Page 1

Volume 25 / Number 1 / 2018

Volume 25 / Number 1 / 2018

European Journal of

Health Psychology

European Journal of Health Psychology

Editor-in-Chief Claus Vรถgele Associate Editors Verena Klusmann Arnold Lohaus Britta Renner Christel Salewski Silke Schmidt Heike Spaderna


Assessment methods in health psychology “This book is an excellent overview of measurement issues that are central to health psychology.” David French, PhD, Professor of Health Psychology, University of Manchester, UK

Yael Benyamini / Marie Johnston / Evangelos C. Karademas (Editors)

Assessment in Health Psychology (Series: Psychological Assessment – Science and Practice – Vol. 2) 2016, vi + 346 pp. US $69.00 / € 49.95 ISBN 978-0-88937-452-2 Also available as eBook

Assessment in Health Psychology presents and discusses the best and most appropriate assessment methods and instruments for all specific areas that are central for health psychologists. It also describes the conceptual and methodological bases for assessment in health psychology, as well as the most important current issues and recent progress in methods. A unique feature of this book, which brings together leading authorities on health psychology assessment, is its emphasis on the bidirectional link between theory and practice. Assessment in Health Psychology is addressed to masters and doctoral students in health psychology, to all

www.hogrefe.com

those who teach health psychology, to researchers from other disciplines, including clinical psychology, health promotion, and public health, as well as to health policy makers and other healthcare practitioners. This latest volume in the series Psychological Assessment – Science and Practice provides a thorough and authoritative record of the best available assessment tools and methods in health psychology, making it an invaluable resource both for students and academics as well as for practitioners in their daily work.


European Journal of

Health Psychology Volume 25 / Number 1 / 2018


Editor-in-Chief

Editorial Office

Claus Vo¨gele, Universite´ du Luxembourg, FLSHASE Campus BELVAL, Maison des Sciences Humaines, 11, Porte de Sciences, L-4366 Esch-sur-Alzette, Luxembourg, Tel. +352 46 6644-9755, E-mail zfgesundheitspsychologie@uni.lu Nicole Knoblauch, Universite´ du Luxembourg, 11, Porte de Sciences, L-4366 Esch-sur-Alzette, Luxembourg, Tel. +352 46 6644-9755, E-mail zfgesundheitspsychologie@uni-lu

Associate Editors

Verena Klusmann, Bremen, Germany Arnold Lohaus, Bielefeld, Germany Britta Renner, Konstanz, Germany

Christel Salewski, Hagen, Gemany Silke Schmidt, Greifswald, Germany Heike Spaderna, Trier, Germany

Editorial Board

Urs Baumann, Salzburg, Austria Elmar Bra¨hler, Leipzig, Germany Birte Dohnke, Schwa¨bisch Gmu¨nd, Germany Michael Eid, Berlin, Germany Heike Eschenbeck, Schwa¨bisch Gmu¨nd, Germany Toni Faltermaier, Flensburg, Germany Dieter Frey, Mu¨nchen, Germany Edgar Geissner, Prien am Chiemsee, Germany Nina Knoll, Berlin, Germany Carl-Walter Kohlmann, Schwa¨bisch Gmu¨nd, Germany Thomas Kubiak, Mainz, Germany

Friedrich Lo¨sel, Cambridge, UK Mike Martin, Zu¨rich, Switzerland Franz Petermann, Bremen, Germany Wolfgang Schlicht, Stuttgart, Germany Silke Schmidt, Greifswald, Germany Urte Scholz, Zu¨rich, Switzerland Ralf Schwarzer, Berlin, Germany Andreas Schwerdtfeger, Graz, Austria Monika Sieverding, Heidelberg, Germany Wolfgang Stroebe, Utrecht, The Netherlands Petra Warschburger, Potsdam, Germany Ju¨rgen Wegge, Dresden, Germany

Publisher

Hogrefe Publishing, Merkelstraße 3, 37085 Go¨ttingen, Germany, Tel. 0551 99950-0, Fax 0551 99950-111, E-mail publishing@hogrefe.com North America: Hogrefe Publishing, 7 Bulfinch Place, Suite 202, Boston, MA 02114, USA, Tel. (866) 823-4726, Fax (617) 354-6875, E-mail publishing@hogrefe.com Regina Pinks-Freybott, Hogrefe Publishing, Merkelstraße 3, 37085 Go¨ttingen, Gemany, Tel. +49551 99950-0, Fax 0551 99950-111, E-mail production@hogrefe.com Hogrefe Publishing, Herbert-Quandt-Straße 4, D-37081 Go¨ttingen, Germany, Tel. +49551 99950-900, Fax 0551 99950-998, E-mail zeitschriftenvertrieb@hogrefe.com Melanie Beck, Hogrefe Publishing, Merkelstraße 3, 37085 Go¨ttingen, Germany, Tel. +49551 99950-423, Fax 0551 99950-111, E-mail marketing@hogrefe.com ISSN-L 2512-8442, ISSN-Print 2512-8442, ISSN-Online 2512-8450

Production Subscriptions Advertising/Inserts ISSN Copyright Information

Publication

Subscription Prices Payment

Electronic Full Text Abstracting Services

Ó 2018 Hogrefe Publishing. This journal as well as the individual contributions and illustrations contained within it are protected under international copyright law. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without prior written permission from the publisher. All rights, including translation rights, reserved. Published in 4 issues per annual volume. European Journal of Health Psychology is the continuation of Zeitschrift fu¨r Gesundheitspsychologie (ISSN 0943-8149), the last annual volume of which (Volume 24) was published in 2016. Calendar year subscriptions only. Rates for 2018: Institutions 268.00/US $343.00; Individuals 94.00/US $120.00 (all plus 12.00/US $16.00 shipping & handling; Germany: 6.00). Payment may be made by check, international money order, or credit card, to Hogrefe Publishing, Merkelstr. 3, D-37085 Go¨ttingen, Germany. US and Canadian subscriptions can also be ordered from Hogrefe Publishing, 7 Bulfinch Place, Suite 202, Boston, MA, 02114, USA. The full text of European Journal of Health Psychology is available online at http://econtent.hogrefe.com and in PsycARTICLES. Abstracted/Indexed in Social Sciences Citation Index (SSCI), Social Scisearch, Journal Citation Report/Social Sciences Edition, PsycINFO, PsycLit, PsyJOURNALS, PSYNDEX, Scopus, IBZ, IBR and Europ. Reference List for the Humanities (ERIH). Impact Factor (2015): 0.529

European Journal of Health Psychology (2018), 25(1)

Ó 2018 Hogrefe Publishing


Contents Editorial

Trends and Transitions

1

Claus Vo¨gele Original Articles

An Examination of the ‘‘Freshman-15’’ in Germany: Predictors of Weight Change in Female University Students Adrian Meule and Petra Platte

2

Momentary Affect and the Optimism-Health Relationship: An Ambulatory Assessment Study Tim Rostalski, Holger Muehlan, and Silke Schmidt

9

Development of Coping Strategies From Childhood to Adolescence: Cross-Sectional and Longitudinal Trends Heike Eschenbeck, Steffen Schmid, Ines Schro¨der, Nicola Wasserfall, and Carl-Walter Kohlmann News and Announcements Call for Papers Meeting Calendar

Ó 2018 Hogrefe Publishing

18

31 32

European Journal of Health Psychology (2018), 25(1)


Editorial Trends and Transitions Claus Vögele Université du Luxembourg, FLSHASE Campus BELVAL, Maison des Sciences Humaines, Luxembourg

The German-language Zeitschrift für Gesundheitspsychologie [Journal of Health Psychology] was established in 1993 to provide a publication platform for the rapidly growing health psychology literature from the German-speaking countries. Although the journal was not affiliated officially with the Division of Health Psychology (Fachgruppe Gesundheitspsychologie) within the German Psychological Society (Deutsche Gesellschaft für Psychologie), it contributed significantly to the establishment of the discipline of health psychology in Germany. Up until today, the Zeitschrift für Gesundheitspsychologie has been one of the most important publication outlets for health psychology research in the German-speaking countries. To improve the accessibility of the research published in the journal to an international audience, authors have for several years been able to publish their articles in English in the journal. In parallel, a complete transformation into an English-language journal was considered and discussed by the previous editors of the journal and with members of the Division of Health Psychology and other colleagues in this field. The outcome of this discussion is now with us: the journal’s language of publication will be English from the current issue onwards. We are confident that this transformation into English will contribute to the visibility and also reception of the research published in the journal. This change of course required some preparation: In 2016, with the publication of volume 24, the last volume in which German-language articles appeared, the subtitle “European Journal of Health Psychology” was introduced. 2017 was spent attracting and reviewing English-language papers. And now from 2018 onwards, starting with volume 25, European Journal of Health Psychology will be the journal’s main title. The choice of the journal’s title was preceded by intensive discussions within the editorial team, a scoping-review by the publisher, and – due to its closeness in name – consultation with the European Health Psychology Society (EHPS). The EHPS Executive Committee

Ó 2018 Hogrefe Publishing

unanimously agreed with this choice of journal title and suggested a closer affiliation of the journal with the EHPS in the future. It’s all in the name – as it is for the European Journal of Health Psychology. As was the case for its predecessor, the European Journal of Health Psychology is dedicated to the publication of empirical or experimental research in health psychology, including applications, methodological developments, and comprehensive and critical literature reviews on the theory and application of psychological approaches to health and disease. The journal welcomes contributions from around the world, but acknowledges its origin in and relatedness to European health psychology. Editors and publisher are convinced that the internationalization of the German Zeitschrift für Gesundheitspsychologie, which will also be reflected in an extended editorial board, will continue to contribute to the success of health psychology research and its implementation in health care. We look forward to numerous contributions to the European Journal of Health Psychology and invite you and/ or your working group to submit papers on all aspects of the field! All manuscripts, including Electronic Supplementary Material (ESM), should be submitted online at http:// www.editorialmanager.com/zgp, where full instructions to authors are also available. On behalf of the editorial board – welcome to the European Journal of Health Psychology! Published online January 15, 2018 Claus Vögele Université du Luxembourg FLSHASE Campus BELVAL Maison des Sciences Humaines 11, Porte de Sciences 4366 Esch-sur-Alzette Luxembourg zfgesundheitspsychologie@uni.lu

European Journal of Health Psychology (2018), 25(1), 1 https://doi.org/10.1027/2512-8442/a000002


Original Article

An Examination of the “Freshman-15” in Germany Predictors of Weight Change in Female University Students Adrian Meule1,2 and Petra Platte3 1

Department of Psychology, University of Salzburg, Austria

2

Center for Cognitive Neuroscience, University of Salzburg, Austria Institute of Psychology, University of Würzburg, Germany

3

Abstract: The “Freshman-15” refers to an expected average weight gain of 15 pounds during the first year at college in US students. Although an overall weight gain during this period can be observed, most studies found that students gain less than 15 pounds on average. Studies in countries other than the US, however, are scarce. In the current study, 120 female freshmen at a German university were tested at the start of the first semester and again at the start of the second semester (after approximately 6 months). Body mass index (BMI) did not differ between measurements, but participants had 0.2% more body fat at the second measurement. Participants with higher BMI at the first measurement lost weight and participants with higher weight suppression (i.e., the difference between an individual’s highest previous weight and current weight) at the first measurement gained weight. Participants who reported to exercise regularly at the first measurement gained weight, but this effect was driven by those who reduced their amount of physical exercise during the first semester. Dietary habits and eating styles at the first measurement were not associated with weight change. To conclude, no evidence was found for an overall weight gain during the first semester in female, German students. Furthermore, weight change was exclusively predicted by BMI, weight suppression, and regular exercise, while eating behaviors were unrelated to weight change. Thus, it appears that variables influencing energy expenditure are more robust predictors of future weight gain than variables influencing energy intake in female freshmen. Keywords: body mass index, body composition, weight gain, weight suppression, Freshman-15

The term “Freshman-15” refers to an alleged average weight gain of 15 pounds (6.80 kg) during the first year at college in US students (Hodge, Jackson, & Sullivan, 1993). Although most studies found that an overall weight gain during this period can be observed, the average weight gain is usually substantially lower than 15 pounds (approximately 3–6 pounds [1.36–2.72 kg]; Casazza et al., 2015; Holm-Denoma, Joiner, Vohs, & Heatherton, 2008; VellaZarb & Elgar, 2009). Nevertheless, the “Freshman-15” is believed to be a serious problem in US students who enter university (Casazza et al., 2015). This is partly due to misinformation by the media about the true extent of weight change, which may potentially increase fear of gaining weight and, subsequently, dysfunctional weight control behaviors (Brown, 2008). The majority of studies on this topic have been conducted in the US and, thus, it cannot be precluded whether freshmen weight gain does similarly occur in other cultures with different educational systems. Yet, only few studies have examined weight change in university students in Europe. For example, a study from the Netherlands found that European Journal of Health Psychology (2018), 25(1), 2–8 https://doi.org/10.1027/2512-8442/a000001

female undergraduate students on average gained 0.4 kg within 1 year (Nederkoorn, Houben, Hofmann, Roefs, & Jansen, 2010). In another study from the Netherlands, university freshmen gained approximately 1 kg during the first 3 months at university (de Vos et al., 2015). In a study from Belgium, it was found that students had an average weight gain of 1 kg during their first semester at university (Deliens, Clarys, Van Hecke, De Bourdeaudhuij, & Deforche, 2013), but only male participants exhibited a continued weight gain (up to 18 months; Deliens, Deforche, De Bourdeaudhuij, & Clarys, 2015). In a study from the United Kingdom, it was found that students reported a gain of 1.5 kg during their first year at university, but data were based on self-reported weight change (Serlachius, Hamer, & Wardle, 2007). In another study from the United Kingdom, which measured body weight objectively, students gained 0.8 kg during the first 3 months, but returned to their baseline weight at 12 months (Finlayson, Cecil, Higgs, Hill, & Hetherington, 2012). To conclude, findings about weight gain in university freshmen in Europe are mixed. It seems as if weight gain particularly occurs in the Ó 2018 Hogrefe Publishing


