied with their appearance. Therefore, women’s attitudes toward surgery may be influenced by the media environment and the frequency with which they are exposed to advertising for cosmetic surgery. Future research should consider the content of cosmetic surgery media to gain an understanding about which types of portrayals have a negative impact on women. Moreover, a focus should be placed on women’s individual trait differences, which may influence how they respond to these types of media.

Acknowledgments We would like to thank all of the women at the University of Neuchâtel School of Law who participated in this research. Special thanks to Prof. Petros Mavroidis, Danilo Facchinetti, Cindy Amez-Droz, and Samuel Monbaron for kindly facilitating our research and for their invaluable help throughout the data collection process.

References American Society of Plastic Surgeons. (2016). 2015 plastic surgery statistics report. Retrieved from https://www.plasticsurgery.org/news/plastic-surgery-statistics Ashikali, E.-M., & Dittmar, H. (2012). The effect of priming materialism on women’s responses to thin-ideal media. British Journal of Social Psychology, 51, 514–533. doi 10.1111/j.20448309.2011.02020.x Ashikali, E. M., Dittmar, H., & Ayers, S. (2015). The impact of advertising for cosmetic surgery on women’s body image and attitudes toward cosmetic surgery. Psychology of Popular Media Culture. doi 10.1037/ppm0000099 British Association of Aesthetic Plastic Surgeons. (2008). Surgeons “name and shame” inappropriate cosmetic surgery ads. Re-

Swiss Journal of Psychology (2017), 76 (1), 13–21

trieved from http://baaps.org.uk/about-us/press-releases/ 405-surgeons-name-and-shame-inappropriate-cosmetic-sur gery-ads British Association of Aesthetic Plastic Surgeons. (2014). Cut it out: Surgery societies and Britain’s largest surgical royal college slam summer discount. Retrieved from http://baaps.org.uk/aboutus/press-releases/1958-cut-it-out-surgery-societies-an dbritain-s-largest-surgical-royal-college-slam-summer-disco unts British Association of Aesthetic Plastic Surgeons. (2015). The wrong reason to pucker up on Valentine’s Day [Press release]. Retrieved from http://baaps.org.uk/about-us/press-releases/2071-the-wrong-reaseon-to-pucker-up-on-valentine-s-d ay British Association of Aesthetic Plastic Surgeons. (2016). SUPER CUTS Daddy makeovers and celeb confessions: Cosmetic surgery procedures soar in Britain. Retrieved from http://baaps.org.uk/ about-us/audit/2268-super-cuts-daddy-makeovers-and-cel eb-confessions-surgery-procedures-soar-in-britain Crockett, R. J., Pruzinsky, T., & Persing, J. A. (2007). The influence of plastic surgery “reality TV” on cosmetic surgery patient expectations and decision making. Plastic and Reconstructive Surgery, 120, 316–324. doi 10.1097/01.prs.0000264339.67451.71 Delinsky, S. S. (2005). Cosmetic surgery: A common and accepted form of self-improvement? Journal of Applied Social Psychology, 35, 2012–2018. doi 10.1111/j.1559-1816.2005.tb02207.x Dittmar, H. (unpublished). The impact of video games on body image. Dittmar, H. (2009). How do “body perfect” ideals in the media have a negative impact on body image and behaviors? Factors and processes related to self and identity. Journal of Social and Clinical Psychology, 28, 1–8. doi 10.1521/jscp.2009.28.1.1 Glaser, D. A., & Kaminer, M. S. (2005). Body dysmorphic disorder and the liposuction patient. Dermatologic Surgery, 31, 559–560. doi 10.1111/j.1524-4725.2005.31161 Halliwell, E., & Dittmar, H. (2006). The role of appearance-related selfdiscrepancies for young adults’ affect, body image, and emotional eating: A comparison of fixed-item and respondent-generated self-discrepancy measures. Personality and Social Psychology Bulletin, 32, 447–458. doi 10.1177/0146167205284005 Henderson-King, D., & Henderson-King, E. (2005). Acceptance of cosmetic surgery: Scale development and validation. Body Image, 2, 137–149. doi 10.1016/j.bodyim. 2005.03.003 Hennink-Kaminski, H., Reid, L. N., & King, K. W. (2010). The content of cosmetic surgery advertisements placed in large city magazines, 1985–2004. Journal of Current Issues & Research in Advertising, 32, 41–57. doi 10.1080/10641734.2010.10505284 International Society of Aesthetic Plastic Surgery. (2011). ISAPS International Survey on Esthetic/Cosmetic Procedures Performed in 2010. Retrieved from http://www.isaps.org/isaps-global-statistics-2011.html Kasser, T., & Ryan, R. M. (1996). Further examining the American dream: Differential correlates of intrinsic and extrinsic goals. Personality and Social Psychology Bulletin, 22, 280–287. doi 10.1177/0146167296223006 Markey, C. N., & Markey, P. M. (2009). Correlates of young women’s interest in obtaining cosmetic surgery. Sex Roles, 61, 158–166. doi 10.1007/s11199-009-9625-5 Markey, C. N., & Markey, P. M. (2010). A correlational and experimental examination of reality television viewing and interest in cosmetic surgery. Body Image, 7, 165–171. doi 10.1016/j.bodyim.2009. 10.006 Markey, C. N., & Markey, P. M. (2012). Emerging adults’ responses to a media presentation of idealized female beauty: An examination of cosmetic surgery in reality television. Psychology of Popular Media Culture, 1, 209–219. Mazzeo, S. E., Trace, S. E., Mitchell, K. S., & Gow, R. W. (2007). Effects of a reality TV cosmetic surgery makeover program on eat-

E.-M. Ashikali et al.: Cosmetic Surgery Advertising, Body Image, and Attitudes

ing disordered attitudes and behaviors. Eating Behaviors, 8, 390–397. doi 10.1016/j.eatbeh.2006.11.016 Meningaud, J.-P., Benadiba, L., Servant, J.-M., Herve, C., Bertrand, J.-C., & Pelicier, Y. (2001). Depression, anxiety and quality of life among scheduled cosmetic surgery patients: Multicentre prospective study. Journal of Cranio-Maxillofacial Surgery, 29, 177–180. doi 10.1054/jcms.2001.0213 Nabi, R. L. (2009). Cosmetic surgery makeover programs and intentions to undergo cosmetic enhancements: A consideration of three models of media effects. Human Communication Research, 35, 1–27. doi 10.1111/j.1468-2958.2008.01336.x Sarwer, D. B., Gibbons, L. M., Magee, L., Baker, J. L., Casas, L. A., Glat, P. M., . . . Young, L. (2005). A prospective, multisite investigation of patient satisfaction and psychosocial status following cosmetic surgery. Esthetic Surgery Journal, 25, 263–269. doi 10.1016/j.asj.2005.03.009 Sarwer, D. B., Wadden, T. A., Pertschuk, M. J., & Whitaker, L. A. (1998). The psychology of cosmetic surgery: A review and reconceptualization. Clinical Psychology Review, 18, 1–22. doi 10.1016/S0272(97)00047-0 Sheldon, K. M., & Kasser, T. (2008). Psychological threat and extrinsic goal striving. Motivation and Emotion, 32(1), 37–45. doi 10.1007/s11031-008-9081-5 Sperry, S., Thompson, J. K., Sarwer, D. B., & Cash, T. F. (2009). Cosmetic surgery reality TV viewership relations with cosmetic surgery attitudes, body image, and disordered eating. Annals of Plastic Surgery, 62, 7–11. doi 10.1097/SAP.0b013e31817e2cb8

Stice, E., & Shaw, H. E. (2002). Role of body dissatisfaction in the onset and maintenance of eating pathology: A synthesis of research findings. Journal of Psychosomatic Research, 53, 985–993. doi 10.1016/S0022-3999(02)00488-9 Swiss Medical Association. (2006). Annexe 2 au code de déontologie: Directives pour l’information et la publicité [Annex 2 to the Code of Conduct: Guidelines for information and publicity]. Retrieved from http://www.fmh.ch/files/pdf5/anhang_2_FR.pdf van Strien, T., Frijters, J. E. R., Bergers, G. P. A., & Defares, P. B. (1986). The Dutch Eating Behavior Questionnaire (DEBQ) for assessment of restrained, emotional, and external eating behavior. International Journal of Eating Disorders, 5, 295–315. doi 10.1002/1098–108X(198602)5:2<295::AID-EAT2260050209>3 .0.CO;2-T Zaninotto, P., Wardle, H., Stamatakis, E., Mindell, J., & Head, J. (2006). Forecasting obesity to 2010. London, UK: Department of Health.

Susan Ayers School of Health Sciences City University London Northampton Square London, EC1V 0HB UK Susan.Ayers.1@city.ac.uk

Swiss Journal of Psychology (2017), 76 (1), 13–21

FAIR-2 Frankfurter Aufmerksamkeits-Inventar 2 2., überarbeitete, ergänzte und normenaktualisierte Auflage H. Moosbrugger / J. Oehlschlägel Das FAIR-2 ist die zweite Auflage eines vielfach bewährten Verfahrens zur Erfassung interindividueller Unterschiede in Aufmerksamkeitsleistung und Konzentrationsfähigkeit. Es eignet sich für Personen im Alter zwischen 9 und 85 Jahren und kann in allen Praxisbereichen der Psychologie sowie in der Pädagogik, Psychiatrie, Pädiatrie, Gerontologie, Sportwissenschaft u. a. eingesetzt werden. Das FAIR-2 ist ein Paper-Pencil-Test und erfordert die genaue und schnelle Diskrimination visuell ähnlicher Zeichen unter gleichzeitiger Ausblendung aufgabenirrelevanter Information. Es sind zwei parallele Testformen A und B enthalten.

Die aktualisierten Normen basieren auf Stichproben mit einem Gesamtumfang von N = 2993. Als zeitökonomische Alternative für die Auswertung ist ein Testauswerteprogramm verfügbar. Test komplett bestehend aus: Manual, 10 Testheften Form A, 10 Testheften Form B, 16 Auswerteschablonen und Box Bestellnummer 03 171 01 € 125,00/CHF 154.00 zusätzlich erhältlich: Testauswerteprogramm Bestellnummer 50 941 02 € 210,00 / CHF 271.00

FAKT-II Frankfurter Adaptiver Konzentrationsleistungs-Test II Grundlegend neu bearbeitete und neu normierte 2. Auflage des FAKT H. Moosbrugger / F. Goldhammer Der FAKT-II ist die grundlegend neu bearbeitete und neu normierte Realisierung eines seit 1997 bewährten computerbasierten Konzepts zur adaptiven Messung der individuellen Konzentrationsfähigkeit. Zur Bestimmung der Konzentrationsfähigkeit werden die Aspekte Konzentrations-Leistung, Konzentrations-Genauigkeit und Konzentrations-Homogenität erfasst. Die Auswertung des Tests erfolgt automatisch und wird entweder auf dem Bildschirm oder auf dem Drucker ausgegeben. Die Testergebnisse können für den wissenschaftlichen Einsatz expor-

www.hogrefe.com

tiert und mit Statistik-Programmen weiterverarbeitet werden. Die aktualisierten Normen basieren auf einer neuen Normierungsstichprobe von N = 859 Probanden beiderlei Geschlechts zwischen 16 und 55 Jahren. HTS 5* Testkit inkl. Manual und 50 Nutzungen Bestellnummer H5 149 01 € 580,00/CHF 748.00 * HTS 5 benötigt eine HTS 5-Edition oder eine entsprechende HTS 5-Jahreslizenz. Mehr Informationen zu HTS 5 erhalten Sie bei Ihrer Testzentrale.

Original Communication

Memory Self-Efficacy and Memory Performance in Older Adults The Mediating Role of Task Persistence Marine Beaudoin1 and Olivier Desrichard2 1

LIP/PC2S, University Savoie Mont Blanc, Chambéry, France

Faculty of Psychology and Sciences of Education, University of Geneva, Switzerland

Abstract. The present research examined the role persistence plays in mediating the positive impact of memory self-efficacy (MSE, i.e., one’s confidence in one’s own memory abilities) on older adults’ memory performance. In three studies, 81 to 264 older adults completed an MSE scale and carried out an explicit episodic memory task, during which we recorded their study time as an indicator of task persistence. We found that higher MSE was indirectly related to better memory performance through greater persistence during encoding, as measured by longer study time. Indirect effects were of medium size, with point estimates ranging from 0.64 to 0.85. This mediation effect was independent of factors that could be confounded with study time: chronological age, memory span, prior level of memory performance, episodic memory ability, and use of learning strategies (encoding strategies and self-testing). When confronted with difficult memory tasks, older adults who lack confidence in their memory abilities cease their efforts prematurely, which contributes to a decrease in their performance. Encouraging older adults to persist in the face of difficulties during encoding and retrieval may help alleviate the negative impact of low MSE on memory performance and allow researchers and clinicians to more accurately estimate older adults’ true memory abilities. Keywords: memory self-efficacy, aging, persistence, memory performance, study time, motivation

Older adults’ poorer memory performance has been attributed to age-related decrements in cognitive resources such as processing speed, working memory, inhibition, and executive functioning (Craik, 2000; Luszcz, 2011; Park, 2000; Salthouse, 1991). However, decreases in memory performance are also partly due to age-related changes in motivational factors, including loss of interest in performing classic laboratory memory tasks (Hess, 2005), a decline in one’s sense of control over memory (Lachman, Neupert, & Agrigoroaei, 2011), and a lack of confidence in one’s ability to use memory effectively in memory-demanding situations (Desrichard & Köpetz, 2005). In some situations, older adults’ lack of motivation to complete memory-demanding tasks has been shown to amplify, or even create, age-related differences in memory performance above and beyond the age-related decline in actual memory abilities (Desrichard & Köpetz, 2005; Germain & Hess, 2007). This could have serious consequences in the clinical diagnosis of memory deficits in older adults, as well as in the validity of laboratory assessments of memory abilities and the resulting conclusions regarding age-related decrements in memory abilities. Hence, in order to develop memory-testing procedures that more accurately reflect people’s real memory abilities, it is important to understand how each motivational variable affects memory performance (Rahhal, Hasher, & Colcombe, 2001). In addition, knowing more about the motivational determi© 2017 Hogrefe

nants of memory performance could help improve the effectiveness of cognitive training programs designed to maintain older adults’ everyday memory functioning (e.g., McDougall, 2009; West, Bagwell, & Dark-Freudeman, 2008). The present research examines the influence of memory selfefficacy (MSE) and task persistence on older adults’ memory performance. MSE is the confidence one has in one’s own memory abilities (Berry, 1999). This notion originated in Bandura’s concept of self-efficacy, which describes “people’s judgments of their capabilities to organize and execute courses of action required to attain designated types of performances” (Bandura, 1994, p. 71). According to Bandura’s self-efficacy theory, people’s beliefs in their capabilities affect how they engage in activities and the level of performance they attain: The more confident one is in one’s ability to succeed, the more resources one engages in succeeding, and the more likely one’s success is (Bandura, 2003). It is thought that one of the reasons for older adults’ poorer memory performance is their belief that their memory is poor (Cavanaugh & Green, 1990), partly due to the endorsement of ageist beliefs about memory decline that lead many older adults to attribute normal memory failures to memory decline (Vallet et al., 2015). Indeed, MSE has been shown to decrease with increasing age and to be positively related to memory performance on a variety of memory tasks (Beaudoin & Desrichard, 2011). Moreover, there is experimenSwiss Journal of Psychology (2017), 76 (1), 23–33 DOI 10.1024/1421-0185/a000188

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

tal evidence that this age-related decrease in MSE contributes to declines in memory performance during aging (Desrichard & Köpetz, 2005). Although the impact of MSE on memory performance is well documented, less is known about the mechanisms responsible for this influence and the relative weight of these mechanisms (Hess, 2005). According to Bandura’s self-efficacy theory, lower MSE should result in less effort expenditure, less persistence in the face of difficulty, lower performance goals, less use of optimal strategies, and higher anxiety during memory tasks. All of these factors are likely to result in poorer performance on difficult memory tasks (Bandura, 2003, 2012; Berry, 1999). Nevertheless, the roles that strategies, goal setting, cognitive effort, persistence, and anxiety play in the MSE-memory performance relationship have not yet been conclusively demonstrated and researchers acknowledge the need for further research in this area. In the present study, we focus on persistence in order to determine whether this variable actually mediates the impact of MSE on memory performance. According to self-efficacy theory, one way people’s self-efficacy influences their performance is by determining “how long they will persist in the face of obstacles and aversive experiences” (Bandura, 1982, p. 123). Persistence is usually operationalized as the time allocated to mastering difficult or unsolvable tasks, or the number of items a person tries to resolve within a given task. Numerous studies provide empirical support for the positive impact of self-efficacy on persistence in a wide variety of activities (e.g., Cervone & Peake, 1986; Damisch, Stoberock, & Mussweiler, 2010; Jacobs, Prentice-Dunn, & Rogers, 1984; Kanfer, Wanberg, & Kantrowitz, 2001; Multon, Brown, & Lent, 1991; Narciss, 2004), but few studies have been conducted in the memory domain. Persistence during word-list or text free-recall tasks has been operationalized as the amount of time spent studying the material to be remembered (i.e., study time), or the time taken to recall the material (i.e., recall time). Published studies of the mediating impact of task persistence in the MSE-memory performance relationship have failed to show any relationship between MSE and study time (Pelegrina, Bajo, & Justicia, 1999, Study 4; Stine-Morrow, Shake, Miles, & Noh, 2006, Study 1) or recall time (Wells & Esopenko, 2008). Gardiner, Luszcz, and Bryan (1997) is the only study to have tested this mediation using an experimental design where task-specific MSE was manipulated by varying the expected difficulty of the memory tasks to be performed, which was actually of identical difficulty between groups. Their manipulation was effective in influencing MSE before task completion, but they did not find any effect on memory performance or on study time. Thus, previous studies provide no evidence for the role of persistence in the impact of MSE on memory performance. However, several characteristics of these studies preclude any firm conclusions being drawn from their findings. Although the participants in Gardiner et al.’s (1997) experimental study reported lower MSE scores when the task to be completed was Swiss Journal of Psychology (2017), 76 (1), 23–33

announced as being difficult (resulting in the effectiveness of the manipulation to affect participants’ MSE before task completion), they may have accurately appraised the actual difficulty of the task at the time of its completion, resulting in the ineffectiveness of the manipulation to affect participants’ MSE during task completion. In addition, the null findings reported by the three correlational studies may be due to insufficient sample size (N = 26; Wells & Esopenko, 2008), insufficient perceived difficulty of the memory tasks used (5- to 10-proposition sentences; Stine-Morrow et al., 2006), or lack of statistical power due to dichotomization of the continuous study-time variable (Pelegrina et al., 1999).

Objective and Overview of the Studies The preceding review highlights the need for further research to test the postulated mediating effect of persistence on the impact of MSE on memory performance. We analyzed three different sets of data in order to examine persistence as a mediator of the MSE-memory performance relationship. The data came from three larger studies that were not uniquely designed to test the mediation-through-persistence hypothesis. Instead, we took advantage of these larger studies to collect the measures used in the present research so as to test the mediation hypothesis. Consequently, the three studies have differences in their samples, measures, and methods that can contribute to the increase in the generalizability of their findings. In all three studies, older adults completed an MSE scale and carried out a memory task, during which we recorded their study times as an indicator of their persistence during encoding. The present studies were conducted on large samples (80 to 250 participants), which overcomes the possible limitations of previous research. Moreover, the memory tasks were difficult enough to maximize the impact of MSE on resource allocation (Beaudoin & Desrichard, 2011). Study time, as measured on a continuous scale, was used as an indicator of persistence. We postulated that persistence – as measured by study time – would mediate the positive relationship between MSE and memory performance.

Method Participants and Procedure Study 1 We examined the data (N = 264, Table 1) obtained from a population-based sample of older adults who had participated in the second wave of a French longitudinal study of normal © 2017 Hogrefe

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

Table 1. Characteristics of the samples Study 1

Study 2

Study 3

264

100

Basic demographics N % Female

Mean age (SD); range

69 (7.76); 57–86

68.52 (7.77); 57–86

62.75 (7.78); 55–94

Mean education (SD); rangea

4.96 (1.77); 1–7

4.95 (1.76); 2–7

6.85 (1.80); 2–9

Maximum number of outliers

Maximum number of missing cases

1 case

0 cases

mental study, we used the active self-perception manipulation method (Salancik & Conway, 1975) to manipulate participants’ MSE. This manipulation was not effective and did not have any influence on the measures collected, so it is not discussed further.1 After the manipulation at the beginning of the testing session, participants were asked to complete a short MSE assessment and to perform a memory task. Their study time was recorded. Strategy use and self-testing were assessed via retrospective reports. At the end of the testing session, the participants were debriefed and thanked. Participants were compensated with EUR 5 for taking part.

Cases deleted from data analyses

Note. 7-point scale ranging from 1 (no formal education) to 7 (some graduate school; Studies 1 and 2); 9-point scale ranging from 1 (no formal education) to 9 (Master’s degree; Study 3).

aging (SAM-CC, Sentiment d’Auto-efficacité Mnesique – Causes et Conséquences; Vallet, 2012). A battery of memory tasks and questionnaires was administered in a larger study on normal aging. After providing written informed consent, each participant was tested during a single, individual interview. Participants were asked to provide a range of sociodemographic (age, level of education, etc.) and medical data (on-going treatments, medical history), and to complete the Mini-Mental Status Examination (MMSE; Folstein, Folstein, & McHugh, 1975). They were then asked to complete several cognitive tests (episodic memory, short-term and working memory spans, speed of processing) and questionnaires assessing MSE, affective state, and health state (depression, trait anxiety, self-esteem), self-perception (social identification as an older adult), and meta-memory (attribution of memory failures, retrospective memory functioning). The order in which the memory tests and questionnaires were administered was varied randomly between participants, with the exception of the episodic memory task used in the present study, which was always administered after the MSE questionnaire. At the end of the interview, participants were debriefed and thanked for participating. Only the measures used for the present study are described below.

Study 2 The participants from Study 1 were offered the opportunity to participate in a subsequent experimental study of MSE and performance in older adults, which started a few months later (Beaudoin, 2008). These participants formed the basis for the sample examined in Study 2 (N = 81, Table 1). In the experi-

Study 3 Because Studies 1 and 2 were conducted with participants from the same sample, we examined the data collected in a third study in order to replicate Study 2 in a different sample of older adults. Participants in Study 3 were 100 community-dwelling older adults (Table 1) who were tested in individual single sessions. After providing informed consent, each participant completed two episodic memory tests, the MSE assessment, and a word-list recall task. Study time was recorded, and strategy use and self-testing were assessed via retrospective reports.

Exclusion Criteria We excluded people who had neurological histories or who were undergoing any form of medical treatment that could affect their cognitive or affective functioning (Studies 1, 2, and 3). We also excluded people who presented with episodic memory deficits (pathological scores on two subtests of the Batterie d’Efficience Mnésique, BEM; Signoret, 1991) from Study 3, or who showed signs of dementia (score < 26 on the Mini-Mental Status Examination, MMSE; Folstein, Folstein, & McHugh, 1975) from Studies 1 and 2. All participants were native French speakers. The data from five participants in Study 2 were discarded due to noncompliance. These participants had not understood that they were to indicate when they stopped studying the word list. In addition, data from two participants in Study 3 were deemed invalid and excluded (one was highly agitated during testing and not willing to complete some of the questionnaires, and one reported having received some very bad news before the testing session and appeared highly distracted during testing).

1 The participants were randomly assigned to one of two conditions either designed to increase (high MSE condition) or decrease (low MSE condition) their MSE. The MSE manipulation was found to have no significant effect on MSE (M = 2.80, SD = 0.64, and M = 2.68, SD = 0.54, t(79) = –0.903, p = .369, for the low MSE and high MSE conditions, respectively), word recall performance (M = 12.93, SD = 4.16, and M = 13.22, SD = 4.49, t(79) = 0.296, p = .768), or study time (M = 172.10, SD = 75.95, and M = 166.24, SD = 96.46, t(79) = –0.294, p = .769).

Swiss Journal of Psychology (2017), 76 (1), 23–33

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

Covariates To ensure that any mediating effect of study time can be attributed to persistence, we measured and controlled two time-consuming behaviors, namely, self-testing (i.e., assessing one’s current level of learning by trying to recall the words already learned; Hager & Hasselhorn, 1992) and the use of encoding strategies (e.g., organizing the material to be remembered by linking items into larger units such as semantic categories, elaborating on individual items by creating mental images; Bailey, Dunloksy, & Hertzog, 2009; Devolder & Pressley, 1992; Hertzog, McGuire, Horhota, & Jopp, 2010; Hertzog, McGuire, & Lineweaver, 1998). These are optimal learning behaviors that take time to be implemented (Bailey, Dunlosky, & Hertzog, 2010; Bryan, Luszcz, & Pointer, 1999; Dunlosky & Hertzog, 1998; Froger, Bouazzaoui, Isingrini, & Taconnat, 2012) and that are likely to mediate the positive impact of MSE on memory performance (Berry, 1999; Hertzog & Dixon, 1994; Hertzog, Dixon, & Hultsch, 1990; Lachman, Bandura, Weaver, & Elliott, 1995). Thus, they could be confounded with persistence as measured with study time. We noticed this potential limitation following analyses of Study 1; therefore, strategy use and self-testing were only included as covariates in Study 2 and Study 3. In addition, we considered that a fair test of the mediating role of persistence in the MSE-performance relationship should control potential confounds that, if not controlled, may create spurious correlations between MSE, persistence, and memory performance. According to self-efficacy theory, previous memory performance (i.e., mastery experiences; Bandura, 1977), which is also a proxy for future performance, affects one’s selfefficacy. Consequently, the impact of MSE on memory performance can be adequately assessed only through partial correlations that control for preperformance assessments of actual memory abilities or previous achievements, as recommended by Bandura (1989). Empirical findings also suggest that MSE and episodic memory performance are independently associated with chronological age (Henry, MacLeod, Phillips, & Crawford, 2004; Hertzog, Lineweaver, & McGuire, 1999; McDonald-Miszczak, Hertzog, & Hultsch, 1995) and depressive symptoms (Burt, Zembar, & Niederehe, 1995; Kalska, Punamaki, Makinen-Pelli, & Saarinen, 1999; Kizilbash, Vanderploeg, & Curtiss, 2002; La Rue et al., 1996). Furthermore, lower persistence could be predicted by increasing age, as suggested by studies reporting shorter study times for older adults than for younger adults (Murphy, Sanders, Gabriesheski, & Schmitt, 1981; Souchay & Isingrini, 2004), and a higher level of depressive affect, as suggested by the Mood Behavior Model (Gendolla, 2000). In all three studies, we collected and controlled participants’ age, depressive affect (Study 1), and actual memory abilities or previous memory achievements, which were as-

sessed using memory span (Study 1), prior performance on a free recall task (Study 2), and standardized tests of episodic memory ability (Study 3). We dropped depressive affect from the covariates included in Study 2 and Study 3 since symptoms of depression were not identified as a potential confound in our analyses of Study 1.

Measures MSE We used the Questionnaire d’Auto-Efficacité Mnésique (QAEM; Beaudoin, Agrigoroaei, Desrichard, Fournet, & Roulin, 2008; Berry, West, & Dennehey, 1989). The QAEM comprises six memory exercises (remembering names, digits, years, shopping lists, object locations, and the location of symbols in a grid), with five levels of difficulty each. Participants are to indicate whether they think they will successfully complete the exercise at each level, for example, “If someone showed you the pictures of 16 familiar objects (e.g., lamp, umbrella, etc.), would you be able to look at them once and recall the names . . . of all 16 objects, of 12 of the 16 objects, of 8 of the 16 objects, of 4 of the 16 objects, of 2 of the 16 objects?” For each exercise, we noted the number of levels the participant felt capable of successfully completing. We averaged the scores across the six exercises to calculate an overall MSE score, ranging from 0 to 5, with higher scores corresponding to higher MSE (α = .80 and α = .88 in Studies 1 and 3, respectively). In Study 2, we used a short version of the QAEM containing three memory exercises (digits, symbols, and shopping lists) selected from the original QAEM.2 We averaged the scores across the three exercises to calculate an overall MSE score ranging from 0 to 5, with higher scores corresponding to higher MSE (α = .67). In Studies 1 and 3, MSE scores obtained from the three items that constituted the short QAEM were highly correlated with the scores obtained from the full-scale QAEM (r = .92 and r = .96, ps < .01, respectively).

Memory Performance Participants were asked to study a list of stimuli, which varied across studies. The stimuli were printed sequentially on a sheet of paper. The participants were told they could study the list for as long as they needed in order to recall as many stimuli as possible in a subsequent, self-paced recall test. When they had finished studying the list, participants performed a 30-s arithmetic distractor task and were then asked to orally recall as many stimuli as they could remember, taking as long as they wished. The number of stimuli each participant correctly re-

2 The QAEM was shortened to 3 items to reduce the time taken to complete the questionnaire before the memory task was administered in order to preserve the MSE manipulation that was planned in the larger experimental study.

Swiss Journal of Psychology (2017), 76 (1), 23–33

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

called from the list was recorded and used to measure memory performance in the three studies. In Study 1, we used a list of 24 concrete nouns from the Word List subtest of the French version of the Wechsler Memory Scale III (Wechsler, 2001) that were printed sequentially on a sheet of paper. In Study 2, participants were asked to study 24 drawings of everyday objects selected from the Picture Recognition subtest of the Rivermead Behavioral Memory Test (RBMT; Vanier & Lemyze, 1994) in order to recall them by name. Exact labels and synonyms were considered correct. In Study 3, we used a list of 25 concrete nouns that we had created.

Persistence The participants were asked to tell the experimenter when they were done studying the list of stimuli. The experimenter was present while they were studying the list and discretely recorded the time taken to study the list (study time, in s). The stopwatch was started when the experimenter gave the list to the participant and stopped as soon as the participant told the experimenter that he/she was done studying. This value was used as a measure of persistence during encoding, following the classical procedure used in previous studies.

Strategy Use (Studies 2 and 3) Strategy use was assessed retrospectively by asking participants to describe any methods they used to study the list of stimuli. The strategy use interview was conducted immediately following the recall phase of the memory task. Retrospective self-reports are often used in the literature on strategy use (e.g., Hertzog et al., 1998, 2010) and there is evidence to support their convergent validity with concurrent reports (Dunlosky & Kane, 2007). Retrospective reports were coded to indicate whether at least one encoding strategy was used (coded 1) or not (coded 0). Strategies for word-list free-recall tasks involve relating words to each other using categorization, the creation of sentences or mental images (e.g., Bryan et al., 1999; Hertzog, Price, & Dunlosky, 2008), item-specific processing such as elaboration of individual items or self-reference encoding, and repetition (e.g., Camp, Markley, & Kramer, 1983; Symons & Johnson, 1997). Effort and concentration were not considered study strategies.

Self-Testing (Studies 2 and 3) The use of self-testing was assessed retrospectively by explicitly asking participants if they had evaluated whether they could remember the stimuli by seeing if they were able to recall, during the study time, the stimuli they were trying to memorize. A © 2017 Hogrefe

binary variable was created indicating whether self-testing was used (coded 1) or not (coded 0).