A. Meule & P. Platte, Freshman-15 in German Students

first months at university, but may be compensated subsequently, particularly in female students. In addition to the question whether the first year at university is a crucial period for a general weight gain, examining individual differences that predict weight change can inform about relevant concepts or individuals who need to be targeted in potential approaches to prevent weight gain. Here, several predictors of weight change in students have been identified. For example, it has been reported for women that baseline body mass index (BMI) negatively predicts weight change, that is, those with a lower BMI at the beginning tend to gain weight (Bodenlos, Gengarelly, & Smith, 2015; Nederkoorn et al., 2010; Wengreen & Moncur, 2009). Similarly, students with higher weight suppression (i.e., the difference between the highest previous weight and current weight) also tend to gain weight (Lowe et al., 2006; Stice, Durant, Burger, & Schoeller, 2011). While findings on the prospective association between physical activity at baseline and weight change have been mixed (Bodenlos et al., 2015; Finlayson et al., 2012; Holm-Denoma et al., 2008), it was found that decreases in physical activity were associated with weight gain (Butler, Black, Blue, & Gretebeck, 2004; Deforche, Van Dyck, Deliens, & De Bourdeaudhuij, 2015; Wengreen & Moncur, 2009). Although changes in dietary habits have been documented in university students, prospective associations between diet quality at baseline and weight change either have not been found or were not tested (e.g., Butler et al., 2004; Deforche et al., 2015; Serlachius et al., 2007; Wengreen & Moncur, 2009). However, several studies tested prospective relationships between certain eating styles at baseline and weight change. For example, it has been found that current dieting predicted weight gain and similar, but less consistent, effects have been reported for restrained eating (Delinsky & Wilson, 2008; Lowe, Doshi, Katterman, & Feig, 2013; van Strien, Herman, & Verheijden, 2014; Vella-Zarb & Elgar, 2009). In order to enable comparability to prior studies, the majority of which investigated women (Brown, 2008), weight change was examined in female, German students during the first semester at university in the current study. Several predictors of weight change were tested, which included variables mainly related to energy expenditure (BMI, weight suppression, exercise) and variables mainly related to energy intake (dietary habits, eating behavior). Based on previous findings (Bodenlos et al., 2015; Lowe et al., 2006; Nederkoorn et al., 2010; Stice et al., 2011; Wengreen & Moncur, 2009), it was expected that a lower BMI and higher weight suppression at baseline would 1

3

predict increases in BMI. Research in eating disorder patients indicates that weight suppression predicts weight gain at low BMI in particular (Berner, Shaw, Witt, & Lowe, 2013; Butryn, Juarascio, & Lowe, 2011; Witt et al., 2014). Thus, it was explored if such an interactive effect between weight suppression and baseline BMI predicted weight gain in students as well. As findings about the prospective association between physical activity and weight change have been mixed (e.g., Bodenlos et al., 2015; Finlayson et al., 2012; Holm-Denoma et al., 2008), a nondirectional hypothesis was tested: regular exercise at baseline may predict weight gain or weight loss. Based on previous findings (Lowe et al., 2013; van Strien et al., 2014), it was expected that current dieting and restrained eating at baseline would predict increases in BMI. Finally, based on cross-sectional associations with BMI, it was explored whether low diet quality (Schröder, Fïto, & Covas, 2007), being an omnivore versus vegetarian/vegan (Spencer, Appleby, Davey, & Key, 2003), low perceived selfregulatory success in eating and weight regulation (Meule, Papies, & Kübler, 2012), frequent food cravings (Meule, Hermann, & Kübler, 2014), night eating (Meule, Allison, Brähler, & de Zwaan, 2014), and eating disorder symptomatology (Kliem et al., 2016) were associated with weight gain.

Methods Participants and Procedure Participants were recruited via postings at the University of Würzburg, Germany. One-hundred thirty-three female university freshmen provided informed consent and participated in the study within the first month of the semester (mid-October to mid-November), where their height and weight was measured and they completed several questionnaires on sociodemographic data, physical activity, dietary habits, and eating behaviors. Of these, n = 121 returned at the start of the second semester (beginning of April to mid-May) for a follow-up measurement of body weight and -composition. Mean period between the two measurements was approximately 6 months (M = 171 days, SD = 8.92). One participant with outlying age (45 years) was excluded from analyses, leaving a final sample of n = 120 participants. Mean age was M = 19.9 years (SD = 1.61) and mean BMI was M = 22.1 kg/m2 (SD = 2.70; Table 1). Fifty participants (41.7%) were psychology students and 70 participants (58.3%) were students of special needs education (German: Sonderpädagogik).1 Five participants (4.20%) reported to be smokers. Thirty-nine participants

Psychology students (M = 1.40, SD = 0.31) had better high school grades than students of special needs education (M = 2.25, SD = 0.47, t(116) = 11.0, p < .001). However, groups did not differ in other variables such as age (t(118) = 0.09, p = .932) or BMI (t(118) = 0.76, p = .447).

Ó 2018 Hogrefe Publishing

European Journal of Health Psychology (2018), 25(1), 2–8


4

A. Meule & P. Platte, Freshman-15 in German Students

Table 1. Descriptive statistics of continuous study variables at the first measurement and their correlations with change in body mass index (BMI) BMI change N = 120 Age (years) High school grade

M

SD

Range

19.9

1.61

18–27

.066

.472

0.59

1.0–3.7

.036

.702

2.70

15.1–29.7

.212

.020

1.89

BMI (kg/m2)

22.1 2

r

p

Weight suppression (kg/m )

0.96

0.97

0.00–3.73

.257

.005

Exercise (times per week)

1.51

1.44

0–5

.128

.165

Diet quality index

14.9

2.79

6–21

.082

.375

Perceived Self-Regulatory Success in Dieting Scale

12.0

3.33

3–20

.153

.095

Three-Factor Eating Questionnaire (restrained eating)

7.43

5.32

0–18

.054

.562

Three-Factor Eating Questionnaire (disinhibition)

7.10

3.16

1–16

.033

.717

Three-Factor Eating Questionnaire (hunger)

6.59

Food Cravings Questionnaire-Trait-reduced

38.5

Night Eating Questionnaire

10.3

Eating Disorder Examination-Questionnaire

1.31

3.28

1–16

.022

.815

20–62

.027

.771

3.65

3–24

.039

.669

1.12

0.00–5.21

.035

.707

10.3

(32.5%) reported to be current dieters. Thirty-four participants (28.3%) reported to be vegetarians and 2 (1.70%) reported to be vegans. Seventy-seven participants (64.2%) reported that they were regularly exercising.

Calculation of fat-free mass in females with this tool is based on the following equation: Fat-free mass = 9.53 + 0.69  stature2/resistance + 0.17  weight + 0.02  resistance; where stature2 is in cm2, resistance is in Ω, and weight is in kg (Sun et al., 2003).

Measures

Physical Activity Participants indicated whether they were regularly exercising (“Do you exercise regularly?” – yes/no). They also indicated their exercising per week on a 6-point scale ranging from “no regular exercise” to “five times per week or more.”

Sociodemographic and Anthropometric Data Participants reported their age (in years), high school grade and indicated whether they were smokers (yes/no). Body height was measured with a body height meter (cm). Body weight was measured with a digital scale (Seca GmbH & Co. KG, Hamburg, Germany). BMI was calculated as the weight in kilogram divided by the squared height in meters (kg/m2). BMI change was calculated by subtracting the BMI at the first measurement from the BMI at the second measurement (i.e., positive values indicate weight gain and negative values indicate weight loss). Participants also reported their highest weight (kg) at their current height. Weight suppression was calculated by subtracting the current BMI from the highest BMI (highest weight divided by squared height). Body composition was measured with bioelectrical impedance analysis (BIA) using the BIA-101 (RJL Systems, Detroit, MI) in supine position, with limbs away from the trunk with two electrodes placed on the dorsal surfaces of the hands and feet on the nondominant side of the body. Hand electrodes were placed at the distal metacarpals and between the distal prominences of the radius and the ulna. Foot electrodes were placed at the distal metatarsals, and between the medial and lateral malleoli of the ankle. Percent body fat was calculated using an online tool (https://www.rjlsystems.com/interactiveonline-bia), in which resistance and reactance (Ω) as well as age, gender, body height, and body weight were entered. European Journal of Health Psychology (2018), 25(1), 2–8

Dietary Habits The Food Frequency List (Winkler & Döring, 1995, 1998) was used for measuring participants’ dietary habits. A diet quality index, which is based on the recommendations of the German Nutrition Society, was calculated with higher scores representing a healthier, balanced diet (Winkler & Döring, 1995, 1998). The foods included in this index were meat, sausages, fish, potatoes, pasta, rice, lettuce, vegetables, fruits, chocolate, pastries, salty snacks, whole-grain bread, cereals, and eggs (Winkler & Döring, 1995, 1998). The diet quality index is calculated based on self-reported consumption frequency of these foods (e.g., eating vegetables every day would be scored as healthier than eating vegetables only once a week or less; eating chocolate every day would be scored as unhealthier than eating chocolate only once a week, etc.). Participants also indicated whether they ate omnivorous, vegetarian, or vegan food. Eating Behavior Dieting status (yes/no) was assessed with a single question (“Are you currently restricting your food intake to control your weight [e.g., by eating less or avoiding certain foods]?”; cf. Meule, Lutz, Vögele, & Kübler, 2012). Ó 2018 Hogrefe Publishing


A. Meule & P. Platte, Freshman-15 in German Students

The Perceived Self-Regulatory Success in Dieting Scale (PSRS; Meule, Papies, et al., 2012) was used for measuring perceived self-regulatory success in eating and weight regulation (α = .682). The Three-Factor Eating Questionnaire (TFEQ; Pudel & Westenhöfer, 1989; Stunkard & Messick, 1985) was used for measuring restrained eating (α = .890), disinhibited eating (α = .726), and hunger (α = .669). The Food Cravings Questionnaire-Trait-reduced (FCQ-T-r; Meule, Hermann, et al., 2014) was used for measuring the frequency of food craving experiences (α = .900). The Night Eating Questionnaire (NEQ; Allison et al., 2008; Meule, Allison, & Platte, 2014) was used for measuring eating large amounts of food after dinner or at night (α = .657). The Eating Disorder Examination-Questionnaire (EDE-Q; Fairburn & Beglin, 1994; Hilbert, Tuschen-Caffier, Karwautz, Niederhofer, & Munsch, 2007) was used for measuring eating disorder symptomatology in the past month (α = .954).

Data Analyses Changes in BMI and percent body fat at the first and second measurement were examined with paired t-tests. For categorical variables, analyses of variance (ANOVAs) for repeated measures were calculated. Specifically, an ANOVA was calculated with a within-subjects factor measurement (BMI at the first and second measurement) and a between-subjects factor dieting status (currently dieting vs. not dieting). Similar ANOVAs were calculated with nutrition (omnivorous vs. vegetarian and vegan) and exercise (regularly exercising vs. not exercising) as between-subjects factors. As there were only two vegans in the present sample, vegetarians and vegans were combined to one group. As only five participants were smokers, no further analyses were conducted with smoking status. For continuous variables (age, high school grade, BMI, weight suppression, exercise per week, diet quality index, PSRS scores, TFEQ scores, FCQ-T-r scores, NEQ scores, EDE-Q scores), correlations with BMI change were computed. In addition, a linear regression analysis was 2

3

5

calculated with BMI at the first measurement, weight suppression, and an interaction term BMI  Weight Suppression as predictor variables and BMI change as the outcome variable. Variables were mean-centered before calculating the product term. Effects with a p-value < .050 were considered as significant.

Results The paired t-tests for BMI and percent body fat showed that BMI did not differ between measurements (t(119) = 1.34, p = .182, d = 0.04; mean BMI change was M = 0.10 kg/m2, Mdn = 0.11, SD = 0.81, range: 2.26 to 2.37). However, participants had higher percent body fat at the second measurement (M = 26.3%, SD = 3.11) than at the first measurement (M = 26.1%, SD = 3.21, t(119) = 2.13, p = .036, d = 0.07).2 The ANOVA with the within-subjects factor measurement and the between-subjects factor dieting status did not show any main effects and no interaction effect on BMI, all Fs(1,118) < 2.33, ps > .129, ηp2 < 0.02. The ANOVA with the within-subjects factor measurement and the betweensubjects factor nutrition showed a main effect of nutrition, F(1,118) = 5.72, p = .018, ηp2 = 0.05, but no main effect of measurement and no interaction effect on BMI, all Fs(1,118) < 1.59, ps > .210, ηp2 < 0.02: vegetarians and vegans (M = 21.3 kg/m2, SD = 2.29) had lower BMI than omnivores (M = 22.5 kg/m2, SD = 2.71), independent of measurement. The ANOVA with the within-subjects factor measurement and the between-subjects factor exercise did not show any main effects, all Fs(1,118) < 0.49, ps > .487, ηp2 < 0.01, but a significant interaction effect on BMI, F(1,118) = 4.67, p = .033, ηp2 = 0.04. Regular exercisers versus non-exercisers did not differ in BMI at the first or second measurement (both ts(118) < 0.33, ps > .741), but differed in BMI change: regular exercisers’ BMI increased (M = 0.22 kg/m2, SD = 0.80, t(76) = 2.38, p = .020, d = 0.09) while it did not change in non-exercisers (M = 0.11 kg/m2, SD = 0.80, t(42) = 0.92, p = .363, d = 0.04).3

Although BMI and percent body fat yielded divergent results here, they were correlated with r = .921 (p < .001) at both measurements. Accordingly, all other analyses were similar whether using BMI or percent body fat and, thus, only results with BMI are reported for all other analyses. To further follow up this interaction, the sample was divided into four groups: participants who indicated to not regularly exercise on both measurements (n = 22), participants who indicated to regularly exercise on both measurements (n = 72), participants who indicated to not regularly exercise at the first measurement, but did so at the second measurement (n = 21), and participants who indicated to regularly exercise at the first measurement, but did not so at the second measurement (n = 5). Running an ANOVA for repeated measures with these groups yielded a significant Interaction Group  Measurement (F(3,116) = 2.95, p = .036). While groups did not differ in BMI at the first and second measurement (all ts < 1.18, ps > .243), BMI change differed between the group who ceased doing sports (M = 0.91 kg/m2, SD = 1.59) and the other three groups (no sports group: M = 0.10 kg/m2, SD = 0.54, t(25) = 2.09, p = .018; began sports group: M = 0.13 kg/m2, SD = 1.02, t(24) = 1.85, p = .077; continued sports group: M = 0.17 kg/m2, SD = 0.72, t(75) = 2.04, p = .045). Furthermore, computing a difference score for times exercising per week at the second measurement minus times exercising per week at the first measurement revealed a negative correlation with BMI change (r = .201, p = .052), indicating that those who reported being more active at the second measurement than at the first measurement tended to lose weight, while those who reported being less active at the second measurement than at the first measurement tended to gain weight.

Ó 2018 Hogrefe Publishing

European Journal of Health Psychology (2018), 25(1), 2–8


6

BMI at the first measurement was weakly, negatively correlated with BMI change (Table 1), indicating that individuals with a higher BMI at baseline tended to lose weight and those with a lower BMI at baseline tended to gain weight. Weight suppression was weakly, positively correlated with BMI change (Table 1), indicating that individuals with a higher weight suppression tended to gain weight. In the regression analysis, weight suppression significantly, positively predicted BMI change while BMI marginally significantly, negatively predicted BMI change, with no interaction effect (Table 2). None of the other continuous study variables were correlated with BMI change (Table 1).

Discussion In the current study, weight change during the first semester at university was investigated in female students in Germany. Contrary to previous findings (e.g., Deliens et al., 2013), mean BMI did not differ between measurements and mean weight change merely was M = 0.28 kg (SD = 2.31). Although there was a statistically significant increase in percent body fat, this increase was small (0.23%, which is less than a third of the fat% gain reported by Deliens et al., 2013). Therefore, the current results neither support the alleged weight gain of 15 pounds in university freshmen (“Freshman-15”) nor do they support the actual weight gain of up to five pounds (“Freshman-5”; Holm-Denoma et al., 2008; Vella-Zarb & Elgar, 2009), which has been reported in studies from the US. This interpretation also holds even when assuming that participants further gained an equal amount of body weight during the second semester, which would still be less than 1 kg within 1 year. In line with previous findings (Bodenlos et al., 2015; Nederkoorn et al., 2010; Wengreen & Moncur, 2009), lower BMI at baseline was associated with weight gain while higher BMI at baseline was associated with weight loss. Because of participants’ young age, it may be that physical maturation is one contributor to weight gain in those with low BMI. Furthermore, it may be speculated that those with high BMI might have adjusted their lifestyles in order to lose weight, which may relate to increased drive for thinness due to living in an environment with predominantly lean fellow students, as has been suggested by others (Nederkoorn et al., 2010). Independent of baseline BMI, higher weight suppression was associated with weight gain, which is in line with previous studies (Lowe et al., 2006; Stice et al., 2011). Thus, it appears that weight reduction below a previous weight generates counteracting processes that drive body weight back toward its initial level. These processes European Journal of Health Psychology (2018), 25(1), 2–8

A. Meule & P. Platte, Freshman-15 in German Students

Table 2. Results from regression analysis for predicting change in BMI from variables at the first measurement β

b

SE

BMI

0.17

0.05

0.03

1.80

t

.074

p

Weight Suppression

0.23

0.19

0.08

2.52

.013

BMI  Weight Suppression

0.01

0.002

0.03

0.09

.931

may include metabolic efficiency, consummatory changes, and increased reward value of food (Lowe, 2015). In line with previous findings (Bodenlos et al., 2015; Holm-Denoma et al., 2008), self-reported regular physical activity at baseline was associated with weight gain. However, follow-up analyses revealed that this association was driven by participants who regularly exercised at baseline, but who reduced their physical activity during the course of the semester (see Footnote 3). Hence, results correspond to previous findings showing that weight gain in university students was accompanied by a reduction of physical activity (Butler et al., 2004; Deforche et al., 2015; Wengreen & Moncur, 2009). Of note, none of the eating-related variables at baseline were associated with weight change. Accordingly, results from previous studies, which examined similar measures of dietary habits and eating behavior, have been mixed (e.g., de Vos et al., 2015; Lowe et al., 2013). There are several possible explanations as to why these measures may not predict weight change, although they are usually associated with BMI cross-sectionally. First, it may be that high scores on some measures (e.g., restrained eating) tend to be a result, but not an antecedent of high BMI (Snoek, van Strien, Janssens, & Engels, 2008). If measures are able to predict weight change, they may do so only after a longer period (e.g., several years) and not after a relatively short period of 6 months (van Strien et al., 2014). Finally, it may also be that most women with unhealthy eating habits may successfully change these during the transition from high school to university (or, specifically, from living at home to living away from home), which would also be supported by the weight loss observed in those with higher BMI in the current study. Interpretation of results is limited to female students. Future studies should also investigate weight changes and predictors thereof in male students, particularly given gender differences reported in previous studies (e.g., Bodenlos et al., 2015; Deforche et al., 2015; Deliens et al., 2015). A further limitation is that data, except height, weight, and body composition, were based on self-reports, which are vulnerable to bias (e.g., overestimation of healthy food intake or exercising). Thus, future studies may assess physical activity and eating habits with more objective measures in daily life (e.g., ambulatory assessment). Finally, weight change was only examined after 6 months and, Ó 2018 Hogrefe Publishing


A. Meule & P. Platte, Freshman-15 in German Students

thus, future studies would benefit from including additional measurements (e.g., after 3 months and after 1 year) in order to capture dynamic changes in weight. This seems particularly necessary as previous studies reported an initial weight gain during the first months at university followed by a return to participants’ initial weight, particularly in women (e.g., Deliens et al., 2015; Finlayson et al., 2012). To conclude, although a small increase in percent body fat was observed, the current findings do not support the “Freshman-15” in female, German students as participants did not gain a significant amount of weight during their first semester at university. Supporting previous findings, variables that were related to body weight (BMI, weight suppression) and physical activity (regularly exercising) at baseline predicted weight change. In contrast, dietary habits and eating behaviors at baseline were not associated with weight change. Thus, it appears that variables that influence energy expenditure (e.g., body composition and physical activity) are more robust predictors of future weight gain than variables that influence energy intake (e.g., nutrition and eating behavior) in female freshmen. Acknowledgment The authors would like to thank Carina Beck Teran, Jasmin Berker, Tilman Gründel, and Martina Mayerhofer for collecting the data.

References Allison, K. C., Lundgren, J. D., O’Reardon, J. P., Martino, N. S., Sarwer, D. B., Wadden, T. A., . . . Stunkard, A. J. (2008). The Night Eating Questionnaire (NEQ): Psychometric properties of a measure of severity of the night eating syndrome. Eating Behaviors, 9, 62–72. https://doi.org/10.1016/j.eatbeh.2007.03.007 Berner, L. A., Shaw, J. A., Witt, A. A., & Lowe, M. R. (2013). The relation of weight suppression and body mass index to symptomatology and treatment response in anorexia nervosa. Journal of Abnormal Psychology, 122, 694–708. https://doi.org/ 10.1037/a0033930 Bodenlos, J. S., Gengarelly, K., & Smith, R. (2015). Gender differences in freshmen weight gain. Eating Behaviors, 19, 1–4. https://doi.org/10.1016/j.eatbeh.2015.06.014 Brown, C. (2008). The information trail of the “Freshman 15” – A systematic review of a health myth within the research and popular literature. Health Information and Libraries Journal, 25, 1–12. https://doi.org/10.1111/j.1471-1842.2007.00762.x Butler, S. M., Black, D. R., Blue, C. L., & Gretebeck, R. J. (2004). Change in diet, physical activity, and body weight in female college freshman. American Journal of Health Behavior, 28, 24–32. https://doi.org/10.5993/AJHB.28.1.3 Butryn, M. L., Juarascio, A., & Lowe, M. R. (2011). The relation of weight suppression and BMI to bulimic symptoms. International Journal of Eating Disorders, 44, 612–617. https://doi.org/ 10.1002/eat.20881 Casazza, K., Brown, A., Astrup, A., Bertz, F., Baum, C., Brown, M. B., . . . Allison, D. B. (2015). Weighing the evidence of common beliefs in obesity research. Critical Reviews in Food

Ó 2018 Hogrefe Publishing

7

Science and Nutrition, 55, 2014–2053. https://doi.org/10.1080/ 10408398.2014.922044 Deforche, B., Van Dyck, D., Deliens, T., & De Bourdeaudhuij, I. (2015). Changes in weight, physical activity, sedentary behaviour and dietary intake during the transition to higher education: A prospective study. International Journal of Behavioral Nutrition and Physical Activity, 12, 1–10. https:// doi.org/10.1186/s12966-015-0173-9 Deliens, T., Clarys, P., Van Hecke, L., De Bourdeaudhuij, I., & Deforche, B. (2013). Changes in weight and body composition during the first semester at university. A prospective explanatory study. Appetite, 65, 111–116. https://doi.org/10.1016/ j.appet.2013.01.024 Deliens, T., Deforche, B., De Bourdeaudhuij, I., & Clarys, P. (2015). Changes in weight, body composition and physical fitness after 1.5 years at university. European Journal of Clinical Nutrition, 69, 1318–1322. https://doi.org/10.1038/ejcn.2015.79 Delinsky, S. S., & Wilson, G. T. (2008). Weight gain, dietary restraint, and disordered eating in the freshman year of college. Eating Behaviors, 9, 82–90. https://doi.org/10.1016/ j.eatbeh.2007.06.001 de Vos, P., Hanck, C., Neisingh, M., Prak, D., Groen, H., & Faas, M. M. (2015). Weight gain in freshman college students and perceived health. Preventive Medicine Reports, 2, 229–234. https://doi.org/10.1016/j.pmedr.2015.03.008 Fairburn, C. G., & Beglin, S. J. (1994). Assessment of eating disorders: Interview or self-report questionnaire? International Journal of Eating Disorders, 16, 363–370. https://doi.org/10.1002/1098108X(199412)16:4<363::AID-EAT2260160405>3.0.CO;2-# Finlayson, G., Cecil, J., Higgs, S., Hill, A., & Hetherington, M. (2012). Susceptibility to weight gain. Eating behaviour traits and physical activity as predictors of weight gain during the first year of university. Appetite, 58, 1091–1098. https://doi.org/ 10.1016/j.appet.2012.03.003 Hilbert, A., Tuschen-Caffier, B., Karwautz, A., Niederhofer, H., & Munsch, S. (2007). Eating Disorder Examination-Questionnaire: Evaluation der deutschsprachigen Übersetzung. Diagnostica, 53, 144–154. https://doi.org/10.1026/0012-1924.53.3.144 Hodge, C. N. Y., Jackson, L. A., & Sullivan, L. A. (1993). The “Freshman 15”: Facts and fantasies about weight gain in college women. Psychology of Women Quarterly, 17, 119–126. https://doi.org/10.1111/j.1471-6402.1993.tb00680.x Holm-Denoma, J. M., Joiner, T. E., Vohs, K. D., & Heatherton, T. F. (2008). The “freshman fifteen” (the “freshman five” actually): Predictors and possible explanations. Health Psychology, 27, S3–S9. https://doi.org/10.1037/0278-6133.27.1.S3 Kliem, S., Mößle, T., Zenger, M., Strauß, B., Brähler, E., & Hilbert, A. (2016). The Eating Disorder Examination-Questionnaire 8: A brief measure of eating disorder psychopathology (EDE-Q8). International Journal of Eating Disorders, 49, 613–616. https:// doi.org/10.1002/eat.22487 Lowe, M. R. (2015). Dieting: Proxy or cause of future weight gain? Obesity Reviews, 16, 19–24. https://doi.org/10.1111/obr.12252 Lowe, M. R., Annunziato, R. A., Markowitz, J. T., Didie, E., Bellace, D. L., Riddell, L., . . . Stice, E. (2006). Multiple types of dieting prospectively predict weight gain during the freshman year of college. Appetite, 47, 83–90. https://doi.org/10.1016/j.appet. 2006.03.160 Lowe, M. R., Doshi, S. D., Katterman, S. N., & Feig, E. H. (2013). Dieting and restrained eating as prospective predictors of weight gain. Frontiers in Psychology, 4, 1–7. https://doi.org/ 10.3389/fpsyg.2013.00577 Meule, A., Allison, K. C., Brähler, E., & de Zwaan, M. (2014). The association between night eating and body mass depends on age. Eating Behaviors, 15, 683–685. https://doi.org/10.1016/ j.eatbeh.2014.10.003

European Journal of Health Psychology (2018), 25(1), 2–8


8

Meule, A., Allison, K. C., & Platte, P. (2014). A German version of the Night Eating Questionnaire (NEQ): Psychometric properties and correlates in a student sample. Eating Behaviors, 15, 523–527. https://doi.org/10.1016/j.eatbeh.2014.07.002 Meule, A., Hermann, T., & Kübler, A. (2014). A short version of the Food Cravings Questionnaire-Trait: The FCQ-T-reduced. Frontiers in Psychology, 5, 1–10. https://doi.org/10.3389/fpsyg. 2014.00190 Meule, A., Lutz, A., Vögele, C., & Kübler, A. (2012). Self-reported dieting success is associated with cardiac autonomic regulation in current dieters. Appetite, 59, 494–498. https://doi.org/ 10.1016/j.appet.2012.06.013 Meule, A., Papies, E. K., & Kübler, A. (2012). Differentiating between successful and unsuccessful dieters: Validity and reliability of the Perceived Self-Regulatory Success in Dieting Scale. Appetite, 58, 822–826. https://doi.org/10.1016/j.appet. 2012.01.028 Nederkoorn, C., Houben, K., Hofmann, W., Roefs, A., & Jansen, A. (2010). Control yourself or just eat what you like? Weight gain over a year is predicted by an interactive effect of response inhibition and implicit preference for snack foods. Health Psychology, 29, 389–393. https://doi.org/10.1037/a0019921 Pudel, V., & Westenhöfer, J. (1989). Fragebogen zum Eßverhalten (FEV) – Handanweisung [Three-Factor Eating Questionnaire – Manual]. Göttingen, Germany: Hogrefe. Schröder, H., Fïto, M., & Covas, M. I. (2007). Association of fast food consumption with energy intake, diet quality, body mass index and the risk of obesity in a representative Mediterranean population. British Journal of Nutrition, 98, 1274–1280. https:// doi.org/10.1017/S0007114507781436 Serlachius, A., Hamer, M., & Wardle, J. (2007). Stress and weight change in university students in the United Kingdom. Physiology & Behavior, 92, 548–553. https://doi.org/10.1016/ j.physbeh.2007.04.032 Snoek, H. M., van Strien, T., Janssens, J., & Engels, R. (2008). Restrained eating and BMI: A longitudinal study among adolescents. Health Psychology, 27, 753–759. https://doi.org/ 10.1037/0278-6133.27.6.753 Spencer, E., Appleby, P., Davey, G., & Key, T. (2003). Diet and body mass index in 38 000 EPIC-Oxford meat-eaters, fish-eaters, vegetarians and vegans. International Journal of Obesity, 27, 728–734. https://doi.org/10.1038/sj.ijo.0802300 Stice, E., Durant, S., Burger, K. S., & Schoeller, D. A. (2011). Weight suppression and risk of future increases in body mass: Effects of suppressed resting metabolic rate and energy expenditure. The American Journal of Clinical Nutrition, 94, 7–11. https://doi. org/10.3945/ajcn.110.010025 Stunkard, A. J., & Messick, S. (1985). The three-factor eating questionnaire to measure dietary restraint, disinhibition and

European Journal of Health Psychology (2018), 25(1), 2–8

A. Meule & P. Platte, Freshman-15 in German Students

hunger. Journal of Psychosomatic Research, 29, 71–83. https:// doi.org/10.1016/0022-3999(85)90010-8 Sun, S. S., Chumlea, W. C., Heymsfield, S. B., Lukaski, H. C., Schoeller, D., Friedl, K., . . . Hubbard, V. S. (2003). Development of bioelectrical impedance analysis prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys. American Journal of Clinical Nutrition, 77, 331–340. van Strien, T., Herman, C. P., & Verheijden, M. W. (2014). Dietary restraint and body mass change. A 3-year follow up study in a representative Dutch sample. Appetite, 76, 44–49. https://doi.org/10.1016/j.appet.2014.01.015 Vella-Zarb, R. A., & Elgar, F. J. (2009). The “Freshman 5”: A metaanalysis of weight gain in the freshman year of college. Journal of American College Health, 58, 161–166. https://doi.org/ 10.1080/07448480903221392 Wengreen, H. J., & Moncur, C. (2009). Change in diet, physical activity, and body weight among young-adults during the transition from high school to college. Nutrition Journal, 8, 1–7. https://doi.org/10.1186/1475-2891-8-32 Winkler, G., & Döring, A. (1995). Kurzmethoden zur Charakterisierung des Ernährungsmusters: Einsatz und Auswertung eines Food-Frequency-Fragebogens [Short measures for characterizing nutrition patterns: Application and evaluation of a food frequency questionnaire]. Ernährungs-Umschau, 42, 289–291. Winkler, G., & Döring, A. (1998). Validation of a short qualitative food frequency list used in several German large scale surveys. Zeitschrift für Ernährungswissenschaft, 37, 234–241. https:// doi.org/10.1007/PL00007377 Witt, A. A., Berkowitz, S. A., Gillberg, C., Lowe, M. R., Råstam, M., & Wentz, E. (2014). Weight suppression and body mass index interact to predict long-term weight outcomes in adolescentonset anorexia nervosa. Journal of Consulting and Clinical Psychology, 82, 1207–1211. https://doi.org/10.1037/a0037484 Received July 27, 2016 Revision received August 8, 2017 Accepted August 10, 2017 Published online January 15, 2018 Adrian Meule Department of Psychology University of Salzburg Hellbrunner Straße 34 5020 Salzburg Austria adrian.meule@sbg.ac.at