Preperformance Assessments The indicators we used to control actual preperformance memory abilities or previous achievements, as recommended by Bandura (1989), varied across studies. In Study 1, participants’ memory span was estimated using four computerized simplespan and complex-span tasks (Fournet et al., 2012). Since the participants’ scores on the four span tasks were positively correlated with each other (.23 < rs < .57), we created a single memory-span score, following a principal components analysis (PCA). Kaiser’s criteria and the result of the parallel analysis (Preacher & MacCallum, 2003) supported a one-factor solution explaining 54.92% of the variance. We used the factor scores obtained from the PCA as an indicator of participants’ memory spans. In Study 2, we followed Bandura’s (1989) recommendation to control for prior achievement. As an indicator of past memory performance, we used participants’ performance on the word-list free-recall tasks they had performed during Study 1. When confronted with a memory task, people are likely to base their MSE evaluation on prior experience with similar tasks (Hertzog et al., 1990). Because the QAEM assesses participants’ MSE by asking them to predict their performance on various laboratory memory tasks, and our participants had previous experience with a laboratory free-recall task, we assumed that they would use this experience in their MSE assessments. Hence, we considered participants’ performance on the free-recall task used in Study 1 a valid indicator of previous achievement. In Study 3, actual memory abilities were assessed using standardized memory tests from the Batterie d’Efficience Mnésique (BEM). The story-recall task involved a 12-sentence story read by the experimenter, immediately followed by an oral free-recall test. In the complex-figure recall task, participants were presented with a picture composed of several geometric shapes, which they studied for 1 minute. They were then asked to reproduce the picture on a blank sheet of paper. Story recall and figure recall scores ranged from 0 to 12. We calculated a composite score by averaging the two scores (r = .27, p = .007). In order to obtain assessments of participants’ memory abilities that were free of any negative influences related to low confidence or stereotype threat, the standard instructions for each of the tasks were preceded by a short description of their objectives that de-emphasized the memory component of the tests (Desrichard & Köpetz, 2005; Rahhal et al., 2001). We did this by presenting the story and figure recall tasks as measuring verbal and visual abilities, respectively. Swiss Journal of Psychology (2017), 76 (1), 23–33

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

Depressive Symptoms (Study 1) We used the Depression subscale of the Symptom Check-List Revised (Derogatis, 1977; Pariente & Guelfi, 1990), which asks participants to indicate the extent to which they are feeling 11 symptoms of depression. This measure gives a depression score of between 1 (slight depression) and 5 (deep depression; subscale α = .85).

Data Analyses Study times were logarithmically transformed to ensure normality and variance homogeneity. Because the distribution of the depression scores presented a positive asymmetry, with 88% of the participants reporting depression scores less than 2, we created a binary variable that indicated whether depressive symptoms were absent (scores less than 2, coded 0) or present (scores from 2 to 5, coded 1). As recommended in the literature (Preacher & Hayes, 2004), we conducted the mediation analyses following two steps. First, we examined the total effect of MSE on recall performance (Step 1) via a multiple regression analysis in which we adjusted for covariates. Then we tested the indirect effect of MSE on performance through study time (Step 2) using the product-of-coefficients method and controlling for covariates. The product-of-coefficients method examines whether the product of the regression coefficients a (MSE predicting study time) and b (study time predicting memory performance, controlling for MSE) is significantly different from 0. This latter stage was done using the bootstrapping method, which is the recommended method for obtaining ac-

curate confidence limits for indirect effects in mediation analyses (Hayes, 2009; Preacher & Hayes, 2008). We carried out the analysis using Preacher and Hayes’ (2008) macro for SPSS, performing 5000 bootstrap resamples, and calculating 95% bias-corrected (BC) confidence intervals (CI) around the estimate of the indirect effect ab. In addition, we computed the κ² effect-size measure for mediation effects (Preacher & Kelley, 2011). This effect-size index estimates the magnitude of the indirect effect relative to the maximum possible indirect effect that could have been obtained given the variables involved in the mediation model. It is interpreted as the proportion of the maximum possible indirect effect. Statistical analyses were carried out after deleting outliers, diagnosed using distance, leverage, and influence statistics (Howell, 1998). Outliers were treated for each analysis separately. Missing cases were handled using pairwise deletion. Indications of the number of participants who were deleted in each study are provided in Table 1. All of the analyses met the linear regression assumptions of linearity, normality, and homogeneity of variance.

Results Descriptive Statistics Descriptive statistics for Studies 1–3 are presented in Tables 2–4, respectively. Of the participants in Study 2, 47% used self-testing during the study and 69% used at least one encoding strategy. Of the participants in Study 3, 42% used selftesting during the study time and 97% used at least one encoding strategy.

Table 2. Descriptive statistics for the study variables and relationships with the control variables, Study 1 β Variable

Range

Age

Depression

Memory span

MSE

3.01

0.56

1.33–4.83

–.20*

–.13*

.17*

Study time (in s)

192.10

123.27

8–880

–.13*

–.08

.11#

Word recall

10.77

4.96

0–24

–.34*

–.18*

.33*

Note. MSE = memory self-efficacy. Standardized regression coefficients were obtained from separate simple regression analyses, with age, depression scores, and memory span as predictors, and MSE, study time, and word recall as outcomes. #p < .10, *p < .05.

Table 3. Descriptive statistics for the study variables and relationships with the control variables, Study 2 β Variable

Range

Age

Past performance

MSE

2.74

0.59

1.33–4

–.19#

.20#

Study time (in s)

211.67

191.58

37–1206

–.19

.34*

Object recall

13.06

4.28

5–24

–.41*

.54*

Age

68.52

7.77

57–86

Past performance

11.16

5.18

0–23

Note. MSE = memory self-efficacy. Standardized regression coefficients were obtained from separate simple regression analyses, with age and past performance as predictors, and MSE, study time, and object recall as outcomes. #p < .10, *p < .05.

Swiss Journal of Psychology (2017), 76 (1), 23–33

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

Table 4. Descriptive statistics for the study variables and relationships with the control variables, Study 3 β Variable

Range

Age

Episodic memory ability

MSE

3.65

0.81

1.67–5

–.15

.32*

Study time (in s)

319.05

228.49

60–1630

–.01

.24*

Word recall

15.19

5.67

2–25

–.23*

.49*

Age

62.75

7.78

55–94

Episodic memory ability

8.86

1.27

5–11.25

Note. MSE = memory self-efficacy. Standardized regression coefficients were obtained from separate simple regression analyses, with age and episodic memory ability as predictors, and MSE, study time, and word recall as outcomes.*p < .05.

Mediation Analyses Coefficients for the a (MSE predicting study time), b (study time predicting memory performance, controlling for MSE), c’ (direct effect of MSE on memory performance, controlling for study time) and c (total effect of MSE on memory performance, not controlling for study time) paths are presented in Figure 1.

In Study 1, MSE scores positively predicted recall scores (β = .14, p = .019) when age, depression symptoms, and preperformance assessment were controlled for. Moreover, the test of the indirect effect of MSE on memory performance via study time indicated a significant indirect effect, with a point estimate of 0.85, 95% CI [0.21, 1.59], when covariates were controlled for. The indirect effect was of medium size (κ² = .11). Both of these results are in line with our predictions. In Studies 2 and 3, we examined the total effect of MSE on memory performance adjusting for age, preperformance assessment, strategy use, and self-testing. In Study 2, as predicted, the relationship between MSE and memory performance was positive, although only marginally significant (β = .19, p = .062). Moreover, the results showed a significant indirect effect of MSE through study time, with a point estimate of 0.64, 95% CI [0.06, 1.53], when covariates were controlled for. The indirect effect was of medium size (κ² = .11). As in Study 2, we planned to adjust for strategy use in Study 3. However, only three participants did not use any effective encoding strategies; therefore, strategy use was not included as a covariate in the analyses, which were conducted on the 97 participants who reported having used a strategy. As predicted, the correlation between MSE and memory performance was positive, although only marginally significant (β = .19, p = .061). Next, we examined the indirect effect of MSE on memory performance through study time, including age, preperformance assessment, and self-testing scores as covariates. The results showed a significant indirect effect of MSE, with a point estimate of 0.82, 95% CI [0.14, 1.76]. The indirect effect was of medium size (κ² = .14).

Relationships Between Main Variables and Potential Confounds

Figure 1. Standardized regression coefficients for the impact of MSE on memory performance as mediated by study time, controlling for covariates. The standardized regression coefficient between MSE and memory performance, controlling for study time, is in parentheses (i.e., c’ path). In Study 3, the mediation analysis was conducted on the 97% of the sample who used at least one encoding strategy. †p < .10, *p < .05.

We carried out additional analyses to examine whether our control variables were actually correlated with MSE, study time, and recall scores. MSE, memory performance, and study time were separately regressed on age, preperformance assessments, and depression (Tables 2–4). As predicted, better preperformance assessments (i.e., higher memory span, better prior performance on a free recall task, better episodic memory Swiss Journal of Psychology (2017), 76 (1), 23–33

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

abilities) predicted higher MSE, better recall performance, and longer study time in all three studies (although the correlations with study time in Study 1 and with prior performance in Study 2 were marginal). In addition, increasing age was associated with poorer recall performance in all three studies. Participants’ age was also negatively associated with MSE and study time; however, these correlations were only significant in Study 1. Symptoms of depression were associated with lower MSE and lower recall scores; however, they were not correlated with study time. We then carried out separate regression analyses to examine whether self-testing and strategy use predicted longer study times and better performance. In Study 2, self-testing predicted better memory performance (β = .25, p = 022) and longer study time (β = .47, p < .001), as predicted. In addition, strategy use predicted longer study time (β = .24, p = .037). However, not using any strategy was not associated with lower recall scores than using a strategy (β = .09, p = .451). Overall, these results are in line with our predictions. They were partially replicated in Study 3, in which self-testing was associated with longer study time (β = .22, p = .031), but was not correlated with memory performance (β = .08, p = .414). We tested the impact of MSE on self-testing and on strategy use via logistic regression analyses. Contrary to our expectations, MSE did not predict self-testing (OR = 0.94, p = .954, and OR = 1.12, p = .664, in Studies 2 and 3, respectively) or strategy use (OR = 2.10, p = .114, Study 2).

Discussion The goal of the present research was to provide a fair test of the mediating role of persistence in the MSE-memory performance relationship by addressing several limitations of previous studies. We used data from three correlational studies to test the hypothesis that the positive impact of MSE on memory performance is mediated by persistence, as measured by study time. In contrast to previous studies, large samples of older adults and difficult memory tasks were used in order to create optimal conditions to observe the mediating role of persistence. In all three studies, we found a significant or marginally significant total effect of MSE on memory performance. Despite the good reliability and validity of our measures, the effect sizes were small, ranging from β = .14 to β = .19. However, they fell within the range of effect sizes found in the literature. Beaudoin and Desrichard’s (2011) meta-analysis on the MSE-memory performance relationship yielded a mean effect size of .15 with a 95% CI from .13 to .17. One explanation of such small values lies in the motivational nature of the processes involved in the MSE-memory performance relationship. Indeed, memory performance is likely to be mostly caused by stable memory abilities or automatic processes, and motivation may play a significant but only secondary role. Swiss Journal of Psychology (2017), 76 (1), 23–33

In addition, the results of our studies confirm that the relationship between older adults’ MSE and memory performance is mediated by their persistence during encoding, as measured by study time. Importantly, the mediating role of study time was found to be significant in the three studies we conducted. This replication increases the robustness of our finding and its generalizability, since the three studies were conducted on two different samples of older adults who were confronted with varying experimental situations and memory assessments. Moreover, the mediating role of study time was found to be robust to the statistical control of interindividual differences in age, depressive affect, and memory span (Study 1), and in age, episodic memory ability/previous achievement, strategy use, and self-testing (Studies 2 and 3). However, the indirect effect sizes were medium at best. This may indicate that persistence is a weak mediator of the MSEperformance relationship, which should stimulate further investigations of the role of other potential mediators. Alternatively, medium effect sizes could be partially due to the use of observed variable analyses. While the observed variable approach maximizes precision and statistical power in mediation analyses, it tends to underestimate indirect effects (Ledgerwood & Shrout, 2011). Now that the present research has confirmed the expected MSE → study time → performance mediation through persistence using an observed variable approach, further studies should be conducted using latent variables and structural equation modeling to get a more accurate estimate of the mediation effect. Memory span (Study 1), past memory performance (Study 2), and episodic memory ability (Study 3) were positively correlated with MSE, study time, and current memory performance, thereby supporting the need to control for actual memory abilities or preperformance assessment, as recommended by Bandura (1989). However, we found only partial support for our hypothesis that age or depressive affect act as confounding variables, as they were not related to study time. The present research was conducted on samples of older adults with a low prevalence of depressive affect, which could have restricted the influence of chronological age and depressive symptoms on persistence. In addition, strategy use and self-testing were not related to MSE. One possible explanation for these results is that some participants, independent of their MSE, may not have known about the strategies we considered effective and consequently did not use them. Alternatively, because we did not code the extent to which participants used each strategy they reported, it may be that participants with low MSE who reported using normatively effective strategies did so for a shorter length of time, or for fewer items, than those with high MSE. Nevertheless, the expected impact of MSE on the spontaneous use of effective encoding strategies has not always been found (Hertzog et al., 1998, 2010; Lachman, Andreoletti, & Pearman, 2006; Wells & Esopenko, 2008). More surprisingly, the relationship between self-testing and memory performance, which © 2017 Hogrefe

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

was observed in Study 2, was not observed in Study 3. Because knowledge of possible control actions is needed in order to regulate learning adequately (Hertzog et al., 2008), it is plausible that previous experience with laboratory memory tasks (as in Study 2) maximizes the impact of self-testing on memory performance. In addition, whereas strategy use was associated with increased study time, we found no relationship between strategy use and memory performance in Study 2. One explanation is that the reporting of a strategy does not necessarily mean that it is correctly utilized, particularly in older adults in whom a utilization deficiency has been found to be related to decreased processing resources (Gaultney, Kipp, & Kirk, 2005; Taconnat et al., 2006). The lack of relationship between MSE and strategy use or self-testing eliminates the possibility that these variables are completely confounded with MSE, study time, and memory performance. Nevertheless, encoding strategies and self-testing are time-consuming behaviors (Studies 2 and 3; Bailey et al., 2010; Bryan et al., 1999; Dunlosky & Hertzog, 1998; Froger et al., 2012) that, if not controlled for, could increase the measurement error resulting from using study time as an indicator of persistence. Unfortunately, previous tests of the MSE → study time → performance hypothesis have rarely controlled for these self-regulation behaviors, which could explain why previous studies have failed to find evidence supporting the mediation. Importantly, our three studies support the mediating role of persistence as measured by study time when timeconsuming behaviors contributing to measurement errors are controlled for. Interpretations of the indirect effect of MSE on memory performance through study time in our studies depend on the assumption that all potential confounds were controlled for. Although we controlled for the influence of several theoretically relevant variables, there may be other variables that also play a role. In order to provide a more definitive test of the MSE → persistence → memory performance mediation hypothesis, it will be necessary to use experimental designs in which MSE is manipulated. This is a difficult task, as suggested by the lack of experimental studies in the MSE literature and previous failures to manipulate MSE using the active self-perception manipulation method (Study 2) or the expected difficulty of memory tests (Gardiner et al., 1997), but we encourage researchers to explore alternative ways in which self-efficacy has been manipulated in other domains (e.g., false feedback; Marquez, Jerome, McAuley, Snook, & Canaklisova, 2002). Furthermore, operationalizing persistence during encoding more strictly as a continued study when subjective difficulties are encountered, rather than total study time, may be a better way of disentangling persistence and time-consuming regulatory behaviors. The present research sheds light on the mediating role of persistence on the impact that older adults’ confidence in their memory abilities has on memory performance. Considering the role of motivation when examining memory test performance is important in the context of clinical diagnoses of memory de© 2017 Hogrefe

ficiencies and with respect to the construct validity of memory assessments in laboratory studies. Due to the small effect sizes observed in our studies, and in the literature in general, such consequences may essentially be expected for people who have mild cognitive symptomatology. The present research suggests that one easy-to-implement way of alleviating the influence of interindividual differences in MSE on memory performance may be to encourage patients and participants to persist during encoding. As well as being useful in ensuring that memory tests are less sensitive to older adults’ low MSE, this approach could also be used to design memory improvement programs that target the proximal motivational determinants of memory performance in order to increase trainees’ confidence in their memory abilities, which has been shown to increase the effectiveness of memory training (Bagwell & West, 2008).