Ó 2018 Hogrefe Publishing


Original Article

Momentary Affect and the Optimism-Health Relationship An Ambulatory Assessment Study Tim Rostalski, Holger Muehlan, and Silke Schmidt Department Health & Prevention, Institute of Psychology, Ernst-Moritz-Arndt-University, Greifswald, Germany Abstract: The aim of this intensive longitudinal study was to examine the moderating effect of affect on the optimism-health relationship and to separately consider valence and arousal, the basic dimensions of affect. For 14 days 45 students answered three times a day a questionnaire regarding affect and health status. Valence interacts with optimism in the prediction of health and tense arousal moderates the pessimism-health relationship. Findings provide support for the relevance of a two-factor model of dispositional optimism and the importance of separate consideration of the basic affect dimensions in the understanding of the processes between optimism and health. Keywords: dispositional optimism, momentary affect, valence, arousal, health status

Background A positive relationship between dispositional optimism, a generalized positive outcome expectancy, and enhanced health has been demonstrated in a variety of studies (Rasmussen, Scheier, & Greenhouse, 2009). This relation has been shown for different physical health outcomes, for example, people with optimistic outcome expectancies report less physical symptoms (Baker, 2007; Lyons & Chamberlain, 1994; Northouse et al., 1999; Pritchard, Wilson, & Yamnitz, 2007), fewer upper respiratory illness (URI) symptoms (Lyons & Chamberlain, 1994), and less pain (Chaney et al., 2004; Ferreira & Sherman, 2007; Lau & Knardahl, 2008). Optimists also report better health-related behaviors (Robbins, Spence, & Clark, 1991), such as greater physical activity (Baker, 2007; Carvajal, 2011) or less alcohol consumption (Carvajal, 2011). They also appraise their self-rated health status (SRHS) as higher than pessimists do (Baker, 2007; Lyons & Chamberlain, 1994; Northouse et al., 1999; Pritchard et al., 2007). The available meta-analytic results confirm these findings. Andersson (1996) reported a weighted combined correlation of .23 for optimism and symptom report (k = 30 studies). Another meta-analytic review conducted by Rasmussen et al. (2009) on the relation between optimism and physical health demonstrated an effect size in a small to moderate range (r = .17, k = 83). Alarcon, Bowling, and Khazon (2013) found a mean sample-weighted Ă&#x201C; 2018 Hogrefe Publishing

corrected correlation of .35 for the relation between optimism and self-rated physical health (k = 36). Although these results provide major evidence for a positive relation between optimism and health, some studies have failed to find a link (e.g., Chamberlain, Petrie, & Azariah, 1992; Costello et al., 2002; Pakenham & Rinaldis, 2001). Several reasons have been proposed to explain the discrepancies, especially the dimensionality of optimism and the role of affect. Dispositional optimism has been conceptualized as a bipolar one-dimensional construct, with optimism and pessimism as opposite poles (Scheier & Carver, 1985). As a standard for the measurement, the Life Orientation Test (LOT; Scheier & Carver, 1985) and the revision the Life Orientation Test-Revised (LOT-R; Scheier, Carver, & Bridges, 1994) have been established. However, factor analyses with items of both the LOT and the LOT-R have often found a solution with two separate factors (Appaneal, 2012; Chang & McBride-Chang, 1996; Herzberg, Glaesmer, & Hoyer, 2006; Kubzansky, Kubzansky, and Maselko, 2004). Moreover, the authors of the initial conceptualization of dispositional optimism (Scheier & Carver, 1985) have since recommended to use both the one-factor model and the two-factor model for analyses (Rasmussen et al., 2009; Scheier et al., 1994). Studies using a two-factor model of optimism have found different relations with health parameters for each factor, with a higher predictive power for optimism than for pessimism. These differences European Journal of Health Psychology (2018), 25(1), 9â&#x20AC;&#x201C;17 https://doi.org/10.1027/2512-8442/a000003


10

T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

in measurement models of optimism may explain the discrepancies in the optimism-health relationship. A second explanation for such differences may be provided by findings from affect research. Some studies have found that affectivity and momentary affect influence the relationship between optimism and health. Some findings indicate that dispositional optimism is related directly and indirectly over positive and negative affectivity. Chang, Sanna, and Yang (2003) initially formulated and proved a mediation model for negative affectivity. This model was confirmed by Baker (2007) for different health outcomes and extended to positive affectivity. Baker also emphasized the role of momentary affect on the intra-individual relationship between optimism and health, proposing fluctuations in momentary affect as the intermediary mechanism, with a moderating effect for positive, but not for negative affect (Baker, 2007). In her study, optimists responding with higher than their average daily positive mood were less likely to drink and more likely to exercise. However, the implicated affect model in Baker’s study makes her findings difficult to interpret. Although the two-dimensional model of state affect (Watson & Tellegen, 1985) and its questionnaire, the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988), are widely used, they do not sufficiently describe the complete affective experience. Schimmack and Grob (2000) showed that such a two-dimensional model only poorly fits affective data. Moreover, the PANAS combines “valence” and “arousal,” the two basic dimensions according to the circumplex model of affect (Russell, 1980). This compound measurement significantly complicates the interpretation of the results (Schimmack, 1999). A second problem with the PANAS model is the onedimensional structure of arousal. It has been shown that a conception of arousal as two separate and largely independent dimensions – energetic arousal (awake-tired) and tense arousal (tense-relaxed) – significantly improved the data fit (Schimmack & Reisenzein, 2002; Steyer, Schwenkmezger, Notz, & Eid, 1994; Thayer, 1989). To account for these problems, Schimmack and Grob (2000) recommend a bipolar three-dimensional affect model consisting of three dimensions: valence, energetic arousal, and tense arousal. In order to investigate the influence of momentary affect, an Ambulatory Assessment (AA) approach is recommended. This approach allows the assessment of self-reports in real time across multiple moments in the everyday life of participants (Fahrenberg, Myrtek, Pawlik, & Perrez, 2007; Mehl & Conner, 2012). These real-time self-reports are a substantial improvement over the more common retrospective self-report methods (Reis, 2012), particularly for reports of momentary affect. According to Robinson and Clore (2002), affective experiences are fleeting and only accessible for introspection until the feeling dissipates. Afterwards, European Journal of Health Psychology (2018), 25(1), 9–17

the affective experience needs to be reconstructed on the basis of episodic or semantic information, which may lead to distortions in the reports. Real-time self-reports, such as AA, on the other hand, offer an opportunity to collect momentary affect reports based on introspective access. Another advantage of the AA approach is its high ecological validity (Reis, 2012). The real-time self-reports are assessed in the everyday life of participants and allow the study of behavior within its natural, spontaneous context. Thus, the results have high generalizability to other settings. Finally results from the AA approach have shown that within-person relationships often differ from betweenperson relations (Affleck, Zautra, Tennen, & Armeli, 1999). As Baker (2007) emphasized, previous research examining the influence of dispositional optimism on health has primarily utilized an interindividual level approach, leading to inconsistent results.

Present Study This study has two major aims. First, the relations between dispositional optimism, affect according to the threedimensional model (valence, energetic arousal, tense arousal), and subjective health (self-rated health status, SRHS) as measured at the intra-individual level were examined. Consistent with theoretical considerations and previous research findings, optimism was assumed to be significantly associated with better health status, higher pleasure, higher alertness, and more relaxation. The second aim of the present study was to explore which, if any, dimensions of affect moderates the relationship between optimism and health status. Following Baker (2007), momentary affect was hypothesized to have an important influence on this relationship. This is only the second study which examines the moderating effect of momentary affect on the optimism-health relationship, and the first which separately considers valence and arousal. This separation allows a more precise investigation into the moderating effect of affect.

Methods Participants Forty-five students (37 of whom were women) took part in this study. The mean age was 23.7 years (SD = 3.83, range 20–43). Study participants were recruited from Health Psychology courses and by email advertising using the university mailing list. Twenty-one participants received psychology course credits for taking part in the study, all others received no incentives. All participants gave their informed consent. The study was conducted in four waves. Ó 2018 Hogrefe Publishing


T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

Procedure Prior to the experience-sampling period, dispositional optimism was assessed one time, together with some further variables not subject of the current study. Afterwards, participants were equipped with an iPod Touch (Apple Inc., Cupertino, CA) for 14 consecutive days. The Ambulatory Assessment procedure was implemented using iDialogPad (Gerhard Mutz, University of Cologne, Germany). A signal-contingent sampling scheme was used that acoustically prompted participants three times per day to complete the questionnaire (with random time windows of ±10 min around 10 a.m., 4 p.m., and 10 p.m.). The time points were chosen to represent the three parts of one day (morning, afternoon, evening). In each data entry trial, they were asked to rate their SRHS and momentary affect.

Measures Optimism Dispositional optimism was assessed with the German version of the Life Orientation Test-Revised (LOT-R; Glaesmer et al., 2012). The LOT-R consists of ten items, rated on a 5-point Likert scale ranged from 0 = strongly agree to 4 = strongly disagree. Four items are just filler items which are not scored. Three of the remaining six items capture optimism (Cronbach’s α = .70) and three capture pessimism (α = .74). A total score is calculated by adding the inverted pessimism score to the optimism score. Cronbach’s α for the total score is .68. SRHS To assess the within-day variation of health, the momentary self-rated health status was measured with the question “How is your health at this moment?” To detect even small fluctuations over the day, a Visual Analog Scale (VAS) with a length of 101 points was used. The anchors were 0 = very poor on the left end and 100 = very good on the right end. This item was equal to the SRHS item in the study of Hasson, Arnetz, Theorell, and Anderberg (2006). Momentary Affect Momentary Affect was assessed with a short form of the Multidimensional Mood State Questionnaire (MDMQ; Steyer, Schwenkmezger, Notz, & Eid, 1997). The German version of the MDMQ consists of 12 items, rated on a 5-point extremely) and consists of three subscales – “valence” (α = .83–.89), “energetic arousal” (α = .74–.83), and “tense arousal” (α = .84–.89). It is important for the interpretation to mention, that high values on these scales mean good mood, alertness, and calmness, whereas low values mean bad mood, tiredness, and tenseness. Four items are assigned Ó 2018 Hogrefe Publishing

11

to each scale, with two items assessing positive affect states and two items assessing negative affect states. Valence is measured with the items “content,” “great,” “bad,” and “uncomfortable,” energetic arousal is formed by the items “rested,” “energetic,” “worn-out,” and “tired,” whereas the scale of tense arousal contains the items “composed,” “relaxed,” “restless,” and “uneasy.” All items appeared in the original order at each trial.

Data Analysis For the calculation of mean values and correlations, the measures over the 14 diary days were aggregated. Diary data share cross-sectional time-series characteristics containing a multilevel structure; in the current study, a two-level model with measurements (level 1) nested within participants (level 2) was used (Bolger, Davis, & Rafaeli, 2003; Nezlek, 2007). Two-level linear-mixed regression models were computed, in which random intercepts for the predictors were introduced. For the prediction of SRHS, four competing models of increasing complexity were computed for each of the three optimism scales (the total score and the two subscales): (1) First a constant-only empty model with the grand mean (γ00), level-2 variance (u0j), and level-1 variance (rij), but without any additional predictors.

γij ¼ γ00 þ u0j þ r ij

ð1Þ

(2) A second model examining the impact of the respective optimism scale. For this purpose optimism as predictor was added at level 2 (between-person level).

γij ¼ γ00 þ γ01 ðOptimismÞ þ u0j þ r ij

ð2Þ

(3) A third model including the affect dimensions at level 1 (within-person level).

γij ¼ γ00 þ γ01 ðOptimismÞ þ u0j þ γ10 ðValenceÞ þ u1j þ γ20 ðEnergetic ArousalÞ þ u2j þ γ30 ðTense ArousalÞ þ u3j þ r ij (4)

ð3Þ

To test the moderation hypothesis, an additional fourth model was computed that included cross-level interactions between optimism and the three affect dimensions.

γij ¼ γ00 þ γ01 ðOptimismÞ þ u0j þ γ10 ðValenceÞ þ γ11 ðOptimism  ValenceÞ þ u1j þ γ20 ðEnergetic ArousalÞ þ γ21 ðOptimism  Energetic ArousalÞ þ u2j þ γ30 ðTense ArousalÞ þ γ31 ðOptimism  Tense ArousalÞ þ u3j þ r ij ð4Þ European Journal of Health Psychology (2018), 25(1), 9–17


12

T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

Table 1. Descriptive characteristics and correlation coefficients for all self-report measures included in the study (SRHS, MDMQ, LOT-R) Mean

SD

Min

Max

1

2

3

4

1. SRHS

80.24

15.08

48.41

99.97

2. Valence

15.33

3.08

6.03

19.36

.49***

3. Energetic arousal

12.62

2.07

7.82

17.31

.52***

.70***

4. Tense arousal

14.28

3.25

4.61

18.88

.37**

.87***

.56***

5. LOT-R total score

13.55

3.09

8

20

.22

.26***

.18

.27

6. LOT-R optimism

8.06

2.97

0

12

.37**

.23

.23

.19

7. LOT-R pessimism

6.51

2.24

0

11

.18

.04

.05

.12

5

6

.73*** .41**

.32*

Notes. SRHS = self-rated health status (single item measure); MDMQ = Multi-Dimensional Mood Questionnaire; LOT-R = Life Orientation Test-Revised. *p < .05, **p < .01, ***p < .001.

To inspect significant continuous by continuous interactions, simple slopes (i.e., the slopes of optimism on SRHS when the respective moderator variable is held constant at different combinations of values from low to high) were computed and the results were graphed. The LOT-R scale scores were grand mean-centered, and the measurement level variables – valence, energetic arousal, and tense arousal – group mean-centered. For all analyses, STATA statistical software (version 11.2; Stata Corporation, College Station, TX) was used with its mixed modeling tool (xtmixed). Simple slopes were estimated with the margins command.

Results A total of 1,732 out of possible 1,890 recordings were obtained, with a dropout rate of 8.4%. These data indicated a rather high compliance with the protocol.

Relationships Between Optimism, Self-Rated Health Status, and Affect States As can be seen in Table 1, the first-order correlations between the variables revealed moderate positive associations between self-rated health status and the subscale optimism of the two-factor optimism scale (r = .37). There were low or insignificant correlations between self-rated health status and both the pessimism subscale (r = .18) and the one-factor optimism scale (r = .22). In analyzing the relation between optimism and affect, only a small but significant positive correlation for the one-factor optimism scale and valence was found (r = .26). Self-related health status was moderately positively associated with all three affect states (r = .37–.52), higher levels of health were related to valence, energetic arousal, and tense arousal.

European Journal of Health Psychology (2018), 25(1), 9–17

Prediction of SRHS To test our hypotheses, random-intercept mixed regression models were computed. Starting with the empty model, the between-person variance amounted to ψ = 207.76 and the within-person variance to θ = 269.71. Thus, 56.5% of the total variance was within-person variance, leaving room for further predictors. To examine the impact of optimism on SRHS, the LOT-R scales were included first. The LOT-R total score did not significantly contribute to the prediction of SRHS (b = 1.34, p = .16) and this model showed no better model fit than the empty model. Compared to the empty model, its log rank was w2(1) = 1.74, p = .19, ψ = 199.65, θ = 269.71. The same results applied to the model with the LOT-R subscale pessimism as the predictor (b = 0.95, p = .18) and for the model fit, w2(1) = 1.93, p = .17, ψ = 198.73, θ = 269.71. The LOT-R subscale optimism predicted significant SRHS (b = 2.53, p = .01) and the model showed a significant better model fit than the empty model, w2(1) = 5.98, p = .01, ψ = 180.98, θ = 269.71. To control for the impact of affect, the MDMQ scales (valence, energetic arousal, and tense arousal) were entered. This led to a superior model for all three analyses using the log rank test in comparison to the previous model, w2(3) = 564.86, p < .001. Moreover, the LOT-R subscale optimism was still a significant predictor (b = 2.53, p = .01).

Affect as Moderator of the OptimismHealth Relationship To investigate the moderation hypotheses, models were computed for all three LOT-R scales, each entering the interactions of optimism with the three MDMQ scales. For the LOT-R total score, the model showed no better model fit than the previous model, w2(3) = 5.70, p = .13, ψ = 201.67, θ = 192.32. As can be seen in Table 2, only the interaction of the LOT-R total score with tense arousal significantly predicted SRHS (b = 0.12, p < .05). If tense Ó 2018 Hogrefe Publishing


T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

13

Table 2. Random-intercept mixed regression for the prediction of SRHS (1,732 observations, N = 45): final models for the moderating role of affect in the optimism-health relationship Model 1: total score Fixed effects

Model 2: optimism

Model 3: pessimism

Est.

SE

Est.

SE

Est.

SE

Constant

80.81***

2.15

90.42***

4.25

90.50***

7.11

Valence

3.49***

0.18

2.66***

0.39

3.30***

0.56

Energetic Arousal

0.23

0.12

0.56*

0.26

0.51

0.45

0.38

1.81**

0.57

Tense Arousal LOT-R Total Score Total Score  Valence

0.67*** 0.95

0.18

0.62

0.71

0.11

0.06

Total Score  Energetic Arousal

0.03

0.04

Total Score  Tense Arousal

0.12*

0.06

LOT-R Optimism

1.75*

Optimism  Valence

0.69

0.14*

0.06

Optimism  Energetic Arousal

0.05

0.04

Optimism  Tense Arousal

0.01

0.06

LOT-R Pessimism Pessimism  Valence Pessimism  Energetic Arousal Pessimism  Tense Arousal Random effects

Var

Var SE

Var

1.34

0.95

0.03

0.07

0.04

0.06

0.16*

0.08

Var SE

Var

Var SE

Constant

201.67

43.58

183.03

39.66

200.78

43.40

Residual

192.32

6.62

192.22

6.62

192.07

6.61

Notes. SRHS = self-rated health status; LOT-R = Life Orientation Test-Revised; Var = Variance. *p < .05, **p < .01, ***p < .001.

arousal was high (this means high calmness) individuals with higher than average optimism reported a better health status than individuals with lower than average optimism (see Figure 1A). But if tense arousal was low (this means high tenseness) this effect of optimism on SRHS was reversed. For the LOT-R subscale optimism, the model fit was not significant, w2(3) = 6.55, p = .09, ψ = 183.03, θ = 192.22, but the interaction with valence was (b = 0.14, p < .05). As can be seen in Figure 1B, the positive effect of optimism on SRHS was strongest if valence was low. This effect decreased with increasing valence. For the LOT-R subscale pessimism, the model showed a better fit than the previous model, w2(3) = 7.89, p < .05, ψ = 200.78, θ = 192.07. Again, only the interaction with tense arousal predicted SRHS (b = 0.14, p < .05). Figure 1C shows that the effect of pessimism on SRHS was almost zero if tense arousal was high. But if tense arousal decreased this effect became more and more positive.

Discussion In considering the relationship between optimism and health, only the relation between the LOT-R subscale Ó 2018 Hogrefe Publishing

optimism and health status was significant. The effect of dispositional optimism on momentary health status remained significant even if affect scores were entered into the model. These results are consistent with findings from previous studies. Kubzansky et al. (2004) found an association between the optimism subscale and self-rated health status but not for the pessimism subscale. This suggests that optimism might have a greater impact on health or at least on perceived health status. The meta-analysis provided by Rasmussen et al. (2009) does not indicate differences between the two optimism subscales in their relation to health. It should be noted, however, that the authors did not account for different health measures applied in the various studies, due to the small number of available studies. Differences between the health measures can be assigned to the one-dimensional optimism scale (Rasmussen et al., 2009), and it can be assumed that this also applies to the two subscales. Even if the correlations are not significant, they are in the range of effects reported in the various meta-analyses (Alarcon et al., 2013; Rasmussen et al., 2009). Regarding the magnitude of correlation coefficients, the relationship between SRHS and the affect was more pronounced than the relationship between SRHS and optimism, as well as the relationship between affect and optimism. This is probably due to the fact that optimism is a one-time measured trait, while SRHS and momentary European Journal of Health Psychology (2018), 25(1), 9–17


14

T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

(A)

(B)

(C)

Figure 1. Interactions predicting SRHS. (A) SRHS as a function of optimism (one-factor model) and tense arousal. (B) SRHS as a function of optimism (two-factor model) and valence. (C) SRHS as a function of pessimism (two-factor model) and tense arousal.

affect are states which have been collected several times. In this regard, it has been shown that affect and health are generally related (e.g., Kelsey et al., 2006; Pressman & Cohen, 2005). The question of how momentary affect moderates the relationship between optimism and health yielded an interesting pattern. First, there is a significant moderating effect for all three optimism scales. Those participants with higher than average optimism (one-factor model) reported a better health when they were in a calm mood. For participants in bad mood, the effect of optimism (two-factor model) on health was strongest. Those participants with higher than average pessimism (two-factor model) reported a better health when they were in a tense mood. Second tense arousal was found to interact both with the total score of the optimism measure and the subscale pessimism. Moreover, valence moderates the relation between subscale optimism and SRHS, thus revealing differences European Journal of Health Psychology (2018), 25(1), 9â&#x20AC;&#x201C;17

for the two dispositional optimism subscales. The total score fails to highlight this process. This finding indicates the potential of the two-factor model of dispositional optimism for future research, because it provides a possible deeper insight into the different effects of the basic affect dimensions on the optimism-health relationship. The onefactor model only partially reproduces this complex relation between optimism, health, and affect. These findings are also relevant to the question regarding the role of the dimensionality of momentary affect in moderating the optimism-health relationship. As the current study measured the two basic affect dimensions, valence and arousal, separately, it could be observed that each affect dimension only interacts with one specific dispositional optimism dimension: There was an interaction between valence and optimism, and between arousal and pessimism. The interaction between optimism and valence had already been observed in previous studies (Baker, 2007; Ă&#x201C; 2018 Hogrefe Publishing


T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

Kubzansky et al., 2004) which showed the effect for positive affect. However, by examining affect in a two-dimensional way this study has shown that valence is the moderating affect dimension. This finding can also be seen as further evidence for a bolstering effect of affect on self-reported and objective health, as noted by other authors (Baker, 2007; Pettit, Kline, Gencoz, Gencoz, & Joiner, 2001). Optimists were more likely to report better health and show more health behavior, although only on days with higher than their average positive mood.

Limitations of This Study With respect to our Ambulatory Assessment approach, SRHS was only assessed with one outcome variable. The general weakness of single items for reliable assessment has been widely discussed (Bowling, 2005; Diamantopoulos, Sarstedt, Fuchs, Wilczynski, & Kaiser, 2012). Nevertheless, this is the standard approach for SRHS assessment, especially regarding validity issues (Bowling, 2005). The validity of the single-item assessment of SHRS has been verified in a number of studies, for example, SRHS assessment has frequently been shown to predict individual life expectancy and mortality (de Boer et al., 2004; Idler & Kasl, 1995; Schoenfeld, Malmrose, Blazer, Gold, & Seeman, 1994). Another limitation of the results presented here is that the data are only based on self-assessment. This is especially true for the assessment of health, which in this study was only assessed over a SRHS. This was done to make the daily burden of participation as small as possible. This approach resulted in a dropout rate of only 8.4% in our study, which is “high compliance” in comparison to other studies in this area (Silvia, Kwapil, Eddington, & Brown, 2013). In future studies, however, additional subjective and objective indicators of health should be used to confirm the validity of the findings.

Conclusions To sum up, findings provide support for the relevance of a two-factor model of dispositional optimism and to justify a separate consideration of the basic affect dimensions in exploring the relationship between optimism and health. We conclude that in future studies interaction dynamics should be further investigated on a multidimensional level within intensive longitudinal studies, including a two-factor model of dispositional optimism and a three-factor model of momentary affect. With regard to practical implications, results should be interpreted with caution. For survey research, it might be considered to assess affect states within health surveys to provide control variables aiming at enhancing validity of self-reported health status measures. Ó 2018 Hogrefe Publishing

15

For health promotion, our study also provides some small hints toward differential treatment of persons with higher dispositional optimism versus persons with higher dispositional pessimism. With respect to potential health outcomes such as subjective health status or health behavior, “optimists” initially should be set into a state of good “positive” mood whereas “pessimists” are probably more susceptible to benefit from an intervention session if set into a state of activation.

References Affleck, G., Zautra, A., Tennen, H., & Armeli, S. (1999). Multilevel daily process designs for consulting and clinical psychology: A preface for the perplexed. Journal of Consulting and Clinical Psychology, 67, 746–754. https://doi.org/10.1037/0022-006X. 67.5.746 Alarcon, G. M., Bowling, N. A., & Khazon, S. (2013). Great expectations: A meta-analytic examination of optimism and hope. Personality and Individual Differences, 54, 821–827. https://doi.org/10.1016/j.paid.2012.12.004 Andersson, G. (1996). The benefits of optimism: A meta-analytic review of the Life Orientation Test. Personality and Individual Differences, 21, 719–725. https://doi.org/10.1016/0191-8869 (96)00118-3 Appaneal, R. N. (2012). A confirmatory factor analysis of the Life Orientation Test-Revised with competitive athletes. Research Quarterly for Exercise and Sport, 83, 592–596. https://doi.org/ 10.5641/027013612804582551 Baker, S. R. (2007). Dispositional optimism and health status, symptoms and behaviours: Assessing idiothetic relationships using a prospective daily diary approach. Psychology & Health, 22, 431–455. https://doi.org/10.1080/ 14768320600941764 Bolger, N., Davis, A., & Rafaeli, E. (2003). Diary methods: Capturing life as it is lived. Annual Review of Psychology, 54, 579–616. https://doi.org/10.1146/annurev.psych.54.101601. 145030 Bowling, A. (2005). Just one question: If one question works, why ask several? Journal of Epidemiology and Community Health, 59, 342–345. https://doi.org/10.1136/jech.2004.021204 Carvajal, S. C. (2011). Global positive expectancies in adolescence and health-related behaviours: Longitudinal models of latent growth and cross-lagged effects. Psychology & Health, 27, 1–22. https://doi.org/10.1080/08870446.2011.633241 Chamberlain, K., Petrie, K., & Azariah, R. (1992). The role of optimism and sense of coherence in predicting recovery following surgery. Psychology & Health, 7, 301–310. https:// doi.org/10.1080/08870449208403159 Chaney, J. M., Mullins, L. L., Wagner, J. L., Hommel, K. A., Page, M. C., & Doppler, M. J. (2004). A longitudinal examination of causal attributions and depression symptomatology in rheumatoid arthritis. Rehabilitation Psychology, 49, 126–133. https://doi.org/10.1037/0090-5550.49.2.126 Chang, E. C., Sanna, L. J., & Yang, K.-M. (2003). Optimism, pessimism, affectivity, and psychological adjustment in US and Korea: A test of a mediation model. Personality and Individual Differences, 34, 1195–1208. https://doi.org/10.1016/ S0191-8869(02)00109-5 Chang, L., & McBride-Chang, C. (1996). The factor structure of the Life Orientation Test. Educational and Psychological Measurement, 56, 325–329. https://doi.org/10.1177/0013164496056002013