References Bagwell, D. K., & West, R. L. (2008). Assessing compliance: Active versus inactive trainees in a memory intervention. Clinical Interventions in Aging, 3, 371–382. doi 10.2147/CIA.S1413 Bailey, H., Dunlosky, J., & Hertzog, C. (2009). Does differential strategy use account for age-related deficits in working memory performance? Psychology and Aging, 24, 82–92. doi 10.1037/a0014078 Bailey, H., Dunlosky, J., & Hertzog, C. (2010). Metacognitive training at home: Does it improve older adults’ learning? Gerontology, 56, 414–420. doi 10.1159/000266030 Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215. doi 10.1037/0033-295X.84.2.191 Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37, 122–147. doi 10.1037/0003-066X.37.2.122 Bandura, A. (1989). Regulation of cognitive processes through perceived self-efficacy. Developmental Psychology, 25, 729–735. doi 10.1037/0012-1649.25.5.729 Bandura, A. (1994). Self-efficacy. In V. S. Ramachandran (Ed.), Encyclopedia of human behavior (Vol. 4, pp. 71–81). New York: Academic Press. Bandura, A. (2003). Auto-efficacité: Le sentiment d’efficacité personnelle [Self-efficacy: The sense of personal efficacy]. Brussels, Belgium: De Boeck. Bandura, A. (2012). On the functional properties of perceived selfefficacy revisited. Journal of Management, 38, 9–44. doi 10.1177/0149206311410606 Beaudoin, M. (2008). Sentiment d’auto-efficacité mnésique et performances de mémoire [Memory self-efficacy and memory performance] (Unpublished doctoral dissertation). Université de Savoie, Chambéry, France. Beaudoin, M., Agrigoroaei, S., Desrichard, O., Fournet, N., & Roulin, J. L. (2008). Validation of the French version of the Memory SelfEfficacy questionnaire. R e v u e E u r o p é e n n e d e P s y c h o logie Appliquée/European Review of Applied Psyc h o l o g y , 5 8 , 165–176. doi 10.1016/j.erap.2007.09.001 Beaudoin, M., & Desrichard, O. (2011). Are memory self-efficacy and memory performance related? A meta-analysis. Psychological Bulletin, 137, 211–241. doi 10.1037/a0022106 Berry, J. M. (1999). Memory self-efficacy in its social cognitive context. In T. M. Hess & F. Blanchard-Fields (Eds.), Social cognition and aging (pp. 69–96). San Diego, CA: Academic Press. Berry, J. M., West, R. L., & Dennehey, D. M. (1989). Reliability and va-

Swiss Journal of Psychology (2017), 76 (1), 23–33

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

lidity of the Memory Self-Efficacy Questionnaire. Developmental Psychology, 25, 701–713. doi 10.1037/0012-1649.25.5.701 Bryan, J., Luszcz, M. A., & Pointer, S. (1999). Executive function and processing resources as predictors of adult age differences in the implementation of encoding strategies. Aging, Neuropsychology, and Cognition, 6, 273–287. doi 10.1076/13825585(199912)06:04;1-B;FT273 Burt, D. B., Zembar, M. J., & Niederehe, G. (1995). Depression and memory impairment: A meta-analysis of the association, its pattern, and specificity. Psychological Bulletin, 117, 285–305. doi 10.1037/0033-2909.117.2.285 Camp, C. J., Markley, R. P., & Kramer, J. J. (1983). Spontaneous use of mnemonics by elderly individuals. Educational Gerontology, 9, 57–71. doi 10.1080/0380127830090106 Cavanaugh, J. C., & Green, E. E. (1990). I believe, therefore I can: Self-efficacy beliefs in memory aging. In E. A. Lovelace (Ed.), Aging and cognition: Mental processes, self-awareness, and interventions (pp. 189–230). Amsterdam, Netherlands: North-Holland. Cervone, D., & Peake, P. K. (1986). Anchoring, efficacy, and action: The influence of judgmental heuristics on self-efficacy judgments and behavior. Journal of Personality and Social Psychology, 50, 492–501. doi 10.1037/0022-3514.50.3.492 Craik, F. I. M. (2000). Age-related changes in human memory. In D. C. Park & N. Schwarz (Eds.), Cognitive aging: A primer (pp. 75–92). Philadelphia, PA: Psychology Press. Damisch, L., Stoberock, B., & Mussweiler, T. (2010). Keep your fingers crossed! How superstition improves performance. Psychological Science, 21, 1014–1020. doi 10.1177/0956797610372631 Derogatis, L. R. (1977). SCL-90: Administration, scoring and procedures manual I for the R(evised) version. Baltimore, MD: John Hopkins University School of Medicine. Desrichard, O., & Köpetz, C. (2005). A threat in the elder: The impact of task-instructions, self-efficacy and performance expectations on memory performance in the elderly. European Journal of Social Psychology, 35, 537–552. doi 10.1002/ejsp.249 Devolder, P. A., & Pressley, M. (1992). Causal attributions and strategy use in relation to memory performance differences in younger and older adults. Applied Cognitive Psychology, 6, 629–642. doi 10.1002/acp.2350060706 Dunlosky, J., & Hertzog, C. (1998). Aging and deficits in associative memory: What is the role of strategy production? Psychology and Aging, 13, 597–607. doi 10.1037/0882-7974.13.4.597 Dunlosky, J., & Kane, M. J. (2007). The contribution of strategy use to working memory span: A comparison of strategy assessment methods. The Quarterly Journal of Experimental Psychology, 60, 1227–1245. doi 10.1080/17470210600926075 Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–198. doi 10.1016/0022-3956(75)90026-6 Fournet, N., Roulin, J.-L.,Vallet, F., Beaudoin, M., Agrigoroaei, S., Paignon, A., . . . Desrichard, O. (2012). Evaluating short-term and working memory in older adults: French normative data. Aging & Mental Health, 16, 922–930. doi 10.1080/13607863.2012.674487 Froger, C., Bouazzaoui, B., Isingrini, M., & Taconnat, L. (2012). Study time allocation deficit of older adults: The role of environmental support at encoding? Psychology and Aging, 27, 577–588. doi 10.1037/a0026358 Gardiner, M., Luszcz, M. A., & Bryan, J. (1997). The manipulation and measurement of task-specific memory self-efficacy in younger and older adults. International Journal of Behavioral Development, 21, 209–227. doi 10.1080/016502597384839 Gaultney, J. F., Kipp, K., & Kirk, G. (2005). Utilization deficiency and working memory capacity in adult memory performance: Not just for children anymore. Cognitive Development, 20, 205–213. doi 10.1016/j.cogdev.2005.02.001 Gendolla, G. H. E. (2000). On the impact of mood on behavior: An

Swiss Journal of Psychology (2017), 76 (1), 23–33

integrative theory and a review. Review of General Psychology, 4, 378–408. doi 10.1037//1089-2680.4.4.378 Germain, C. M., & Hess, T. M. (2007). Motivational influences on controlled processing: Moderating distractibility in older adults. Aging, Neuropsychology, and Cognition, 14, 462–486. doi 10.1080/13825580600611302 Hager, W., & Hasselhorn, M. (1992). Memory monitoring and memory performance: Linked closely or loosely? Psychological Research/Psychologische Forschung, 54, 110–113. doi 10.1007/ BF00937139 Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76, 408–420. doi 10.1080/03637750903310360 Henry, J. D., MacLeod, M. S., Phillips, L. H., & Crawford, J. R. (2004). A meta-analytic review of prospective memory and aging. Psychology and Aging, 19, 27–39. doi 10.1037/0882-7974.19.1.27 Hertzog, C., & Dixon, R. A. (1994). Metacognitive development in adulthood and old age. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 227–251). Cambridge, MA: MIT. Hertzog, C., Dixon, R. A., & Hultsch, D. F. (1990). Relationships between metamemory, memory predictions, and memory task performance in adults. Psychology and Aging, 5, 215–227. doi 10.1037/0882-7974.5.2.215 Hertzog, C., Lineweaver, T. T., & McGuire, C. L. (1999). Beliefs about memory and aging. In T. M. Hess & F. Blanchard-Fields (Eds.), Social cognition and aging (pp. 43–68). San Diego, CA: Academic Press. Hertzog, C., McGuire, C. L., Horhota, M., & Jopp, D. (2010). Does believing in “use it or lose it” relate to self-rated memory control, strategy use, and recall? International Journal of Aging & Human Development, 70, 61–87. doi 10.2190/AG.70.1.c Hertzog, C., McGuire, C. L., & Lineweaver, T. T. (1998). Aging, attributions, perceived control, and strategy use in a free recall task. Aging, Neuropsychology & Cognition, 5, 85–106. doi 10.1076/ anec.5.2.85.601 Hertzog, C., Price, J., & Dunlosky, J. (2008). How is knowledge generated about memory encoding strategy effectiveness? Learning and Individual Differences, 18, 430–445. doi 10.1016/j.lindif.2007.12.002 Hess, T. M. (2005). Memory and aging in context. Psychological Bulletin, 131, 383–406. doi 10.1037/0033-2909.131.3.383 Howell, D. C. (1998). Méthodes statistiques en sciences humaines [Statistical methods for the social sciences]. Paris, France: De Boeck. Jacobs, B., Prentice-Dunn, S., & Rogers, R. W. (1984). Understanding persistence: An interface of control theory and self-efficacy theory. Basic & Applied Social Psychology, 5, 333–347. doi 10.1207/s15324834basp0504_6 Kalska, H., Punamaki, R.-L., Makinen-Pelli, T., & Saarinen, M. (1999). Memory and metamemory functioning among depressed patients. Applied Neuropsychology, 6, 96–107. doi 10.1207/ s15324826an0602_5 Kanfer, R., Wanberg, C. R., & Kantrowitz, T. M. (2001). Job search and employment: A personality-motivational analysis and meta-analytic review. Journal of Applied Psychology, 86, 837–855. doi 10.1037/0021-9010.86.5.837 Kizilbash, A. H., Vanderploeg, R. D., & Curtiss, G. (2002). The effects of depression and anxiety on memory performance. Archives of Clinical Neuropsychology, 17, 57–67. doi 10.1016/S08876177(00)00101-3 Lachman, M. E., Andreoletti, C., & Pearman, A. (2006). Memory control beliefs: How are they related to age, strategy use and memory improvement? Social Cognition, 24, 359–385. doi 10.1521/soco.2006.24.3.359 Lachman, M. E., Bandura, M., Weaver, S. L., & Elliott, E. (1995). Assessing memory control beliefs: The Memory Controllability In-

M. Beaudoin & O. Desrichard: Memory Self-Efficacy and Task Persistence

ventory. Aging & Cognition, 2, 67–84. doi 10.1080/ 13825589508256589 Lachman, M. E., Neupert, S. D., & Agrigoroaei, S. (2011). The relevance of control beliefs for health and aging. In K. W. Schaie & S. L. Willis (Eds.), Handbook of the psychology of aging (7th ed., pp. 175–190). New York: Elsevier. La Rue, A., Small, G., McPherson, S., Komo, S., Matsuyama, S. S., & Jarvik, L. F. (1996). Subjective memory loss in age-associated memory impairment: Family history and neuropsychological correlates. Aging, Neuropsychology, and Cognition, 3, 132–140. doi 10.1080/13825589608256618 Ledgerwood, A., & Shrout, P. E. (2011). The trade-off between accuracy and precision in latent variable models of mediation processes. Journal of Personality and Social Psychology, 101, 1174–1188. doi 10.1037/a0024776 Luszcz, M. A. (2011). Executive function and cognitive aging. In K. W. Schaie & S. L. Willis (Eds.), Handbook of the psychology of aging (7th ed., pp. 59–72). San Diego, CA: Elsevier. Marquez, D. X., Jerome, G. J., McAuley, E., Snook, E. M., & Canaklisova, S. (2002). Self-efficacy manipulation and state anxiety responses to exercise in low active women. Psychology & Health, 17, 783–791. doi 10.1080/0887044021000054782 McDonald-Miszczak, L., Hertzog, C., & Hultsch, D. F. (1995). Stability and accuracy of metamemory in adulthood and aging: A longitudinal analysis. Psychology and Aging, 10, 553–564. doi 10.1037/0882-7974.10.4.553 McDougall, G. J., Jr. (2009). A framework for cognitive interventions targeting everyday memory performance and memory self-efficacy. Family & Community Health, 32(Suppl. 1), S15–S26. doi 10.1097/01.FCH.0000342836.20854.fb Multon, K. D., Brown, S. D., & Lent, R. W. (1991). Relation of self-efficacy beliefs to academic outcomes: A meta-analytic investigation. Journal of Counseling Psychology, 38, 30–38. doi 10.1037/0022-0167.38.1.30 Murphy, M. D., Sanders, R. E., Gabriesheski, A. S., & Schmitt, F. A. (1981). Metamemory in the aged. Journal of Gerontology, 36, 185–193. doi 10.1093/geronj/36.2.185 Narciss, S. (2004). The impact of informative tutoring feedback and self-efficacy on motivation and achievement in concept learning. Experimental Psychology, 51, 214–228. doi 10.1027/16183169.51.3.214 Pariente, P., & Guelfi, J. D. (1990). Inventaires d’auto-évaluation de la psychopathologie chez l’adulte. 1ère partie: Inventaires multidimensionnels [Self-report symptom inventories for adults. I. Multidimensional questionnaires]. Psychiatrie & Psychobiologie, 5(1), 49–63. Park, D. C. (2000). The basic mechanisms accounting for age-related decline in cognitive function. In D. C. Park & N. Schwarz (Eds.), Cognitive aging: A primer (pp. 3–21). Philadelphia, PA: Psychology Press. Pelegrina, S., Bajo, M. T., & Justicia, F. (1999). Allocation of time in self-paced memory tasks: The role of practice, instructions, and individual differences in optimizing performance. Learning & Individual Differences, 11, 401–429. doi 10.1016/S10416080(99)80011-9 Preacher, K. J., & Hayes, A. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717–731. doi 10.3758/BF03206553 Preacher, K. J., & Hayes, A. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. doi 10.3758/BRM.40.3.879 Preacher, K. J., & Kelley, K. (2011). Effect size measures for mediation models: Quantitative strategies for communicating indi-