European Journal of Health Psychology (2018), 25(1), 9–17


16

T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

Costello, N. L., Bragdon, E. E., Light, K. C., Sigurdsson, A., Bunting, S., Grewen, K., & Maixner, W. (2002). Temporomandibular disorder and optimism: Relationships to ischemic pain sensitivity and interleukin-6. Pain, 100, 99–110. https:// doi.org/10.1016/S0304-3959(02)00263-4 de Boer, A. G. E. M., van Lanschot, J. J. B., Stalmeier, P. F. M., van Sandick, J. W., Hulscher, J. B. F., de Haes, J. C. J. M., & Sprangers, M. A. G. (2004). Is a single-item visual analogue scale as valid, reliable and responsive as multi-item scales in measuring quality of life? Quality of Life Research, 13, 311–320. https://doi.org/10.1023/B:QURE.0000018499.64574.1f Diamantopoulos, A., Sarstedt, M., Fuchs, C., Wilczynski, P., & Kaiser, S. (2012). Guidelines for choosing between multi-item and single-item scales for construct measurement: A predictive validity perspective. Journal of the Academy of Marketing Science, 40, 434–449. https://doi.org/10.1007/s11747-0110300-3 Fahrenberg, J., Myrtek, M., Pawlik, K., & Perrez, M. (2007). Ambulatory Assessment – monitoring behavior in daily life settings: A behavioral-scientific challenge for psychology. European Journal of Psychological Assessment, 23, 206–213. https://doi.org/10.1027/1015-5759.23.4.206 Ferreira, V. M., & Sherman, A. M. (2007). The relationship of optimism, pain and social support to well-being in older adults with osteoarthritis. Aging & Mental Health, 11, 89–98. https:// doi.org/10.1080/13607860600736166 Glaesmer, H., Rief, W., Martin, A., Mewes, R., Brähler, E., Zenger, M., & Hinz, A. (2012). Psychometric properties and population-based norms of the Life Orientation Test Revised (LOT-R). British Journal of Health Psychology, 17, 432–445. https://doi.org/10.1111/j.2044-8287.2011.02046.x Hasson, D., Arnetz, B. B., Theorell, T., & Anderberg, U. M. (2006). Predictors of self-rated health: A 12-month prospective study of IT and media workers. Population Health Metrics, 4, 8. https://doi.org/10.1186/1478-7954-4-8 Herzberg, P. Y., Glaesmer, H., & Hoyer, J. (2006). Separating optimism and pessimism: A robust psychometric analysis of the Revised Life Orientation Test (LOT-R). Psychological Assessment, 18, 433–438. https://doi.org/10.1037/1040-3590.18.4.433 Idler, E. L., & Kasl, S. V. (1995). Self-ratings of health: do they also predict change in functional ability? The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 50, 344–353. https://doi.org/10.1093/geronb/50B.6.S344 Kelsey, K. S., DeVellis, B. M., Begum, M., Belton, L., Hooten, E. G., & Campbell, M. K. (2006). Positive affect, exercise and self-reported health in blue-collar women. American Journal of Health Behavior, 30, 199–207. https://doi.org/10.5993/AJHB.30.2.9 Kubzansky, L. D., Kubzansky, P. E., & Maselko, J. (2004). Optimism and pessimism in the context of health: Bipolar opposites or separate constructs? Personality and Social Psychology Bulletin, 30, 943–956. https://doi.org/10.1177/ 0146167203262086 Lau, B., & Knardahl, S. (2008). Perceived job insecurity, job predictability, personality, and health. Journal of Occupational and Environmental Medicine, 50, 172–181. https://doi.org/ 10.1097/JOM.0b013e31815c89a1 Lyons, A., & Chamberlain, K. (1994). The effects of minor events, optimism and self-esteem on health. The British Journal of Clinical Psychology, 33, 559–570. https://doi.org/10.1111/ j.2044-8260.1994.tb01152.x Mehl, M. R., & Conner, T. S. (2012). Handbook of research methods for studying daily life. New York, NY: Guilford Press. Nezlek, J. B. (2007). A multilevel framework for understanding relationships among traits, states, situations and behaviours. European Journal of Personality, 21, 789–810. https://doi.org/ 10.1002/per.640

European Journal of Health Psychology (2018), 25(1), 9–17

Northouse, L. L., Caffey, M., Deichelbohrer, L., Schmidt, L., Guziatek-Trojniak, L., West, S., . . . Mood, D. (1999). The quality of life of African American women with breast cancer. Research in Nursing & Health, 22, 435–448. https://doi.org/10.1002/ (SICI)1098-240X(199912)22:6<435::AID-NUR3>3.0.CO;2-N Pakenham, K. I., & Rinaldis, M. (2001). The role of illness, Resources, appraisal, and coping strategies in adjustment to HIV/AIDS: The direct and buffering effects. Journal of Behavioral Medicine, 24, 259–279. https://doi.org/10.1023/ A:1010718823753 Pettit, J. W., Kline, J. P., Gencoz, T., Gencoz, F., & Joiner, T. E. Jr. (2001). Are happy people healthier? The specific role of positive affect in predicting self-reported health symptoms. Journal of Research in Personality, 35, 521–536. https://doi.org/10.1006/ jrpe.2001.2327 Pressman, S. D., & Cohen, S. (2005). Does positive affect influence health? Psychological Bulletin, 131, 925–971. https://doi.org/ 10.1037/0033-2909.131.6.925 Pritchard, M. E., Wilson, G. S., & Yamnitz, B. (2007). What predicts adjustment among college students? A longitudinal panel study. Journal of American College Health, 56, 15–21. https:// doi.org/10.3200/JACH.56.1.15-22 Rasmussen, H., Scheier, M., & Greenhouse, J. (2009). Optimism and physical health: A meta-analytic review. Annals of Behavioral Medicine, 37, 239–256. https://doi.org/10.1007/s12160009-9111-x Reis, H. T. (2012). Why researchers should think “real-world”: A conceptual rationale. In M. R. Mehl & T. S. Conner (Eds.), Handbook of research methods for studying daily life (pp. 3–21). New York, NY: Guilford Press. Robbins, A. S., Spence, J. T., & Clark, H. (1991). Psychological determinants of health and performance: The tangled web of desirable and undesirable characteristics. Journal of Personality and Social Psychology, 61, 755–765. https://doi.org/ 10.1037//0022-3514.61.5.755 Robinson, M. D., & Clore, G. L. (2002). Belief and feeling: Evidence for an accessibility model of emotional self-report. Psychological Bulletin, 128, 934–960. https://doi.org/10.1037/00332909.128.6.934 Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. https://doi. org/10.1037/h0077714 Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4, 219–247. https://doi.org/ 10.1037/0278-6133.4.3.219 Scheier, M. F., Carver, C. S., & Bridges, M. W. (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology, 67, 1063–1078. https://doi.org/10.1037/0022-3514.67.6.1063 Schimmack, U. (1999). Strukturmodelle der Stimmungen: Rückschau, Rundschau und Ausschau [Structural models of mood: Review, overview, and outlook]. Psychologische Rundschau, 50, 90–97. https://doi.org/10.1026/0033-3042.50.2.90 Schimmack, U., & Grob, A. (2000). Dimensional models of core affect: A quantitative comparison by means of structural equation modeling. European Journal of Personality, 14, 325–345. https://doi.org/10.1002/1099-0984(200007/08)14:4< 325::AID-PER380>3.0.CO;2-I Schimmack, U., & Reisenzein, R. (2002). Experiencing activation: Energetic arousal and tense arousal are not mixtures of valence and activation. Emotion, 2, 412–417. https://doi.org/ 10.1037/1528-3542.2.4.412 Schoenfeld, D. E., Malmrose, L. C., Blazer, D. G., Gold, D. T., & Seeman, T. E. (1994). Self-rated health and mortality in the

Ó 2018 Hogrefe Publishing


T. Rostalski et al., Momentary Affect and the Optimism-Health Relationship

high-functioning elderly – a closer look at healthy individuals: MacArthur field study of successful aging. Journal of Gerontology, 49, 109–115. https://doi.org/10.1093/geronj/49. 3.M109 Silvia, P. J., Kwapil, T. R., Eddington, K. M., & Brown, L. H. (2013). Missed beeps and missing data: Dispositional and situational predictors of nonresponse in experience sampling research. Social Science Computer Review, 31, 471–481. https://doi.org/ 10.1177/0894439313479902 Steyer, R., Schwenkmezger, P., Notz, P., & Eid, M. (1994). Testtheoretische Analysen des Mehrdimensionalen Befindlichkeitsfragebogen (MDBF) [Test-theoretical analyses of the Multidimensional Mood State Questionnaire]. Diagnostica, 40, 320–328. Steyer, R., Schwenkmezger, P., Notz, P., & Eid, M. (1997). Der Mehrdimensionale Befindlichkeitsfragebogen (MDBF) [Multidimensional Mood Questionnaire]. Göttingen, Germany: Hogrefe. Thayer, R. E. (1989). The biopsychology of mood and arousal. Oxford, UK: Oxford University Press. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The

Ó 2018 Hogrefe Publishing

17

PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063 Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological Bulletin, 98, 219–234. https://doi.org/ 10.1037/0033-2909.98.2.219 Received April 24, 2017 Revision received October 11, 2017 Accepted October 12, 2017 Published online January 15, 2018 Tim Rostalski Department Health & Prevention Institute of Psychology Ernst-Moritz-Arndt-University Robert-Blum-Str.13 17487 Greifswald Germany tim.rostalski@uni-greifswald.de

European Journal of Health Psychology (2018), 25(1), 9–17


Original Article

Development of Coping Strategies From Childhood to Adolescence Cross-Sectional and Longitudinal Trends Heike Eschenbeck, Steffen Schmid, Ines Schröder, Nicola Wasserfall, and Carl-Walter Kohlmann Department of Psychology, University of Education Schwäbisch Gmünd, Germany Abstract: Extensive research exists on coping in children and adolescents. However, developmental issues have only recently started to receive more attention. The present study examined age differences and developmental changes in six coping strategies (social support seeking, problem solving, avoidant coping, palliative emotion regulation, anger-related emotion regulation, and media use) assessed by a coping questionnaire (German Stress and Coping Questionnaire for Children and Adolescents, SSKJ 3–8; Lohaus, Eschenbeck, Kohlmann, & Klein-Heßling, 2006) in middle/late childhood and early adolescence. At the initial assessment, 917 children from grades 3 to 7 (age range 8–15 years) were included (cross-sectional sample). Three cohorts (grades 3–5 at baseline) were traced longitudinally over 1½ years with four assessments (longitudinal sample: n = 388). The cross-sectional coping data showed significant effects for grade level in four coping strategies. Older children scored higher in problem solving and media use, and lower in avoidant coping. Seventh graders scored lower than fourth and fifth graders in social support seeking. Longitudinal data confirmed time effects and cohort effects indicating developmental changes. Increases over time were found for problem solving and media use; decreases were found for avoidant coping and anger-related emotion regulation. For social support seeking, an increase within the youngest cohort (grades 3–5) was found. Developmental trends (in cross-sectional and longitudinal data), with especially strong increases for problem solving or declines in avoidant coping in the youngest cohort, differed for the two studied stressful situations (social, academic) but were independent of the child’s gender. To conclude, particularly in the age range of 9–11 years relevant developmental changes toward a more active coping seem to appear. Keywords: coping, children, adolescents, cross-sectional, longitudinal, gender

There is currently a substantial corpus of research on the ways in which children and adolescents cope with stressful events (for reviews, see Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001; Zimmer-Gembeck & Skinner, 2011). However, progress regarding developmental patterns in coping across childhood and early adolescence from longitudinal research is relatively small. Therefore, the goal of the present cross-sectional and longitudinal study is to examine age differences and developmental changes in coping in children and adolescents.

Coping Strategies in Children and Adolescents Coping is an active, purposeful process that consists of cognitive changes and behavioral adaptation when handling specific external or internal demands that are evaluated as something exceeding the resources of the person European Journal of Health Psychology (2018), 25(1), 18–30 https://doi.org/10.1027/2512-8442/a000005

(Lazarus & Folkman, 1984). Several potential ways of coping have been identified. Widely used conceptualizations of coping are problem- versus emotion-focused coping (Lazarus & Folkman, 1984), approach versus avoidance coping (Krohne, 1993; Roth & Cohen, 1986), and engagement versus disengagement coping (Compas et al., 2001). The coping strategies most often considered theoretically and empirically are as follows: social support seeking (including instrumental as well as emotional support from others), problem solving (including approach and problem-focused strategies), avoidance (including efforts to disengage from the stressor), distraction (including a wide variety of alternative pleasurable activities), and emotion regulation (including efforts to palliate emotions to return to a stable pleasurable mood or keeping calm; Compas et al., 2001; Skinner, Edge, Altman, & Sherwood, 2003; Zimmer-Gembeck & Skinner, 2011). Accordingly, in the present study, coping was operationalized by six dimensions, including five strategies that were similar to the most Ó 2018 Hogrefe Publishing


H. Eschenbeck et al., Coping in Children and Adolescents

common strategies cited in the literature on children and adolescents: problem solving that reflects a strategy directly related to the stressful situation (in terms of problemfocused, approach, or engagement). Avoidant coping, palliative emotion regulation (related to relaxation and distraction), and anger-related emotion regulation (related to externalizing feelings of anger and fury) represent three more indirect coping strategies to adjust to the stressor (in terms of emotion-focused, avoidance, or disengagement). Social support seeking as a means of a mixed strategy including direct instrumental efforts focusing on the stressor as well as indirect emotional efforts to adjust to the stressor. In addition, electronic media use as an indirect coping strategy not focusing on the stressor was assessed. Children spend a lot of time using electronic media such as television, computers, and cell phones (MPFS, 2017; Ofcom, 2016). Among others, one function of electronic media use is to cope with stress, including more indirect efforts to disengage from the stressor or to palliate feelings (Leiner, Argus-Calvo, Peinado, Keller, & Blunk, 2014; Lohaus, Ball, Klein-Heßling, & Wild, 2005). Confirmatory factor analyses in third- to eighth-grade children have provided a good fit for the five factors (without electronic media use as coping strategy, comparative fit indices (CFIs) from .94 to .96; Eschenbeck, Kohlmann, Lohaus, & KleinHeßling, 2006) as well as for the six factors including the coping strategy related to electronic media (CFIs between .92 and .93; Eschenbeck, Heim-Dreger, Tasdaban, Lohaus, & Kohlmann, 2012; Eschenbeck, Kohlmann, & Meier, 2010). Considering validity, media use was positively correlated with avoidant coping, palliative emotion regulation, and anger-related emotion regulation, and it correlated negatively with problem solving (Eschenbeck et al., 2010, 2012). Regarding health outcomes, especially anger-related emotion regulation was positively related to physical and emotional stress symptoms as well as behavioral problems. A similar pattern was shown for the coping strategy media use (Eschenbeck et al., 2012). Also, the coping strategy media use was especially high for obese adolescent boys who evaluated themselves as obese compared to obese adolescent boys with unrealistically positive self-evaluations who did not evaluate themselves as obese and to normal weight boys and girls (Meier, Kohlmann, Eschenbeck & Gross, 2010). On the contrary, problem solving was negatively correlated with behavioral problems and positively with prosocial behavior (e.g., Eschenbeck et al., 2012).

Developmental Changes in Coping According to Zimmer-Gembeck and Skinner (2011) the transition from late childhood to early adolescence (covering the ages from approximately 10 to 12 years) represents Ó 2018 Hogrefe Publishing

19

a relevant window for changes in the strategies of coping with stress. It is an age period of multiple developmental changes including physical, cognitive (e.g., more abstract and complex, meta-cognitive skills), emotional (e.g., differentiation of emotions, intrapersonal regulation of emotions and distress), and social changes (e.g., social skills, friendships, to become an autonomous person; see Lerner, 2015; for developmental tasks, see Havighurst, 1972). Significant challenges occur, for example, at school (e.g., transition from primary to secondary schools), through involvement with peers in and out of school and in family/parent relationships (e.g., Byrne, Thomas, Burchell, Olive, & Mirabito, 2011). Summarizing age trends in coping, Zimmer-Gembeck and Skinner (2011) found predominantly decreases in use or stability (i.e., no differences) with age in studies on support seeking (including one longitudinal study, Vierhaus & Lohaus, 2009). For problem solving, most studies (with mainly cross-sectional designs, including only one longitudinal study, Vierhaus & Lohaus, 2009) found increases with age during childhood. These increases were observed especially when examining smaller age gaps and more controllable stressors. For escape coping (including cognitive escape or a combination of cognitive and behavioral escape) during middle and late childhood, studies reported stability in the use of this strategy (i.e., no age differences) or decreases with age (especially during late childhood when dealing with social stressors). For distraction, findings commonly showed increases in use in dealing with uncontrollable stressors or stability (the latter for many different kinds of stressors). There is a clear lack of studies on externalizing emotional coping (opposition, aggression), although there is evidence for increases with age from late childhood to adolescence (Zimmer-Gembeck & Skinner, 2011). Findings that support significant age differences or changes in central coping strategies result mainly from cross-sectional studies or initial evidence from longitudinal research (Vierhaus, Lohaus, & Ball, 2007). Thereby longitudinal studies that describe stability and change in coping are to be considered as more conclusive. The goal of our study was to analyze developmental changes in central coping strategies (i.e., seeking social support, problem solving, avoidant coping, palliative emotion regulation, anger-related emotion regulation, and media use) in more detail by making use of small age gaps within the relevant age group of middle/late childhood and early adolescence. Three longitudinal cohorts (covering grades 3–5, ages between 8 and 13 years at baseline) were traced over 1½ years for four assessments at small intervals of 6 months. A cross-sectional sample of grades 3–7 was included to compare the longitudinal results on developmental changes with the cross-sectional age differences. Following the literature, in our first research question improvements with age European Journal of Health Psychology (2018), 25(1), 18–30


20

in the child’s ability to cope with normative stressful events are expected. We hypothesized developmental changes especially for both problem solving (increase) and avoidant coping (decrease), for which the age window for changes within middle/late childhood and adolescence needs to be more precisely specified (Cohort  Time interaction). Overall, this pattern of changes in coping indicates improvements favoring more engagement and problemfocused strategies.