rect effects. Psychological Methods, 16, 93–115. doi 10.1037/a0022658 Preacher, K. J., & MacCallum, R. C. (2003). Repairing Tom Swift’s electric factor analysis machine. Understanding Statistics, 2, 13–43. doi 10.1207/S15328031US0201_02 Rahhal, T. A., Hasher, L., & Colcombe, S. J. (2001). Instructional manipulations and age differences in memory: Now you see them, now you don’t. Psychology and Aging, 16, 697–706. doi 10.1037/0882-7974.16.4.697 Salancik, G. R., & Conway, M. (1975). Attitude inferences from salient and relevant cognitive content about behavior. Journal of Personality and Social Psychology, 32, 829–840. doi 10.1037/00223514.32.5.829 Salthouse, T. A. (1991). Mediation of adult age differences in cognition by reductions in working memory and speed of processing. Psychological Science, 2, 179–183. doi 10.1111/j.1467-9280. 1991.tb00127.x Signoret, J. L. (1991). Batterie d’Efficience Mnésique (BEM 144) [Memory Efficiency Battery (BEM 144)]. Paris, France: Elsevier. Souchay, C., & Isingrini, M. (2004). Age related differences in metacognitive control: Role of executive functioning. Brain and Cognition, 56, 89–99. doi 10.1016/j.bandc.2004.06.002 Stine-Morrow, E. A. L., Shake, M. C., Miles, J. R., & Noh, S. R. (2006). Adult age differences in the effects of goals on self-regulated sentence processing. Psychology and Aging, 21, 790–803. doi 10.1037/0882-7974.21.4.790 Symons, C. S., & Johnson, B. T. (1997). The self-reference effect in memory: A meta-analysis. Psychological Bulletin, 121, 371–394. doi 10.1037/0033-2909.121.3.371 Taconnat, L., Baudouin, A., Fay, S., Clarys, D., Vanneste, S., Tournelle, L. & Isingrini, M. (2006). Aging and implementation of encoding strategies in the generation of rhymes: The role of executive functions. Neuropsychology, 20, 658–665. doi 10.1037/08944105.20.6.658 Vallet, F. (2012). Comment évalue-t-on l’efficience de notre mémoire: Le rôle des attributions causales et des théories naïves [How do people assess their memory efficiency: The role of causal attributions and lay theories] (Unpublished doctoral dissertation). University of Savoy, Chambéry, France. Vallet, F., Agrigoroaei, S., Beaudoin, M., Fournet, N., Paignon, A., Roulin, J.-L., & Desrichard, O. (2015). Older adults’ beliefs about forgetting and aging predict memory self-efficacy above and beyond actual memory performance and mental health. International Review of Social Psychology, 28(4), 57–79. Vanier, M., & Lemyze, C. (1994). Rivermead Behavioral Memory Test (2nd ed.) [Measurement instrument]. Bury St. Edmunds, UK: Thames Valley Test Company. Wechsler, D. (2001). MEM-III: Manuel de l’Echelle Clinique de Mémoire [WMS-III: Wechsler Memory Scale] (3rd ed.) [Measurement instrument]. Paris, France: Les Editions du Centre de Psychologie Appliquée. Wells, G. D., & Esopenko, C. (2008). Memory self-efficacy, aging, and memory performance: The roles of effort and persistence. Educational Gerontology, 34, 520–530. doi 10.1080/ 03601270701869386 West, R. L., Bagwell, D. K., & Dark-Freudeman, A. (2008). Self-efficacy and memory aging: The impact of a memory intervention based on self-efficacy. Aging, Neuropsychology, and Cognition, 15, 302–329. doi 10.1080/13825580701440510 Marine Beaudoin LIPPC2S University of Savoy Mont Blanc 73000 Chambéry France marine.beaudoin@univ-smb.fr

Swiss Journal of Psychology (2017), 76 (1), 23–33

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.

Original Communication

Psychometric Qualities of the Creative Process Engagement Scale in a Malaysian Undergraduate Sample Chee-Seng Tan1 and Timothy Teo2 1

Department of Psychology and Counseling, Universiti Tunku Abdul Rahman, Kampar, Malaysia

Faculty of Education, University of Macau, Taipa, Macau Abstract. The present study examines the psychometric properties of the Creative Process Engagement Scale (CPES) among Malaysian undergraduates. A total of 377 undergraduates whose ages ranged from 18 to 43 years participated in the study and were presented with the CPES and self-perceived creativity. Confirmatory factor analyses supported the hierarchical four-factor structure of the CPES, which consisted of three first-order factors and one second-order factor. In addition, we found that the CPES has sound internal consistency as well as criterion-related validity. Furthermore, the results of measurement invariance testing supported the hypothesis of equivalent factor loading and intercepts for this hierarchical four-factor structure across gender. Latent mean analysis revealed that female students showed less engagement in creativity-related activities than male students. The findings shed light on the psychometric qualities of the CPES and confirm that the CPES measures involvement in creative activities across gender in undergraduate students. Keywords: creative process engagement, creativity, confirmatory factor analysis, internal consistency, validity, Malaysia

Decades of research have consistently shown a link between creativity and many positive outcomes, such as job performance (Zhang & Bartol, 2010b), problem-solving ability (Tan & Hashim, 2009), and short-term mating success (Beaussart, Kaufman, & Kaufman, 2012). For instance, Tan and Hashim (2009) found a positive correlation between self-perceived creativity and problem-solving ability among undergraduates in Malaysia. Moreover, students with high problem-solving ability also reported a low level of stress. While the majority of studies have focused on creative outcomes, little attention has been given to the creative process (Shalley & Zhou, 2008). Consequently, little is known about the impact of creative process engagement and especially the procedures and behaviors required to achieve creative outcomes. According to the classical view proposed by Wallas (1926), the creative thinking process consists of four stages: preparation, incubation, illumination, and verification. Similarly, Amabile (1983) indicated that the three stages of the creative process are problem or task presentation, preparation, and response generation. Although a number of models have been proposed to account for cognitive processes associated with creativity, Reiter-Palmon and Illies (2004) indicated that it is generally agreed upon that problem identification and construction, information search and encoding, solution generation, and idea evaluation are the core processes for generating creative outcomes. Research has shown that engaging in the creative process plays a central role in the generation of creative outcomes (e.g., Gilson & Shalley, 2004). In her compo© 2017 Hogrefe

nential framework of creativity, Amabile (1983) proposed that level of creativity is a function of domain-relevant skills, creativity-relevant skills, and task motivation. Specifically, these three factors influence one’s engagement in problem identification, information searching, and idea generation, respectively. In other words, the quality of creative outcomes relies on the level of engagement in creative process. Indeed, Amabile (1983) argued that engaging in creative process activities has as great an influence on creative behaviors as intrinsic motivation. On the basis of the notion proposed by Amabile (1983) and Reiter-Palmon and Illies (2004), Zhang and Bartol (2010a) defined creative process engagement as the extent to which an individual engages in creativity-relevant methods or processes. This involvement can be indexed by three dimensions: problem identification, information searching, and idea/alternative generation. Problem identification refers to the construction of a problem by means of identification of the goals, procedures, limitations, and information necessary to solve the problem. Information searching reflects the processes of connecting, integrating, and encoding information that is useful for creative production, while idea generation is the stage at which alternative solutions or outcomes are produced. It is noteworthy that creative process engagement is conceptually different from the standard problem solving. One of the distinctions between the two constructs is that creative process engagement consists of searching, encoding, and reorganizing while problem solving requires recall and using previously acquired procedures and solutions (Lubart, 2001). To facilitate a better understanding Swiss Journal of Psychology (2017), 76 (1), 35–42 DOI 10.1024/1421-0185/a000189

of the creative process, Zhang and Bartol (2010a) developed the Creative Process Engagement Scale (CPES). Zhang and Bartol (2010a) administered the CPES to 498 professional-level employees to investigate the underlying mechanism of the relationship between empowering leadership and employee creativity. They found that creative process engagement not only significantly predicted creativity, but also mediated the relationship between empowering leadership and employee creativity and that between intrinsic motivation and creativity. Moreover, confirmatory factor analysis (CFA) on the CPES supported the model of three first-order factors and one second-order factor and showed good internal consistency (Zhang & Bartol, 2010a). In a follow-up study on the relationship between creative process engagement and job performance, Zhang and Bartol (2010b) administered an online survey, including the CPES, to employees who worked in complex jobs requiring high creativity (e.g., software engineers). Based on the responses from 367 employees, Zhang and Bartol found a curvilinear (i.e., an inverted-U shaped) relationship between creative process engagement and employee job performance. In particular, individuals have the highest job performance when creative process engagement levels are moderate rather than low or high. This is because, when creative process engagement level is low, people tend to have low activation and put little effort into the task. When creative process engagement level is high, people tend to have high activation, which results in difficulties when their limited attention capacity and cognitive resources do not allow them to cope with the high task demands. The CPES has also been used as a proxy for creativity. To, Fisher, Ashkanasy, and Rowe (2012) investigated the impact of mood on creativity in a sample of employees using the shortened, six-item CPES (i.e., two items for each dimension) to reduce participants’ response burden. To and colleagues found that the shortened scale was highly correlated with the original CPES (r = .95). Although past studies provided support for the CPES as a psychometrically sound tool for assessing one’s engagement in the creativity-related process, these studies were mainly conducted on employees (e.g., Zhang & Bartol, 2010a, 2010b). Because perception of creativity varies from one population to another (Mumford, Whetzel, & Reiter-Palmon, 1997; Tan & Viapude, 2015; Wang & Greenwood, 2013; Yue, Bender, & Cheung, 2011), it is necessary to examine the psychometric properties of the CPES (e.g., factor structure) in other populations and in different contexts (e.g., education). In addition, we urgently need a psychometrically sound tool for measuring creative process engagement to understand the impact of creative process engagement on students’ creativity. Studies on academic performance have shown that academic engagement is positively associated with academic achievement (e.g., Hughes & Coplan, 2010). It is, therefore, reasonable to assume that creative process engagement is conducive to students’ creativity. Nevertheless, it is difficult to examine the hypothetical relationship between creative process engagement and creativity withSwiss Journal of Psychology (2017), 76 (1), 35–42

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

out an evidence-supported measurement of students’ creative process engagement. The present study aims to evaluate the usability of the CPES in the educational context by examining the psychometric properties of the CPES in a sample of Malaysian undergraduates. Specifically, CFA was used to assess the factor structure of the CPES. Moreover, the evidence based on relationships between the CPES and other variables was examined by testing the association between the CPES and self-report creativity. Theoretically, individuals who are active in creative processes, such as searching for information from multiple sources, consider questions from different perspectives and, hence, are more likely to generate creative ideas. In addition, self-report scales for assessing creativity are reliable and valid (Silvia, Wigert, ReiterPalmon, & Kaufman, 2012).

Method Participants and Procedure A group of 380 undergraduates from two universities in Malaysia participated in this study in exchange for either course credit or five Malaysian ringgit (US$1.40). Because initial analyses in the two groups showed support for the theoretical second-factor model, data from the two groups were combined for further analysis. Because of unexpected technical problems, three participants failed to complete the survey and were excluded from further analysis, resulting in a sample of 377 participants, of which 256 were female (67.9%) and 120 were male (31.8%) (one missing value). The participants were 18 to 43 years of age (M = 21.00, SD = 1.90); one participant did not report her age. The sample consisted mainly of Chinese (89.1%), followed by Malays (5.3%), Indians (4.0%), and others (1.6%). Of the participants, 35.3% were freshmen, 46.4% were sophomores, 14.9% were juniors, and 3.4% were seniors; they had different majors (e.g., psychology, education, marketing, management). The participants answered the online questionnaire in groups of five to 20 in a computer laboratory. Anonymity was assured because the researchers were not able to identify participants through their responses. On average, it took 20 minutes for the participants to complete the survey.

Measurements Creative Process Engagement Scale (CPES) The CPES (Zhang & Bartol, 2010a) consisted of three dimensions: problem identification (three items), information searching (three items), and idea generation (five items). Participants responded on a 5-point Likert scale, ranging from 1 (never) to 5 (very frequently). Sample items for the problem identification, © 2017 Hogrefe

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

information searching, and idea generation subscales are “I think about the problem from multiple perspectives,” “I search for information from multiple sources,” and “I generate a significant number of alternatives to the same problem before I choose the final solution,” respectively. A composite score was derived by averaging the item scores. A higher score indicated more engagement in the creativity process. The internal consistency values are discussed in the results section.

Self-Perceived Creativity (Zhou & George, 2001) This 13-item scale was initially designed for supervisors to rate their subordinates’ creativity. We modified the items so that respondents could evaluate their own creativity. The participants reported the extent to which they thought they were creative by responding to the items on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). A sample item was “I am a good source of creative ideas.” Item scores were summed up to generate a total score, whereby higher scores indicate a higher level of creativity. The internal consistency of the scale was good, w = .90; the 95% bias-corrected confidence interval based on 10,000 bootstrap samples was 95% CI [.88, .92].

Results Confirmatory Factor Analyses The means and standard deviations for the items of the CPES ranged from 3.39 to 3.77 and from 0.77 to 0.87, respectively. The skewness of the CPES items ranged from –0.32 to 0.03, and kurtosis ranged from –0.47 to –0.03. The data are deemed to be normally distributed because the absolute values of the skewness and kurtosis scores are less than 3 and 8, respectively (Kline, 2011). We assessed the factor structure of the CPES in our sample and compared it to the suggested structure using CFA with maximum likelihood (ML) estimation. The fitness of the model was examined with several indices, including model χ², Tucker-Lewis index (TLI), comparative fit index (CFI), root-mean square error of approximation (RMSEA), and the standardized rootmean-square residual (SRMR). A model is considered a poor fit if the χ² value is large and statistically significant. Given that the χ² value is highly sensitive to sample size, the ratio of χ² value divided by degrees of freedom has been used as an index of model fit. A ratio below 3 is considered acceptable (Tabachnick & Fidell, 2007). For the TLI and CFI, values greater than .95 indicate a good fit, and values .90 are considered an accept-

able fit. The RMSEA value should be less than .05 for good model fit, but values less than .08 are acceptable. The SRMR should be less than .08 for good model fit (Hu & Bentler, 1999). The theoretical model of three first-order factors and one second-order factor (Model 1)1 was examined first. The loadings of the first item on each first-order factor (Items 1, 4, 7), as well as the loading on Problem Identification, were fixed to 1 for purposes of model identification. Model 1 yielded acceptable fit statistics (see Table 1) with the exception of the ratio of χ² value divided by degrees of freedom. Examination of modification indices suggested adding one error covariance between Item 9 and Item 10. Despite the modification resulting in a better fit (see Model 1a), the model was not respecified because the modification was not justified by strong theoretical rationale. In addition, we examined three two-factor models to determine whether these alternative models were better than the theoretical model. In each of the two-factor models, we kept one of the three subdimensions of the CPES and combined the other two subdimensions. In Model 2a, for example, the Problem Identification factor was retained while items of the Information Searching and Idea Generation factors were combined to load on a single factor. Similarly, Model 2b consisted of the Information Searching factor and a composite factor, while Model 2c consisted of the Idea Generation factor and a composite factor. We also examined the one-factor model with 11 items (Model 3) and the one-factor model with six items (Model 3a) shown in a study by To and colleagues (2012). The model statistics for Model 2b, Model 2c, Model 3, and Model 3a indicated a poor fit with the data, though three fit indices (i.e., TLI, CFI, & SRMR) for Model 2b, Model 2c, and Model 3 were acceptable. Model 2a, on the other hand, showed acceptable fit statistics except for the ratio of χ² value divided by degrees of freedom, suggesting that both Model 1 and Model 2a are acceptable. However, compared to Model 2a, the fit statistics of Model 1 were closer to the suggested values. Moreover, Model 1 was found to have lower Akaike information criterion (AIC) and Bayes information criterion (BIC) values than Model 2a: AIC = 176.87 and BIC = 275.17 for Model 1; AIC = 185.14 and BIC = 275.58 for Model 2a. Taken together, the results indicated that Model 1 (see Figure 1) was the best fitting model.

Internal Consistency of the CPES McDonald’s omega was used to assess the internal consistency of the hierarchical four-factor structure of the CPES, composed of 3 first-order factors and one second-order factor with no error covariance. We found that the CPES and the three subscales had good internal consistency: .89, 95% CI [.87–.91] for overall CPES; .72, 95% CI [.66–.77] for Problem Identification;

1 We did not test a model that consists of 3 first-order factors that are intercorrelated because the three covariances between the factors can be reflected by the three higher-order loadings in Model 1. In other words, the three-correlated-factor model is empirically equivalent to Model 1.

Swiss Journal of Psychology (2017), 76 (1), 35–42

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

Table 1. Goodness-of-fit indices for the Creative Process Engagement Scale (N = 377) Model

χ²

χ²/df

TLI

CFI

RMSEA [90% CI]

SRMR

126.87***

3.09

.929

.947

.075 [.060, .090]

.043

102.28***

2.56

.947

.961

.064 [.048, .080]

.040

139.14***

3.24

.924

.940

.077 [.063, .092]

.046

160.99***

3.74

.906

.927

.085 [.072, .100]

.049

157.48***

3.66

.909

.929

.084 [.070, .098]

.048

173.18***

3.94

.900

.920

.088 [.075, .102]

.051

53.30***

5.92

.875

.925

.114 [.086, .145]

.049

Note. Model 1 = model of 3 first-order factors and one second-order factor; Model 1a = modified Model 1 with one error covariance added; Model 2a = model of two factors with problem identification factor and a composite factor; Model 2b = model of two factors with information searching factor and a composite factor; Model 2c = model of two factors with idea generation factor and a composite factor; Model 3: model of a single factor; Model 3a = model of a single factor with six items; TLI = Tucker-Lewis index; CFI = comparative fit index; RMSEA = root-mean-square error of approximation; SRMR = standardized rootmean-square residual. ***p < .001.

Figure 1. Path diagram with standardized factor loadings for Model 1 (3 first-order factors and 1 second-order factor model) and error variances. All factor loadings were significant at p < .001 level. CPE = Creative Process Engagement; Problem = Problem Identification; Information = Information Searching; Idea = Idea Generation.

Swiss Journal of Psychology (2017), 76 (1), 35–42

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

Table 2. Interitem correlation matrix for creative process engagement scale Item

1.000

.487

1.000

.464

.427

1.000

.398

.381

.486

1.000

.373

.399

.366

.508

1.000

.362

.412

.382

.482

.522

1.000

.380

.397

.369

.444

.494

.590

1.000

.406

.424

.347

.390

.520

.558

.566

1.000

.351

.477

.375

.386

.406

.445

.425

.405

1.000

.333

.353

.344

.315

.340

.406

.380

.541

1.000

.378

.435

.368

.337

.448

.459

.493

.514

.512

.527

.75, 95% CI [.70–.79] for Problem Searching; .82, 95% CI [.78–.95] for Idea Generation. In addition, consistent with the CFA results, the interitem correlation matrix (see Table 2) showed high correlations between Item 9 and Item 10. Furthermore, the correlations between Item 9 and the other items were slightly higher than those between Item 10 and the other items.

Validity of the CPES To examine the validity of the CPES, we computed correlations among the CPES overall score and subscales and self-perceived creativity. The results showed that the CPES overall score was significantly related to self-perceived creativity (r = .63, p < .001). Consistent with the relevant literature (e.g., Zhang & Bartol, 2010a, 2010b), significant relationships were also observed between self-perceived creativity and each of the CPES dimensions: problem identification, r = .47, p < .001; information searching, r = .53, p < .001; idea generation, r = .62, p < .001. We also conducted a stepwise multiple linear regression to examine the extent to which the three dimensions of CPES predicted self-perceived creativity. We found that idea generation (β = .50, p < .001) and information searching (β = .18, p = .002), but not problem identification, significantly predicted self-perceived creativity. The two-predictor model was able to account for 40.1% of the total variance of self-perceived creativity, F(2, 374) = 126.84, p < .001. Taken together, the results suggest that CPES is a valid measure of engagement in the creative process.

Measurement Invariance Although our findings lend support to the theoretical model of the CPES, it was not clear whether the hierarchical four-factor structure of the CPES is equivalent across gender. The second goal of the study, therefore, was to investigate measurement invariance of the structure of the CPES across gender. Testing © 2017 Hogrefe

1.000

factorial invariance was conducted with Amos 21 (Arbuckle, 2012) in a four-step process following the procedures suggested by Byrne and Stewart (2006) and Chen, Sousa, and West (2005). First, we identified the baseline model for the two gender groups. Then, we began the analyses by examining configural invariance, whereby the same factor structure was tested simultaneously in male and female students without placing constraints on the parameters (Model 1). If equal factor structure is established, the next step is to compare the configural invariance model with a more stringent model in which factor loadings are constrained to be equal across gender. Specifically, Model 2 constrained the loadings of the measured variables on the first-order factors to be equal across groups, while Model 3 imposed equality constraints on both first-order and secondorder factor loadings. If equal factor loadings are established, the last step is to compare the fit of Model 3 to the fit of a model in which the intercepts of measured variables are constrained to be equal (Model 4). Because there were 14 intercepts (11 intercepts for measured variables and 3 intercepts for first-order factors), but only 11 measured variables, the intercepts of the first-order factors were not constrained but fixed to zero for purposes of model identification (Byrne & Stewart, 2006). Finally, the latent mean difference was investigated if invariance of factor loadings and intercepts were supported. We did not test equivalence in residual and factor covariance before testing latent mean difference because “the covariances and residuals have been factored out of the latent means before they are compared” (Guo, Suarez-Morales, Schwartz, & Szapocznik, 2012, p. 59). The models were compared using two widely used criteria: the likelihood ratio test and CFI. The likelihood ratio test compares two nested models through the difference of the χ² values and degrees of freedom. A statistically significant χ² difference value (Δχ²) indicates a decrement in fit, that is, the constraints imposed on the restrictive model are not consistent with data. In contrast, the measurement invariance is supported if the Δχ² is not statistically significant. Similarly, the ΔCFI compares two models by computing the difference in the CFI (ΔCFI) of the Swiss Journal of Psychology (2017), 76 (1), 35–42

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

Table 3. Goodness-of-fit indices for Tests of Invariance of Creative Process Engagement Scale Hierarchical Structure for Men and Women Model and description

χ²

CFI

Model comparison

Δχ²

ΔCFI

Baseline model Male

69.033**

.931

–

Female

96.992***

.953

–

Model 1: Configural invariance

166.097***

.947

–

Model 2: First-order factor loadings invariant

175.406***

.946

2 vs. 1

9.309

.32

.001

Model 3: First- and second-order factor loadings invariant

176.467***

.947

3 vs. 2

1.061

.59

.001

Model 4: First- and second-order factor loadings and intercepts of measured variables invariant

196.477***

103

.941

4 vs. 3

.045

.006

20.01

Note. CFI = comparative fit index. **p < .01, ***p < .001.

models. Measurement invariance is supported when the absolute value of ΔCFI is less than 0.01 (Cheung & Rensvold, 2002). Because Δχ² is sensitive to sample size, the decision is based on CFI when the results of the two criteria contradict each other. Table 3 shows the results of the tests of measurement invariance. First, the testing of the four-factor hierarchical model, consisting of three first-order factors and one second-order factor, yielded an acceptable fit in men and women, respectively. Therefore, this model was used as the baseline model. The results for Model 1, which tested the baseline model simultaneously in men and women, indicated an acceptable fit model. The results support configural invariance. The results for Model 2, in which the first-order factor loadings were constrained to be equal, suggested a well-fitting model. Similarly, the results for Model 3, which tested equality for second-order factor loading, yielded a well-fitting model. The results suggest that equalities are established for the first- and second-order factor loadings. Finally, the results for Model 4, which imposed constraints on the intercepts of the measured variables, indicated a deterioration in model fit, showing a significant, large increase in value. The ΔCFI, however, was below the recommended value for rejecting the hypothesis of measurement invariance. Therefore, we concluded that the item intercepts were equal. To examine the latent mean difference, male students served as the reference group and their latent factor means were set to zero, whereas the latent means in females were freely estimated. We found that females scored significantly lower than males on creative process engagement, estimated mean = – 0.18, SE = 0.06, p = .001.

Discussion Previous creativity studies focused more on the outcome than on the role of engagement in the creativity-relevant process (e.g., Shalley & Zhou, 2008). Zhang and Bartol (2010a) developed the CPES to investigate the role of creative process engagement in creative performance. Although past studies on Swiss Journal of Psychology (2017), 76 (1), 35–42

employees found the CPES to be psychometrically sound, the validity of the CPES has been underinvestigated in the educational context. The present study is the first to examine the psychometric properties of the CPES in a Malaysian undergraduate sample. Compared to the alternative models (i.e., two-factor models and single-factor models), our results support the hypothesis that the CPES consists of 3 first-order factors (i.e., Problem Identification, Information Searching, and Idea Generation) and a general factor, though the χ² to degrees of freedom ratio is slightly beyond the suggested cut-off value. Indeed, Model 1 (i.e., hierarchical four-factor model) is considered the best fit model, although the two-factor Model 2a also showed acceptable fit to the data. This is because Model 1 is not only theoretically supported but the fit statistics, including the AIC and BIC, are also better than those for Model 2a. More studies, however, are needed to further investigate the fit of the two models. The inconsistent fit index of the theoretical model (i.e., Model 1) was resolved and model fit was improved after adding error covariance between Item 9 and Item 10. This suggests that Item 9 (“I generate a significant number of alternatives to the same problem before I choose the final solution”) and Item 10 (“I try to devise potential solutions that move away from established ways of doing things” may overlap conceptually, at least in our sample. The model, however, was not respecified because the modification was not supported by theoretical rationale. Future studies should reexamine the performance of Item 9 and Item 10 to avoid item redundancy and to improve model fit. Indeed, the interitem correlation matrix shows that Item 9 was perceived as being more relevant to other items than Item 10. Therefore, one of the possibilities for improving model fit is to remove Item 10. The present study contributes to the literature by providing empirical support for the CPES as a useful measurement for educational purposes. Specifically, the present study builds on previous studies using the CPES by extending its use with undergraduates. Further studies, however, are needed to replicate the findings of the present study in other populations and traits across diverse settings (e.g., adolescents’ artistic performance). Furthermore, it is noteworthy that creative process may vary © 2017 Hogrefe

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

from one domain to another (e.g., Botella et al., 2013). Bourgeois-Bougrine et al. (2014) interviewed 22 recognized French screenplay writers and found that the creative process of screenplay writing consists of three general phases: impregnation (e.g., understanding the demand of sponsors and motivation behind the film), structuration (e.g., writing synopsis or outline), and production (e.g., writing and rewriting script). The differences imply that the CPES may not be able to fully capture engagement in the creative process of other domains and the possibility that the CPES may be more suitable (i.e., have higher validity) for one domain than others. Nonetheless, studies have found that creativity is both domain-general and domain-specific (Tan & Qu, 2012). In other words, there are also similarities across different domains of creativity. For instance, the production phase of screenplay writing is conceptually similar to the idea generation dimension of the CPES. It is, therefore, believed that the CPES can be the blueprint for measurement of engagement in the creative process of different domains. Researchers may modify the CPES items according to the domain of study. The positive associations between each dimension of CPES and self-perceived creativity provide preliminary support for the notion that CPES could be used to assess engagement in creativity-related processes in undergraduates. Interestingly, regression analysis showed that only information searching and idea generation are predictive of creativity, which implies that engaging in information searching and idea generation is more important than problem identification for producing creative responses. More studies are needed to replicate the findings and investigate the reason problem identification had no impact on self-perceived creativity. It is also essential to examine the validity of the CPES more closely by assessing the role of the three CPES factors in alternative criterion measures, such as the Creative Achievement Questionnaire (Carson, Peterson, & Higgins, 2005) and the Alternate Uses Task (Wallach & Kogan, 1965), in future studies. Furthermore, we recommend that researchers extend the scope of investigation by examining the discriminant validity of the CPES as well as the mechanism underlying the relationship between engagement and creativity. The most important finding of the present study is the evidence for measurement invariance. To our knowledge, this study is the first to examine the measurement invariance of the CPES across gender in a student sample. Our results showed provisional evidence of equal factor structure, factor loadings, and item intercepts across gender. However, we found an inconsistency between the χ² test and the difference in CFI when testing intercept equivalence. Although the discrepancy could be due to the sensitivity of χ² to sample size, future studies should investigate whether the findings can be replicated. In addition, the measurement invariance suggests that it is important to examine mean differences between groups. In our sample, female students showed lower engagement in creativity-related activities than male students. Future studies should investigate the potential factors that contribute to the gender © 2017 Hogrefe

differences in creativity process engagement. In addition, future studies should address equivalence across cultures. In summary, the present study provides evidence that the CPES has good psychometric properties and cross-gender measurement invariance. Future studies should use the CPES to assess and compare individual engagement in creativity-relevant activities between genders.

Acknowledgments The data used in this study were part of a larger project that was funded by a grant awarded to the first author by the Universiti Tunku Abdul Rahman Research Fund (IPSR/RMC/ UTARRF/2014-C1/T02). We thank Ms. Yian-Thin Kung for helping with data collection.

References Amabile, T. M. (1983). The social psychology of creativity: A componential conceptualization. Journal of Personality and Social Psychology, 45, 357–376. doi 10.1037/0022-3514.45.2.357 Arbuckle, J. L. (2012). Amos (Version 21.0) [Computer Program]. Chicago, IL: IBM SPSS. Beaussart, M. L., Kaufman, S. B., & Kaufman, J. C. (2012). Creative activity, personality, mental illness, and short-term mating success. The Journal of Creative Behavior, 46, 151–167. doi 10.1002/jocb.11 Botella, M., Glaveanu, V., Zenasni, F., Storme, M., Myszkowski, N., Wolff, M., & Lubart, T. (2013). How artists create: Creative process and multivariate factors. Learning and Individual Differences, 26, 161–170. doi 10.1016/j.lindif.2013.02.008 Bourgeois-Bougrine, S., Glaveanu, V., Botella, M., Guillou, K., De Biasi, P. M., & Lubart, T. (2014). The creativity maze: Exploring creativity in screenplay writing. Psychology of Esthetics, Creativity, and the Arts, 8, 384–399. doi 10.1037/a0037839 Byrne, B. M., & Stewart, S. M. (2006). The MACS approach to testing for multigroup invariance of a second-order structure: A walk through the process. Structural Equation Modeling: A Multidisciplinary Journal, 13, 287–321. doi 10.1207/s15328007sem1302_7 Carson, S. H., Peterson, J. B., & Higgins, D. M. (2005). Reliability, validity, and factor structure of the Creative Achievement Questionnaire. Creativity Research Journal, 17, 37–50. doi 10.1207/s15326934crj1701_4 Chen, F. F., Sousa, K. H., & West, S. G. (2005). Testing measurement invariance of second-order factor models. Structural Equation Modeling: A Multidisciplinary Journal, 12, 471–492. doi 10.1207/s15328007sem1203_7 Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9, 233–255. doi 10.1207/S15328007SEM0902_5 Gilson, L. L., & Shalley, C. E. (2004). A little creativity goes a long way: An examination of teams’ engagement in creative processes. Journal of Management, 30, 453–470. doi 10.1016/j.jm.2003.07.001 Guo, X., Suarez-Morales, L., Schwartz, S. J., & Szapocznik, J. (2012). Some evidence for multidimensional biculturalism: Confirmatory factor analysis and measurement invariance analysis on the Bicultural Involvement Questionnaire – Short Version. Journal of Latina/o Psychology, 1(S), 52–65. doi 10.1037/2168-1678.1.s.52