Gender and Stressful Situation as Influencing Factors Gender and the characteristics of the stressful situation may influence developmental changes in coping strategies. For gender, the most consistent result revealed differences for social support seeking, followed by problem solving and avoidant coping (e.g., Eschenbeck, Kohlmann, & Lohaus, 2007). Females more often report social support seeking and problem solving, whereas males more often use avoidance. Therefore, in terms of age or developmental changes, the empirical evidence for gender differences is more consistent for adolescence than for childhood (Eschenbeck et al., 2007). This leads to our second research question of whether gender-specific developmental changes and stability exist in coping during middle/late childhood and early adolescence. We assumed that, as puberty is reached, the gender differences between boys and girls become more striking and, for example, as documented above, the development of coping may differ for boys and girls for social support seeking (greater decrease in boys or increase in girls) or avoidant coping (no differences or increase in boys, decrease or no differences in girls). Coping behavior should be understood in relation to the type of stressful situation (for adults, see Folkman & Moskowitz, 2004). Direct, problem-focused strategies to alter the stressful situation are especially appropriate in more controllable situations that allow changing the stressful situation, whereas more indirect coping efforts are more appropriate in less controllable situations (for a review, see Compas et al., 2001). Studies that examine the controllability of stressful events (both objectively defined controllability and participants’ perceived controllability) are rare (see Zimmer-Gembeck & Skinner, 2016). There is initial evidence for situational influences on developmental changes in coping. Vierhaus et al. (2007) indicated situation-specific changes in coping, especially for problem solving, social support seeking, and avoidant coping. For problem solving, increases with age during childhood were found for more controllable stressors but not for uncontrollable or social problems. For avoidant coping, European Journal of Health Psychology (2018), 25(1), 18–30

H. Eschenbeck et al., Coping in Children and Adolescents

decreases were found when dealing with social stressors (see Zimmer-Gembeck & Skinner, 2011). Coping questionnaires for children and adolescents (i.e., stimulus-response inventories) often include normative social stressors (e.g., conflict with a friend) and normative academic stressors (e.g., difficult exam or too much homework; Causey & Dubow, 1992; Eschenbeck et al., 2007; Hampel & Petermann, 2005; for a review, see Compas et al., 2001) as ego-threatening situations. Consequently, a third research question is whether developmental changes in coping depend on the type of stressor. The present study examined developmental changes and age differences in coping in response to two different normative social or academic stressors (peer conflict and school problems). In line with Zimmer-Gembeck and Skinner (2011) situational influences on developmental changes were expected especially for avoidant coping (i.e., decreases for the social stressor peer conflict) and for problem solving (i.e., increases for the homework stressor but not for the social peer stressor).

Migration Background Differences in sociocultural contexts and ethnic backgrounds have also been linked to the choice of coping strategies (for an overview, see Yasui & Dishion, 2007). However, current research on culturally specific coping strategies is limited. Despite the increasing diversity of German schoolchildren and its potential influence on child and adolescent coping, a detailed analysis of sociocultural influence on coping and developmental changes in the ways that children and adolescents deal with everyday stressors has not been conducted. However, given the potential sociocultural influence of cultural background, in supplementary analyses, we additionally included migration background as a control variable for the studied developmental changes or age differences.

Present Study The present study performed a detailed analysis of age differences and developmental changes in central coping strategies (i.e., social support seeking, problem solving, avoidant coping, palliative emotion regulation, angerrelated emotion regulation, and media use) in middle/late childhood and early adolescence. First, we evaluated cohort and time effects in the ways of coping. In addition, we investigated whether developmental changes were modified by gender and the type of stressful situation (social and academic stressor). Finally, we controlled for migration background to test whether effects regarding age differences and developmental changes remain. Ó 2018 Hogrefe Publishing


H. Eschenbeck et al., Coping in Children and Adolescents

Method Participants and Procedure Participants were schoolchildren recruited from primary schools (grade levels 3 and 4) and secondary schools (grade levels 5 and above) in Schwäbisch GmĂźnd. Overall, 12 schools participated in the initial assessment (two grammar schools [Gymnasium], two secondary modern schools [Realschule], five primary and lower secondary schools [Grund- und Hauptschule], and three primary schools [Grundschule]). Participants and their parents provided their written informed consent prior to the start of the study. Children and adolescents completed a selfreport coping questionnaire in their classes. The measures were administered by trained students. The study was part of a project on stress, coping, and activity in schoolchildren (BUS [Bewegung und Umgang mit Stress] Study). The proposal of the study was approved by the Ethics Committee of the German Psychological Society. At the initial assessment, cross-sectional data were collected from 940 children and adolescents (50% girls, 43% with migration background1). A total of 917 children and adolescents (50% girls, 43% with migration background) had complete questionnaires (cross-sectional sample). Of these, 294 (32%) children came from primary schools, 623 (68%) from secondary schools. Within the secondary schools, 166 (27%) children came from lower secondary schools, 260 (42%) from secondary modern schools, and 197 (32%) from grammar schools. Children were from grades 3 to 7 with a mean age of 10.84 years (SD = 1.59, range 8â&#x20AC;&#x201C;15 years), including 167 third graders (52% girls), 127 fourth graders (50% girls), 252 fifth graders (52% girls), 198 sixth graders (47% girls), and 173 seventh graders (49% girls; see Figure 1). There were no differences between grade levels 3 and 7 regarding gender (p = .87) or migration background (p = .21). Three cohorts from the cross-sectional sample (third graders [cohort 3], fourth graders [cohort 4], and fifth graders [cohort 5]) were followed up longitudinally with a total of four measures at intervals of 6 months (T1â&#x20AC;&#x201C;T4) taking place between spring 2011 and fall 2012. Thus, children were in grades 3â&#x20AC;&#x201C;5 at T1, in grades 4â&#x20AC;&#x201C;6 at T2 and T3, and in grades 5â&#x20AC;&#x201C;7 at T4. Children were included in the longitudinal study if they had participated in all four assessments. The longitudinal sample consisted of 388 children and adolescents out of 546 children and adolescents from grades 3 to 5 at T1 (71% of the initial sample, 50% girls, 38% with a migration background), with

1

21

a mean age of 9.96 years (SD = 1.07, range 8â&#x20AC;&#x201C;13 years) at initial assessment (see Figure 1). There were no differences between cohorts regarding gender (p = .74) or migration background (p = .57). Comparisons (w2 test, analysis of variance [ANOVA]) between children with complete data for the four assessments (n = 388) and children who could not be followed over all four assessments (n = 158) did not show differences (either for the whole sample or for the three cohort subsamples) for gender (ps > .23) and migration background (ps > .14). However, children who dropped out between the assessments were more frequently third graders at the initial assessment (i.e., cohort 3, p < .001, especially for T4 after the school transition from primary to secondary school). With regard to the dependent variables almost no group differences were found between children included (n = 388) and those not included in the analyses (n = 158). Only in cohort 3, those who dropped out had lower scores in palliative emotion regulation than children with complete data (p < .001). No differences were found for the other coping strategies in cohort 3 nor for all six coping strategies in cohorts 4 and 5 (ps between .12 and .92).

Measures Coping strategies were assessed by a stimulus-response inventory, the revised German Stress and Coping Questionnaire for Children and Adolescents (SSKJ 3â&#x20AC;&#x201C;8; Lohaus, Eschenbeck, Kohlmann & Klein-HeĂ&#x;ling, 2006, see also Eschenbeck et al., 2006, 2010). Participants indicated on a 5-point scale ranging from never (1) to always (5) how often they used a variety of coping strategies in response to the following two common stressful situations: (a) having an argument with a friend (a social stressor) or (b) having problems with doing the homework (an academic stressor). Cross-situational coping scores across the two stressful situations (mean of both stressors) were computed, as well as situation-specific coping scores for the social stressor and the academic stressor. For each stress situation, 30 coping items were provided representing six dimensions of coping (i.e., 5 items per dimension):  social support seeking (e.g., â&#x20AC;&#x153;I ask someone for helpâ&#x20AC;?; the internal consistencies were between Cronbachâ&#x20AC;&#x2122;s Îą = .78 and .82 for the cross-situational coping subscale),  problem solving (e.g., â&#x20AC;&#x153;I try to think of different ways to solve itâ&#x20AC;?; between Îą = .85 and .92),

Migration background was not assessed in more detail. Census data for the region around Schwäbisch Gmßnd (where the study was carried out) report the most common countries of origin to be Turkey, Kazakhstan, Russia, Poland, and Romania (with decreasing frequency; State Statistics Office of Baden-Wßrttemberg, 2011). These countries are the origin of two thirds of the individuals with a migrant background in the abovementioned area.

Ă&#x201C; 2018 Hogrefe Publishing

European Journal of Health Psychology (2018), 25(1), 18â&#x20AC;&#x201C;30


22

H. Eschenbeck et al., Coping in Children and Adolescents

n = 940 at T1 Excluded from the analysis: n = 23 due to incomplete data

Figure 1. Sampling procedure for cross-sectional and longitudinal analyses. Grade level refers to the initial assessment (T1). c3 = cohort 3, c4 = cohort 4, c5 = cohort 5.

Cross-sectional sample n = 917 at T1 grade 3: n = 167 grade 4: n = 127 grade 5: n = 252 grade 6: n = 198 grade 7: n = 173

6th and 7th graders were not investigated longitudinally

n = 546 at T1 from grades 3 to 5 for longitudinal analysis

Longitudinal sample n = 388 with T1, T2, T3, T4 cohort 3 (grade 3): n = 89 cohort 4 (grade 4): n = 101 cohort 5 (grade 5): n = 198

Excluded from the analysis: n = 158 due to missing data (drop out) during longitudinal data collection

T2: n = 50 (c3: 6; c4: 16; c5: 28) T3: n = 20 (c3: 5; c4: 4; c5: 11) T4: n = 88 (c3: 67; c4: 6, c5: 15)

 avoidant coping (e.g., â&#x20AC;&#x153;I tell myself it doesnâ&#x20AC;&#x2122;t matterâ&#x20AC;?; between Îą = .71 and .81),  palliative emotion regulation (e.g., â&#x20AC;&#x153;I try to relaxâ&#x20AC;?; between Îą = .84 and .90),  anger-related emotion regulation (e.g., â&#x20AC;&#x153;I get mad and break somethingâ&#x20AC;?; between Îą = .80 and .85), and  media use (e.g., â&#x20AC;&#x153;I go onlineâ&#x20AC;?; between Îą = .76 and .83).

Statistical Analyses Missing values were imputed with the Predictive Mean Matching method using the R package mice (Multivariate Imputation by Chained Equations; van Buuren & Groothuis-Oudshoorn, 2011) if less than 10% of the values were missing for each time point. Cross-Sectional Data To analyze age differences in coping strategies at T1 taking gender and stressful situation into account, six 5  2 repeated-measures ANOVAs with grade level (grades 3â&#x20AC;&#x201C;7) and gender as between-subject factors and situation (social stressor, academic stressor) as the within-subject factor were conducted separately for each coping strategy (social support seeking, problem solving, avoidant coping, palliative emotion regulation, anger-related emotion regulation, European Journal of Health Psychology (2018), 25(1), 18â&#x20AC;&#x201C;30

and media use) as a dependent variable. For significant main effects and interactions post hoc tests were applied using the estimated marginal means option (Bonferroni adjusted). Follow-up ANOVAs were conducted with the independent variables (grade level, gender, situation) described above and migration background (with/without migration background) as an additional between-subject factor to test whether significant effects remain when migration background was controlled for. A migration background was assumed when at least one parent was not born in Germany. Longitudinal Data To investigate developmental changes in coping over time in parallel to the cross-sectional analysis separately for each coping strategy, six 3  2 repeated-measures ANOVAs with cohort (cohorts 3â&#x20AC;&#x201C;5) and gender as between-subject factors and situation (social stressor, academic stressor) and time (T1â&#x20AC;&#x201C;T4) as within-subject factors were performed. For significant main effects and interactions post hoc tests were applied using the estimated marginal means option (Bonferroni adjusted). Follow-up ANOVAs were conducted with the independent variables described above and migration background (with/without migration background) as an additional between-subject factor to test whether Ă&#x201C; 2018 Hogrefe Publishing


H. Eschenbeck et al., Coping in Children and Adolescents

significant effects remain when migration background was controlled for.