Swiss Journal of Psychology (2017), 76 (1), 35–42

Hu, L.-t., & 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 Hughes, K., & Coplan, R. J. (2010). Exploring processes linking shyness and academic achievement in childhood. School Psychology Quarterly, 25, 213–222. doi 10.1037/a0022070 Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New York: Guilford. Lubart, T. I. (2001). Models of the creative process: Past, present, and future. Creativity Research Journal, 13, 295–308. doi 10.1207/S15326934CRJ1334_07 Mumford, M. D., Whetzel, D. L., & Reiter-Palmon, R. (1997). Thinking creatively at work: Organization influences on creative problem solving. The Journal of Creative Behavior, 31, 7–17. doi 10.1002/ j.2162-6057.1997.tb00777.x Reiter-Palmon, R., & Illies, J. J. (2004). Leadership and creativity: Understanding leadership from a creative problem-solving perspective. The Leadership Quarterly, 15, 55–77. doi 10.1016/j.leaqua.2003.12.005 Shalley, C. E., & Zhou, J. (2008). Organizational creativity research: A historical overview. In J. Zhou & C. E. Shalley (Eds.), Handbook of organizational creativity (pp. 3–31). New York: Erlbaum. Silvia, P. J., Wigert, B., Reiter-Palmon, R., & Kaufman, J. C. (2012). Assessing creativity with self-report scales: A review and empirical evaluation. Psychology of Esthetics, Creativity, and the Arts, 6, 19–34. doi 10.1037/a0024071 Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson/Allyn & Bacon. Tan, C.-S., & Hashim, I. H. M. (2009). An investigation on relationships between creativity, problem solving and life stress: A study on Malaysian university students. In J. L. S. Jaafar & S. McCarthy (Eds.), Building Asian families and communities in the 21st century: Selected proceedings of the 2nd Asian Psychological Association Conference, Kuala Lumpur, Malaysia, June, 2008 (pp. 338–360). Newcastle, UK: Cambridge Scholars Publishing. Tan, C.-S., & Qu, L. (2012). Generality and specificity: Malaysian undergraduate students’ self-reported creativity. The International Journal of Creativity & Problem Solving, 22(2), 19–30. Tan, C.-S., & Viapude, G. N. (2015, June). The best-known creator in the eye of Malaysian undergraduates. Paper presented at the 7th

Swiss Journal of Psychology (2017), 76 (1), 35–42

C.-S. Tan & T. Teo: Psychometric Qualities of the CPES

International Conference on Humanities and Social Sciences, Songkhla, Thailand. To, M. L., Fisher, C. D., Ashkanasy, N. M., & Rowe, P. A. (2012). Within-person relationships between mood and creativity. Journal of Applied Psychology, 97, 599–612. doi 10.1037/a0026097 Wallach, M. A., & Kogan, N. (1965). Modes of thinking in young children. New York: Holt, Rinehart and Winston. Wallas, G. (1926). The art of thought. New York: Harcourt-Brace. Wang, B., & Greenwood, K. M. (2013). Chinese students’ perceptions of their creativity and their perceptions of Western students’ creativity. Educational Psychology, 33, 628–643. doi 10.1080/01443410.2013.826345 Yue, X. D., Bender, M., & Cheung, C.-K. (2011). Who are the bestknown national and foreign creators: A comparative study among undergraduates in China and Germany. The Journal of Creative Behavior, 45, 23–37. doi 10.1002/j.2162-6057.2011. tb01082.x Zhang, X., & Bartol, K. M. (2010a). Linking empowering leadership and employee creativity: The influence of psychological empowerment, intrinsic motivation, and creative process engagement. Academy of Management Journal, 53, 107–128. doi 10.5465/ AMJ.2010.48037118 Zhang, X., & Bartol, K. M. (2010b). The influence of creative process engagement on employee creative performance and overall job performance: A curvilinear assessment. Journal of Applied Psychology, 95, 862–873. doi 10.1037/a0020173 Zhou, J., & George, J. M. (2001). When job dissatisfaction leads to creativity: Encouraging the expression of voice. Academy of Management Journal, 44, 682–696. doi 10.2307/3069410 Chee-Seng Tan PF-004 Department of Psychology and Counseling Faculty of Arts and Social Science Universiti Tunku Abdul Rahman Jalan Universiti Bandar Barat 31900 Kampar, Perak Malaysia tcseng@utar.edu.my

GISC-EL Gießener Screening zur Erfassung der erweiterten Lesefähigkeit Nils Euker / Arno Koch / Jan Kuhl Das GISC-EL ist ein Einzeltest zur Erfassung der Lesekompetenz bei Kindern und Jugendlichen mit dem Förderschwerpunkt geistige Entwicklung. Es kann an Förderschulen und im Inklusiven Unterricht eingesetzt werden, bei Schülerinnen und Schülern mit dem Förderschwerpunkt geistige Entwicklung über alle Altersstufen hinweg (ca. 6–20 Jahre), bei Schülern mit dem Förderschwerpunkt Lernen im Grundstufenalter (ca. 6–10 Jahre).

Test komplett Bestehend aus: Manual, Stimulusbuch, 10 Protokollbogen, 10 Klassenauswertungen und Box. Best.-Nr. 03 224 01 € 135,00 / CHF 165.00

www.hogrefe.com

Das GISC-EL erfasst die verschiedenen Facetten des Lesens im weiteren (Lesen ikonischer und symbolischer Zeichen) und engeren Sinne (Lesen der Alphabetschrift) und berücksichtigt auch relevante Vorläuferkompetenzen (phonologische Bewusstheit und Buchstabenkenntnis). Das Verfahren folgt den Erwerbsstufen des erweiterten Lesens und erfasst insgesamt 8 Kompetenzstufen, die vom Lesen fotorealistischer Abbildungen bis hin zum sinnentnehmenden Textlesen reichen und jeweils durch einen Subtest repräsentiert sind. Die Aufgabenformate sind leicht verständlich und auch bei Schülern mit stärkerer kognitiver Beeinträchtigung einsetzbar. Die Testergebnisse er-

möglichen eine gezielte Planung der schriftsprachlichen Förderung und der lebenspraktischen Unterstützung. Eine zusätzliche Evaluationsstudie zeigte, dass das Verfahren auch bei Kindern mit Förderschwerpunkt Lernen eingesetzt werden kann. Zuverlässigkeit Die Werte für die innere Konsistenz der Testbereiche liegen zwischen α = .83 und .99. Die Retest-Reliabilität beträgt für die Subtests zum Lesen im weiteren Sinne rtt = .83, für die Subtests zum Lesen im engeren Sinne rtt = .98 und für den Gesamttest rtt = .98. Gültigkeit Es liegen positive Befunde zur inhaltlichen und zur konvergenten und diskriminanten Validität vor. Normen Aufgrund der großen Heterogenität der Zielgruppe wurde auf eine Normierung verzichtet. Die Auswertung erfolgt kriterial. Bearbeitungsdauer Je nach Lesekompetenz ca. 15 bis 35 Minuten.

Franz Petermann / Manfred Döpfner / Anja Görtz-Dorten

Franz Petermann Manfred Döpfner Anja Görtz-Dorten

Ratgeber aggressives und oppositionelles Verhalten bei Kindern Informationen für Betroffene, Eltern, Lehrer und Erzieher 3., überarbeitete Auflage

Ratgeber aggressives und oppositionelles Verhalten bei Kindern Informationen für Betroffene, Eltern, Lehrer und Erzieher

Aggressivoppositionelles Verhalten im Kindesalter

Franz Petermann Manfred Döpfner Anja Görtz-Dorten

3., überarbeitete Auflage

Franz Petermann / Manfred Döpfner / Anja Görtz-Dorten

Aggressivoppositionelles Verhalten im Kindesalter

Leitfaden Kinder- und Jugendpsychotherapie

(Ratgeber zur Reihe: „Ratgeber Kinder- und Jugendpsychotherapie“, Band 3) 3., überarb. Auflage 2016, 47 Seiten, Kleinformat, € 8,95 / CHF 11.90 ISBN 978-3-8017-2649-2 / Auch als eBook erhältlich

(Reihe: „Leitfaden Kinder- und Jugendpsychotherapie“, Band 1) 3., überarb. Auflage 2016, X/181 Seiten, € 24,95 / CHF 32.50 ISBN 978-3-8017-2648-5 / Auch als eBook erhältlich

Der Ratgeber informiert über aggressives Verhalten bei Kindern und gibt Hinweise, wie man in Familie, Schule oder Kindergarten mit dieser Problematik besser klarkommen kann.

Der Band beschreibt Leitlinien zur Diagnostik und Therapie aggressiv-oppositioneller Störungen bei Kindern.

Verlängerte Konfrontationstherapie für Jugendliche mit einer Posttraumatischen Belastungsstörung

Edna B. Foa Kelly R. Chrestman Eva Gilboa-Schechtmann

Die emotionale Verarbeitung traumatischer Erfahrungen Deutsche Übersetzung und Bearbeitung von Anne Boos, Theres Gläser und Sabine Schönfeld

Edna B. Foa / Kelly R. Chrestman / Eva Gilboa-Schechtman

Verlängerte Konfrontationstherapie für Jugendliche mit einer Posttraumatischen Belastungsstörung

Störung des Sozialverhaltens bei Jugendlichen

Rudolf Eigenheer Bruno Rhiner Marc Schmid Edith Schramm

Die Multisystemische Therapie in der Praxis

Rudolf Eigenheer / Bruno Rhiner / Marc Schmid / Edith Schramm

Störung des Sozialverhaltens bei Jugendlichen Die Multisystemische Therapie in der Praxis

Die emotionale Verarbeitung traumatischer Erfahrungen

Therapeutische Praxis

Praxis der Paarund Familientherapie

(Reihe: „Therapeutische Praxis“) 2016, 142 Seiten, Großformat, inkl. CD-ROM, € 44,95 / CHF 55.90 ISBN 978-3-8017-2630-0 Auch als eBook erhältlich

(Reihe: „Praxis der Paar- und Familientherapie“, Band 10) 2016, X/289 Seiten, € 29,95 / CHF 39.90 ISBN 978-3-8017-2528-0 Auch als eBook erhältlich

Das Manual beschreibt die Durchführung der Verlängerten Konfrontationstherapie nach Edna Foa für traumatisierte Jugendliche zwischen 13 und 18 Jahren.

Der Band stellt ein Therapieverfahren zur Behandlung der Störung des Sozialverhaltens bei Jugendlichen, die Multisystemische Therapie, vor.

www.hogrefe.com

P. C. Mefoh and V.Sw C. issJournal Ezeh: Co gnitive of Psychology Style Versus (2017), ©Memory 2017 76 (1), Hogrefe 43–46 Slips

Short Research Note

Effect of Cognitive Style on Prospective-Retrospective Memory Slips Unipolar Approach Philip C. Mefoh and Valentine C. Ezeh Department of Psychology, University of Nigeria, Nsukka, Nigeria

Abstract. We examined the effect of cognitive style on prospective and retrospective memory slips using the Group Embedded Figures Test (GEFT) and the Prospective and Retrospective Memory Questionnaire (PRMQ). A group of 233 undergraduate students (55% women) of the University of Nigeria, Nsukka, whose mean age was 19.66 years (SD = 3.02), participated in this study. Using bivariate linear regression to analyze the data, we found that cognitive style accounted for 7% of the variation in prospective memory slips and 21% of the variation in retrospective memory slips. The findings demonstrated that cognitive style significantly negatively predicted prospective and retrospective memory slips: As field independence increased, prospective and retrospective memory slips decreased. Keywords: field-dependent/independent cognitive style, lapsus memoriae, memory slips, prospective-retrospective memory, recall

Previous research has shown that field-dependent/independent cognitive styles constitute important aspects of individual differences among students with respect to the way they acquire and process information (Graff, 2003; Li, Dong, & Gong, 2009; Nigro, Cicogna, D’Olimpio, & Cosenza, 2012; Witkin, Moore, Goodenough, & Cox, 1977). Handal and Herrington (2004) found that students with a field-independent cognitive style were more efficient in comprehending hypermedia instruction than their field-dependent counterparts. Other studies (Douglas & Riding, 2007; Lopez-Ruperez, Palacios, & Sanchez, 1991; Tinajero & Paramo, 1997) found that learners with different cognitive styles pay attention to different aspects of information. They encode, store, and recall information differently and also think and comprehend in different ways. For example, Tinajero and Paramo (1997) investigated the relationship between cognitive style and student achievement in several subject domains, including English, mathematics, and Spanish. They found that cognitive style was a significant source of variation in students’ overall performance. Students who learn about their own cognitive style frequently become better learners, obtain higher grades, are more motivated, and have a more positive attitude about their studies (Sternberg & Zhang, 2001; Zhang & Sternberg, 2006). The field-dependent/independent model of cognitive style proposes that people differ with respect to their level of dependence on context. Accordingly, individuals who are more fieldindependent (have scores in the upper range) process information more semantically and/or analytically than those who are more field-dependent (have scores in the lower range). The two styles have stable and predictable attributes (Wapner, 1986; © 2017 Hogrefe

Witkin et al., 1977). The Group Embedded Figures Test (GEFT) is one of the most widely used instruments for classifying cognitive style as field-dependent or field-independent. However, some researchers (e.g., Kozhevnikov, Kosslyn, & Shephard, 2005) maintain that the classification procedure used by the GEFT may be somewhat confounded because the dimensions on which cognitive processes vary may overlap. Consequently, many of the previous studies (e.g., Amazue, 2006; Annis, 1979; Douglas & Riding, 2007; Fritz, 1992) may have suffered from arbitrary distinctions between field-dependent and field-independent cognitive styles. This has probably made the study of field-dependent/independent cognitive style less attractive. The present study was designed to overcome this problem. GEFT scores were treated as continuous rather than dichotomous data and range from zero to a maximum value, whereby higher scores indicate greater field independence. The present research examines the effect of cognitive style on prospective and retrospective memory slips/lapses in everyday life. Prospective memory is a memory for future events, plans, or intentions, while retrospective memory is a memory for past events or actions (Maylor & Logie, 2010). Prospective and retrospective memory slips or lapsus memoriae (i.e., minor errors in recall) is measured with the Prospective and Retrospective Memory Questionnaire (PRMQ). The PRMQ has been used with participants in a number of European countries and has been recommended for use in other cultures (Crawford, Henry, Ward, & Blake, 2006; Crawford, Smith, Maylor, Della Sala, & Logie, 2003; Gonzalez-Ramirez & Mendoza-Gonzalez, 2011; Kliegel & Martin, 2003). By administering the PRMQ to a Nigerian sample, this study follows the call to internationalize Swiss Journal of Psychology (2017), 76 (1), 43–46 DOI 10.1024/1421-0185/a000190

the PRMQ. We hypothesize that cognitive style predicts prospective and retrospective memory slips.

Method Participants A group of 233 undergraduate students (106 male, 127 female) of University of Nigeria, Nsukka, participated in this study. Their ages ranged from 18 to 22 years (M = 19.66, SD = 3.02). They were predominantly right-handed (95%) and were attending the General Psychology class. Most of the students in the General Psychology class were psychology majors, but there were a few students who were combining psychology with another social science discipline (e.g., psychology and economics; psychology and political science) who offer the course as an ancillary subject. All of the participants participated in partial fulfillment of the course requirement.

P. C. Mefoh and V. C. Ezeh: Cognitive Style Versus Memory Slips

reliability (.89 on test-retest over a three-year period; Witkin et al., 1971). In Nigeria, the GEFT has a reliability index of .69 (Amazue, 2006). The PRMQ is a self-report measure of prospective and retrospective memory slips in everyday life. It consists of 16 questions about minor memory mistakes that people make from time to time. Eight of the questions concern prospective memory slips, while the other eight concern retrospective memory slips. On a 5-point Likert scale ranging from 1 (very often) to 5 (never), respondents rate how often each of the incidents has happened to them. Thus, the minimum and maximum total scores are 16 and 80, respectively. Sample items are: “Do you decide to do something in a few minutes’ time and then forget to do it?” (prospective item), and “Do you fail to recognize a place you have visited before?” (retrospective item). The Cronbach’s α of the total scale and the prospective and retrospective subscales were .89, .84, and .80, respectively (Smith et al., 2000). A pilot study conducted with 76 participants to standardize the instrument in a Nigerian sample showed that the PRMQ is a valid and reliable measure of the two kinds of memory: Cronbach’s α was .69 and .77 for prospective and retrospective memory slips, respectively.

Instruments Two instruments were used in this study: the Group Embedded Figures Test (GEFT; Witkin, Oltman, Raskin, & Karp, 1971) and the Prospective and Retrospective Memory Questionnaire (PRMQ; Smith, Della Sala, Logie, & Maylor, 2000). The GEFT is a perceptual test frequently used to classify individuals as field-dependent or field-independent. Each item of the test requires one to locate a simple figure that is embedded in a complex pattern. The test consists of 25 items organized into three sections. The first section, which has seven items, is used for practice. The other two sections, containing nine items each, are scored. Sample GEFT items are: “Find simple form ‘C,’( “Find simple form ‘F,’” “Find simple form ‘A.’” GEFT scores reflect the level of perceptual disembedding ability. Test-takers receive one point for each simple figure that they correctly locate within the complex pattern. Total scores range from 0 to 18. The GEFT is one of the most widely known instruments for measuring cognitive style, but it has been criticized for containing an element of ability and so may not measure cognitive style alone (Kirton, 2003). This problem notwithstanding, the GEFT has remained the chief instrument for classifying people with respect to field-dependent/independent cognitive style. Dichotomizing cognitive style into two distinct groups seems to be somewhat problematic. Some researchers (e.g., Kozhevnikov et al., 2005) have criticized the GEFT measure for turning a continuum of uncertainty into a dichotomous field-dependent versus field-independent cognitive style. The present study treats GEFT scores as continuous data. Higher scores reflect greater skill at locating simple figures and thus greater perceptual disembedding ability. The GEFT has satisfactory validity (a correlation of .82 between the two major subsections) and Swiss Journal of Psychology (2017), 76 (1), 43–46

Procedure Participants were presented with the GEFT and the PRMQ in a quiet classroom. Before they completed the measures, the participants were told that the tests were not an examination and that participation was anonymous. All 233 participants completed the two instruments. They completed the GEFT first. The test consists of three sections; the first section is for practice and the two other sections are for actual scoring. The researchers gave verbal instructions regarding Section 1 and demonstrated how a figure could be traced over the lines of a complex figure. Before turning over to the timed sections (i.e., Sections 2 and 3) of the test, the researchers reminded the participants about the instructions for completing the test. The instructions read as follows: “This is a test of your ability to find a simple form when it is hidden within a complex pattern. Try to find the simple form in the complex figure and trace it in pencil directly over the lines of the complex figure. It is the same size, in the same proportions, and faces in the same direction within the complex figure as when it appeared alone.” After all the participants had sufficiently practiced the examples in Section 1, they were allowed 30 minutes to complete Sections 2 and 3 without assistance. When the test duration had elapsed, participants were instructed to turn the GEFT facedown and to get ready for the second test. Then the researchers informed the participants that the items on the PRMQ were stated in the form of questions about minor memory mistakes that everyone makes from time to time. They were to indicate how often (very often or quite often) or infrequently (sometimes, rarely, or never) they make the mistakes. After the instructions © 2017 Hogrefe

P. C. Mefoh and V. C. Ezeh: Cognitive Style Versus Memory Slips

Table 1. Means, standard deviations, and the coefficient details of the unstandardized and standardized weights and t-values for the criterion variables (N = 233) Variables

Cognitive style

12.80

3.91

–

Prospective memory

20.29

5.14

–.34

.08

–.26

–4.08*

.26

.07

Retrospective memory

17.12

4.76

–.56

.07

–.46

–7.91*

.46

.21

Note. *p < .001.

had been given, participants were asked to complete the questionnaire. No time restrictions were imposed, but the last participant completed the PRMQ before the expiration of 5 minutes. Ethical approval for this research was granted by the Faculty of the Social Sciences, University of Nigeria, Nsukka, Ethics Committee.

Results and Discussion The participants in this study were not classified into the usual field-dependent and field-independent cognitive styles as in most previous studies. Instead, GEFT scores were treated as continuous (on a unipolar dimension). The data used in the analysis were based on the scores obtained from the 233 undergraduate students in the General Psychology class who completed the PRMQ. The data were analyzed using bivariate linear regression to determine the effect of cognitive style on prospective and retrospective memory slips. The analysis was run with SPSS, version 16. The descriptive statistics show that the predictor variable was negatively correlated with prospective memory slips (r = –.26, p < .001) as well as retrospective memory slips (r = –.46, p < .001). The results of the bivariate linear regression (Table 1) seem to support the hypothesis that cognitive style predicts prospective and retrospective memory slips. Cognitive style significantly predicted prospective memory slips (β = –.34, t = –4.08, p < .001), accounting for 7% of the variation in prospective memory slips. As field independence increased, prospective memory slips decreased by .34. This is evident in the confidence interval, 95% CI = [–.82, –.48]. Since the confidence interval for prospective memory slips encompassed a negative value, we concluded that the population regression coefficient for prospective memory slip is negative. Similarly, cognitive style significantly predicted retrospective memory slips (β = –.56, t = –7.91, p < .001), indicating that, as field independence increased, retrospective memory slips dropped by .56. Cognitive style accounted for 21% of the variation in retrospective memory slips and the confidence interval, 95% CI = [–.70, –.42], demonstrated that the relationship is negative. In summary, cognitive style significantly and negatively predicted prospective and retrospective memory slips (or lapsus memoriae), suggesting that as field independence increases, prospective and retrospective memory slips decreases. The association between cognitive style and retrospective memory © 2017 Hogrefe

slips seems stronger (β = –.56) than the association between cognitive style and prospective memory slips (β = –.34). This study examined the influence of cognitive style on prospective and retrospective memory slips in everyday life. The goal of the study was to determine whether cognitive style, considered as a unipolar dimension, predicts minor errors in recall of prospective and retrospective materials. Data analysis using bivariate linear regression showed that the hypothesis tested in the study was not rejected. Cognitive style statistically and negatively predicted both kinds of memory slips. More importantly, our finding is consistent with previous research on the field-dependence/independence phenomenon (e.g., Amazue, 2006; Annis, 1979; Graff, 2003; Handal & Herrington, 2004; Zhang & Sternberg, 2006). That is, there seems to be, from several sources, a consensus that the field-independent cognitive style, because it uses an active reasoning pattern, tends to be more successful in processing information. The convergence of the present finding with those of previous studies seems to downplay the apprehension expressed by Kozhevnikov et al. (2005) that the classification of cognitive style into field-dependent and field-independent cognitive styles is fraught with arbitrariness. Until there is converging evidence against the field-dependent/independent model of cognitive style, the present result tends to support previous studies in demonstrating that the model is well established (Wapner, 1986; Witkin et al., 1977). The model is also relatively parsimonious in understanding an individual’s consistent approach to organizing and processing information. Although the result of the present research seems to be consistent with those of previous studies, the authors believe that self-report and objective memory measures would not yield similar correlations. This is because objective or actual memory performance usually produces higher correlations than merely asking about people’s beliefs about their own memory ability (Morris, 1984).

References Amazue, L. O. (2006). Role of cognitive style, locality, and gender in concept learning among secondary schools. Nigerian Journal of Psychological Research, 5, 82–94. Annis, L. F. (1979). Effect of cognitive style and learning passage organization on study technique effectiveness. Journal ofEducational Psychology, 71, 620–626. doi 10.1037/0022-0663.71.5.620

Swiss Journal of Psychology (2017), 76 (1), 43–46

Crawford, J. R., Henry, J. D., Ward, A. L., & Blake, J. (2006). The Prospective and Retrospective Memory Questionnaire (PRMQ): Latent structure, normative data and discrepancy analysis for proxy-ratings. British Journal of Clinical Psychology, 45, 83–104. doi 10.1348/014466505x28748 Crawford, J. R., Smith, G., Maylor, E. A., Della Sala, S., & Logie, R. H. (2003). The Prospective and Retrospective Memory Questionnaire (PRMQ): Normative data and latent structure in a large nonclinical sample. Memory, 11, 261–275. doi 10.1080/09658210244000027 Douglas, G., & Riding, R. J. (2007). The effect of pupil cognitive style and position of prose passage title on recall. Educational Psychology, 13, 385–393. doi 10.1080/0144341930130314 Fritz, R. L. (1992, December). A study of gender differences in cognitive style and conative volition. Paper presented at the American Vocational Education Research Association Session at the American Vocational Association Convention, St. Louis, MO. Gonzalez-Ramirez, M. T., & Mendoza-Gonzalez, M. E. (2011). Spanish version of the Prospective and Retrospective Memory Questionnaire (PRMQ-S). Spanish Journal of Psychology, 14, 385–391. doi 10.5209/rev_sjop.2011.v14.n1.35 Graff, M. (2003). Learning from web-based instructional system and cognitive style. British Journal of Educational Technology, 34, 407–418. doi 10.1111/1467-8535.00338 Handal, B., & Herrington, A. (2004). On being dependent or independent in computer based learning environments. E-Journal of Instructional Science and Technology, 7(2), 1–10. Kirton, M. J. (2003). Adaptation-innovation: In the context of diversity and change. London, UK: Routledge. Kliegel, M., & Martin, M. (2003). Prospective memory research: Why is it relevant? International Journal of Psychology, 38, 193–194. doi 10.1080/00207590344000114 Kozhevnikov, M., Kosslyn, S., & Shephard, J. (2005). Spatial versus object visualizers: A new characterization of visual cognition style. Memory & Cognition, 33, 710–726. doi 10.3758/bf03195337 Li, S.-X., Dong, L.-I., & Gong, D.-Z. (2009). Attention, cognitive style and TAP effect of prospective memory. Acta Psychologica Sinica, 40, 1149–1157. doi 10.3724/sp.j.1041.2008.01149 Lopez-Ruperez, F., Palacios, C., & Sanchez, J. (1991). Relation of field independence and test-items format to student performance on written Piagetian tests. Journal of Research in Science Teaching, 28, 389–400. doi 10.1002/tea.3660280503

Swiss Journal of Psychology (2017), 76 (1), 43–46

P. C. Mefoh and V. C. Ezeh: Cognitive Style Versus Memory Slips

Maylor, E. A., & Logie, R. H. (2010). A large-scale comparison of prospective and retrospective memory development from childhood to middle age. The Quarterly Journal of Experimental Psychology, 63, 442–451. doi 10.1080/17470210903469872 Morris, P. E. (1984). The validity of subjective reports on memory. In J. E. Harris & P. R. Morris (Eds.), Everyday memory, actions and absent-mindedness (pp. 153–172). London, UK: Academic Press. Nigro, G., Cicogna, P. C., D’Olimpio, F., & Cosenza, M. (2012). The role of visual perceptual style and personality disorder traits in event-based prospective memory. Personality and Individual Differences, 53, 912–916. doi 10.1016/j.paid.2012.07.002 Smith, G., Della Sala, S., Logie, R. H., & Maylor, E. A. (2000). Prospective and retrospective memory in normal aging and dementia: A questionnaire study. Memory, 8, 311–321. doi 10.1080/09658210050117735 Sternberg, R. J., & Zhang, L. F. (Eds.). (2001). Perspectives on thinking, learning, and cognitive styles. Mahwah, NJ: Erlbaum. Tinajero, C., & Paramo, M. F. (1997). Field dependence-independence and academic achievement: A reexamination of their relationship. British Journal of Educational Psychology, 67, 199–212. doi 10.1111/j.2044-8279.1997.tb01237.x Wapner, S. (1986). Introductory remarks. In M. Bertini, L. Pizzamiglio, & S. Wapner (Eds.). Field dependence in psychological theory, research, and application (pp. 1–4). Hillsdale, NJ: Erlbaum. Witkin, H. A., Moore, C. A., Goodenough, D. R., & Cox, P. W. (1977). Field-dependent and field-independent cognitive styles and their educational implications. Review of Educational Research, 47, 1–64. doi 10.3102/00346543047001001 Witkin, H. A., Oltman, P. K., Raskin, E., & Karp, S. A. (1971). A manual for the embedded figures tests. Palo Alto, CA: Consulting Psychologists Press. Zhang, L. F., & Sternberg, R. J. (2006). The nature of intellectual styles. Mahwah, NJ: Erlbaum. Philip C. Mefoh Department of Psychology University of Nigeria Nsukka, 41000 Enugu state Nigeria philip.mefoh@unn.edu.ng

Das Programm gegen Mobbing in Schule und Kindergarten

Françoise D. Alsaker

Mutig gegen Mobbing in Kindergarten und Schule 2., unveränd. Aufl. 2017. 272 S., 18 Abb., 2 Tab., Kt € 29,95 / CHF 39.90 ISBN 978-3-456-85667-4 Auch als eBook erhältlich

Mobbing unter Kindern und Jugendlichen hat viele Gesichter. Es kann grob und offensichtlich sein, aber ebenso gut auch subtil und versteckt. Es kann in der Schule, auf dem Spielplatz, im Internet oder per SMS stattfinden. „Mutig gegen Mobbing“ legt den heutigen Kenntnisstand umfassend dar – und präsentiert ein wissenschaftlich fundiertes sowie in der Praxis erprobtes Programm gegen Gewalt in Kindergärten und Schulen. Es bietet Fachpersonen wie Lehrerinnen, Psychologinnen

www.hogrefe.com

und Sozialarbeitern sowie Eltern ein umfangreiches Instrumentarium, damit sie präventiv gegen Mobbing vorgehen und bei Mobbing erfolgreich intervenieren können. Das Buch soll Mut machen: denn der Umgang mit Mobbing ist keine Zauberkunst. Wenn man bereit ist, eigene Vorstellungen zu überdenken, Handlungsmuster zu ändern und miteinander über unangenehme Themen zu reden, dann kann mit etwas Mut viel erreicht werden.

Das neue Lehrbuch zur Psychologischen Diagnostik

Gerhard Stemmler / Jutta Margraf-Stiksrud (Hrsg.)

Lehrbuch Psychologische Diagnostik

Gerhard Stemmler Jutta Margraf-Stiksrud

Stemmler / Margraf-Stiksrud (Hrsg.)

Lehrbuch Psychologische Diagnostik

Herausgeber

2015. 386 S., 28 Abb., 37 Tab., Gb € 49,95 / CHF 65.00 ISBN 978-3-456-85518-9 Auch als eBook erhältlich

Die Psychologische Diagnostik ist eine zentrale angewandte Querschnittsdisziplin der Psychologie. Sie dient der regelgeleiteten Sammlung und Verarbeitung von gezielt erhobenen Informationen, die für die Beschreibung und Prognose menschlichen Erlebens und Verhaltens bedeutsam sind. Die Psychologische Diagnostik ist in allen Anwendungsfächern von großer Bedeutung.

Lehrbuch Psychologische Diagnostik

25.03.15 16:19

Das neue Lehrbuch zur Psychologischen Diagnostik ist konsequent an den erforderlichen Übungen im Studium ausgerichtet, aber auch interessante für die Fort- und Weiterbildung. Es orientiert sich inhaltlich konsequent an den erforderlichen Übungen im Psychologiestudium: • Verhaltensbeobachtung • Diagnostische Interviews • Testkonstruktion • Testverfahren • Das Psychologische Gutachten Es bietet damit die passgenaue Vorbereitung für die Studierenden in Bachelor und Master.