Results Cross-Sectional Data Results for the cross-sectional coping data with grade level, gender, and situation as independent variables are shown in Table 1. Grade Level With reference to Research Question 1, we first examined age-related differences in the ways of coping among children and young adolescents. The analyses showed significant main effects for grade level in four out of six coping strategies (see Figure 2). In social support seeking, seventh graders scored lower than fourth and fifth graders (ps < .05 Bonferroni adjusted). In problem solving, fifth and seventh graders scored higher than third graders (ps  .01 Bonferroni adjusted). In avoidant coping, seventh graders scored lower than third graders (p < .05 Bonferroni adjusted). In media use as a coping strategy, seventh graders scored especially higher than the younger children (from grades 3 to 5; ps < .05 Bonferroni adjusted). For palliative emotion regulation and anger-related emotion regulation, no significant effect for grade level was found. Stressful Situation and Gender The six coping strategies (social support seeking, problem solving, avoidant coping, palliative emotion regulation, anger-related emotion regulation, and media use) varied as a function of situation (see Figure 2). Children and adolescents generally reported higher scores in the social situation (argument with a friend) than in the academic situation (problems with homework). Significant main effects for gender were found for social support seeking and problem solving, with higher scores for girls, whereas boys yielded higher scores in avoidant coping, palliative emotion regulation, and media use as coping strategies. A significant interaction between gender and situation was observed for social support seeking, problem solving, and avoidant coping; gender differences (throughout significant, ps  .01) were stronger for the social argument situation compared to the academic homework situation. Interactions Between Grade Level, Stressful Situation, and Gender Regarding Research Questions 2 and 3, we took into account the role of gender and stressful situation for age-related differences in coping. For two coping strategies, problem Ă&#x201C; 2018 Hogrefe Publishing

23

solving and media use, the main effects for grade level were modified by a significant interaction between grade level and situation (see Figure 2). For problem solving, differences between younger and older children were stronger in the academic homework situation (p < .001) compared to the social argument situation (p = .03), with third graders scoring lowest in problem solving in the homework situation (ps < .02 Bonferroni adjusted). In addition, a three-way interaction among grade level, gender, and situation resulted. For the social argument situation, significant gender differences were found with girls reporting higher scores than boys at all grade levels (p < .05 Bonferroni adjusted). As described above, for the academic homework situation, the findings revealed significant grade level differences. Boys and girls in grade 3 reported the lowest scores in problem solving compared to fourth (boys only, p < .01 Bonferroni adjusted), fifth and seventh graders (boys and girls, ps < .03 Bonferroni adjusted). Gender differences were limited to fifth and sixth graders, with girls reporting higher scores. Media use as a coping strategy was reported most frequently by the older group (i.e., grades 6 and 7), but differences were stronger, and the level was higher for the social argument situation compared to the academic homework situation. Migration Background Finally, after repeating the ANOVAs with migration background as an additional independent variable, the above-described main effects and interactions remained significant, except for the main effect for grade level in avoidant coping (p = .08). For three coping strategies (social support seeking, palliative emotion regulation, media use), migration background showed additional significant effects. In all cases, the effect size was below Ρ2 = .02. Thus, effects of migration background were minor.

Longitudinal Data Results for the longitudinal coping data are presented in Table 2. Developmental changes for the coping strategies are shown in Figure 3. Developmental Trends With reference to Research Question 1, we first investigated developmental changes in coping. Time effects and/or cohort effects indicating developmental differences in coping were shown for five out of six coping strategies: social support seeking, problem solving, avoidant coping, angerrelated emotion regulation, and media use. In general, with regard to time effects and/or cohort effects, two- and/or three-way interactions with stressful situation (i.e., Cohort  Situation, Time  Situation, Cohort  Time  Situation; see Research Question 3) were significant for European Journal of Health Psychology (2018), 25(1), 18â&#x20AC;&#x201C;30


24

H. Eschenbeck et al., Coping in Children and Adolescents

Table 1. Summary of ANOVA effects (F-values) and effect sizes (η2) for the cross-sectional data at T1

Coping strategies Factor

Seeking social support F

η

2

Problem solving F

η

2

Avoidant coping F

η

2

Palliative emotion regulation F

η

2

Anger-related emotion regulation F

η

2

Media use F

η2

Grade

4.43**

.019

4.99***

.022

2.70*

.012

1.08

.005

0.76

.003

9.38***

.040

Gender

108.87***

.107

26.97***

.029

41.33***

.044

7.30**

.008

1.17

.001

27.87***

.030

Situation

9.05**

.010

18.68***

.020

186.19***

.170

9.14**

.010

.079

283.75***

.234

Grade  Gender

0.30

.001

0.72

.003

2.08

.009

0.84

.004

0.46

.002

0.79

.003

Grade  Situation

1.25

.005

6.11***

.026

2.24

.010

0.42

.002

1.60

.007

3.77**

.016

.042

14.54***

.016

.012

0.23

.000

0.55

.001

0.08

.000

.002

2.50*

.011

.005

1.20

.005

1.32

.006

1.55

.007

Gender  Situation Grade  Gender  Situation

40.02*** 0.53

11.43*** 1.21

77.65***

Notes. N = 917. *p < .05, **p < .01, ***p  .001.

Figure 2. Coping strategies in the cross-sectional assessment as a function of grade level, gender (& = boys, h = girls), and situation (social: argument with a friend, academic: homework). Depicted are the estimated marginal means.

European Journal of Health Psychology (2018), 25(1), 18–30

Ó 2018 Hogrefe Publishing


H. Eschenbeck et al., Coping in Children and Adolescents

25

Figure 3. Developmental trends for the six coping strategies in longitudinal assessments separately for situation (social: argument with a friend, academic: homework), gender and cohort (& = 3rd cohort boys, d = 4th cohort boys, N = 5th cohort boys, h = 3rd cohort girls, s= 4th cohort girls, Δ = 5th cohort girls). Depicted are the estimated marginal means.

problem solving, avoidant coping, anger-related emotion regulation, and media use. Regarding gender, the corresponding two- and three-way interactions (i.e., Cohort  Gender, Time  Gender, Cohort  Time  Gender, Time  Situation  Gender; see Research Question 2) were not significant. For three strategies (social support seeking, problem solving, and anger-related emotion regulation), the four-way interactions between cohort, time, gender, and situation were significant with effect sizes below η2 = .02. In all of the follow-up tests, Bonferroni adjusted interaction effects with gender were insignificant. Thus, while developmental trends were influenced by the stressful

Ó 2018 Hogrefe Publishing

situation (Research Question 3), they were independent of the child’s gender (Research Question 2). In the following section, interactions of cohort or time with gender are not considered. For social support seeking, Cohort  Time interactions indicate an increase within the youngest cohort (i.e., children in grades 3–5; see Figure 3). Regarding problem solving, effects for time and cohort were significant (interaction with situation) only for the academic homework situation (contrary to the social argument situation). Thus, for problem solving, results showed positive linear trends for time and cohort indicating an increase in the use of

European Journal of Health Psychology (2018), 25(1), 18–30


26

H. Eschenbeck et al., Coping in Children and Adolescents

Table 2. Summary of ANOVA effects (F-values) and effect sizes (η2) for the longitudinal data (T1–T4)

Coping strategies

Seeking social support

Problem solving

η2

Avoidant coping η2

0.03

.000

1.44

.007

1.66

6.14*** .046

1.43

.011

9.32*** .069

6.66*** .050

0.33

.001

0.34

8.61**

Cohort

1.19

.006

Time

0.51

.004 11.45*** .083

74.22*** .163 20.34*** .051

30.48*** .074

Gender

1.88

.010

F

η2

Media use

η2

F

η2

Anger-related emotion regulation

F

Factor

F

Palliative emotion regulation

F

.009 .001

F

η2

2.54

.013

Situation

4.53*

.012

0.72

Cohort  Time

2.25*

.017

3.91*** .030

3.12**

.024

0.70

.005

3.44**

.026

1.31

.022

.002 206.07*** .350 16.29*** .041 21.82*** .054 228.88*** .375 .010

Cohort  Gender

0.03

.000

0.57

.003

0.08

.000

1.81

.009

0.51

.003

0.10

.001

Cohort  Situation

1.81

.009

5.19**

.026

1.18

.006

0.05

.000

0.18

.001

4.14*

.021

1.53

.012

Time  Gender

0.69

.005

Time  Situation

0.59

.005 11.49*** .083

Gender  Situation

28.82*** .070

0.21

.002

0.39

.003

0.86

.007

2.42

.019

3.34*

.026

0.97

.008

3.23*

.025

0.30

.002

11.92*** .030

1.50

.004

2.56

.007

2.11

.006

0.81

.002

Cohort  Time  Gender

1.08

.008

0.53

.012

0.79

.006

0.98

.008

0.40

.003

0.64

.005

Cohort  Time  Situation

1.12

.009

5.20*** .039

0.84

.007

1.60

.012

1.13

.009

2.57*

.020

Cohort  Gender  Situation

1.21

.006

1.65

.009

1.46

.008

0.50

.003

0.91

.005

0.03

.000

Time  Gender  Situation

2.57

.020

0.85

.007

0.51

.004

1.78

.014

1.35

.011

0.21

.002

Cohort  Time  Gender  Situation

2.12*

.016

2.47*

.019

0.74

.006

1.31

.010

2.19*

.017

1.34

.010

Notes. N = 388. *p < .05, **p < .01, ***p  .001.

problem solving with age. Problem solving was less frequently used in the youngest cohort (i.e., cohort 3 compared to cohorts 4 and 5). Moreover, there was a Cohort  Time interaction with the youngest cohort (i.e., children in grades 3–5) showing the strongest increase over time (see Figure 3). For avoidant coping and anger-related emotion regulation, the trend is negative for time. Both strategies were less frequently used over time, especially for the argument situation (Time  Situation interaction). Cohort  Time interactions indicate that the decrease in use of avoidant coping and anger-related emotion regulation was especially strong for the youngest cohort (i.e., children in grades 3–5; see Figure 3). For media use as a coping strategy, the trend is positive for time. The increase in media use was especially strong in the oldest cohort (i.e., children in grades 5–7) for the social argument situation (see Figure 3). Stressful Situation and Gender Except for problem solving, the coping strategies (social support seeking, avoidant coping, palliative emotion regulation, anger-related emotion regulation, and media use) varied as a function of situation (see Table 2). Children and adolescents generally reported higher scores in the social situation (argument with a friend) than in the academic situation (problems with homework, see Figure 3). Significant effects for gender were found for social support seeking (with stronger differences for the social argument situation; Gender  Situation interaction) and problem solving, with higher scores for girls. Boys showed higher

European Journal of Health Psychology (2018), 25(1), 18–30

scores in avoidant coping (especially in the social argument situation; Gender  Situation interaction) and media use as means of coping (see Figure 3). Migration Background Finally, when ANOVAs were repeated with migration background as an additional independent variable, the above-described main effects and interactions remained significant, except for the two Cohort  Situation interactions for problem solving (p = .06) and media use (p = .06). For four coping strategies (social support seeking, problem solving, avoidant coping, media use), migration background showed single additional significant effects. In all cases, the effect size was below η2 = .03. Thus, effects of migration background were negligible.

Discussion Developmental Changes: Correspondence Between Cross-Sectional and Longitudinal Findings The cross-sectional and longitudinal data largely corresponded regarding developmental differences or changes, especially increases for problem solving (with small to medium effect sizes) and decreases for avoidant coping strategies (yielding small effect sizes). The direction of

Ó 2018 Hogrefe Publishing


H. Eschenbeck et al., Coping in Children and Adolescents

these effects confirmed previous cross-sectional findings (see Zimmer-Gembeck & Skinner, 2011). More specifically, for problem solving, this age difference was particularly marked in the academic homework situation contrary to the more interpersonal social argument situation. Avoidant coping was less frequently used over time, especially for the social argument situation. Overall, changes in coping (with increases for problem solving and declines in avoidant coping) were especially strong in the youngest cohort (3rd–5th graders), representing an age window of particular importance. Also Vierhaus et al. (2007) reported similar increases for problem solving and decreases for avoidant coping in their longitudinal analyses. The transition from primary to secondary school seems to be crucial for these developmental changes. However, also within the same school environment, Ben-Eliyahu and Kaplan (2015) showed a longitudinal increase in academic positive coping (a combination of problem-focused and positive reframing strategies) from third to fifth grade. These results support the view that middle/late childhood to early adolescence (approximately ages 9–11, grades 3–5) marks a relevant age period for a variety of changes (e.g., cognitive, emotional, social; see Zimmer-Gembeck & Skinner, 2011), including changes in central coping strategies. Consequently, an increase in problem solving along with a decrease in avoidant coping allows for a more adaptive engagement and problem-focused coping. Cross-sectional and longitudinal results were also similar for palliative emotion regulation (related to relaxation and distraction), identifying no developmental changes or age differences. This is in line with Zimmer-Gembeck and Skinner (2011) and also with Vierhaus et al. (2007) for middle to late childhood. Finally, in accordance with a general increase in electronic media usage from younger to older children and adolescents (e.g., MPFS, 2016, 2017; Robert Koch Institute, 2015), media use as a coping strategy was reported more frequently in adolescents (grades 6 and above), especially in the social situation (with small effect sizes). Similar effects were shown by Eschenbeck et al. (2010). Longitudinal data confirmed increases, especially in the older cohort (grades 5–7). Media use may be a helpful strategy to distract from the immediate impact of a stressful social interaction (Knobloch, 2003).

Differences Between Cross-Sectional and Longitudinal Findings Age effects or developmental patterns were less clear for social support seeking and anger-related emotion regulation. For social support seeking, cross-sectional data showed higher scores for fourth and fifth graders compared

Ó 2018 Hogrefe Publishing

27

to early adolescents (seventh graders). However, longitudinally, a slight increase within the youngest cohort (grades 3–5) was found (in each case yielding small effect sizes), yet not representing a contradiction. Anger-related emotion regulation showed a decline across time (especially in the youngest cohort of third to fifth graders; with small to medium effect sizes) and cross-sectional data found no age differences. These patterns are compatible in part with the results of Vierhaus et al. (2007), who reported lower levels in support seeking in older children and decreases for externalizing coping, the latter especially for younger children. With an increase in problem-focused coping (see above), the importance of palliative strategies seems to diminish for stressful events with a friend or demands of school. With regard to social support a differentiation in future research between emotional support and instrumental support seems to be helpful for getting a clearer picture.

Influencing Factors While age differences and developmental changes for problem solving, avoidant coping, and media use differed between the two studied stressful situations (argument with a friend and problems with homework), the assumption that developmental changes in coping strategies differed for boys and girls was not strongly confirmed. Crosssectional increases slightly differed for boys and girls only for problem solving. Overall, during middle/late childhood and early adolescence, gender-specific developmental changes in coping were not of great significance. This finding is unexpected because gender differences seem to become more evident in adolescence compared to childhood (see Eschenbeck et al., 2007). This suggests that future research on developmental changes in coping should incorporate children younger than those in this study (approximately 9 years of age, grade 3), especially with regard to the two strategies of social support seeking and avoidant coping. Considering migration background as an additional influencing factor, the documented developmental changes or age differences in coping do not seem to have been affected. To clarify the developmental patterns documented in the present study (that altogether fit into the pattern reviewed by Zimmer-Gembeck & Skinner, 2011 for very heterogeneous studies), future research should take into account individual and contextual factors (in addition to gender, stressful situation, and migration background) that might influence developmental changes or stability. For example, what is the impact of the new school context (e.g., best friends move away, academic challenges), after

European Journal of Health Psychology (2018), 25(1), 18–30


28

the school transition from primary to secondary school for changes or stabilities in the child’s coping?

Gender Differences In accordance with prior research (e.g., Eschenbeck et al., 2007), gender differences in coping were clearly supported. Girls scored higher in social support seeking (with medium to large effect sizes) and problem solving, while boys scored higher in avoidant coping (for the latter two strategies with small to medium effect sizes). In parallel, these three gender differences (except for problem solving in the longitudinal sample) were also stronger for the social argument situation than for the academic homework situation. Also in line with our earlier study (Eschenbeck et al., 2007), no gender differences occurred for anger-related emotion regulation for the cross-sectional and longitudinal data, respectively. A more heterogeneous pattern emerged for palliative emotion regulation. Contrary to Vierhaus et al. (2007) and Eschenbeck et al. (2007), in the present study, boys reported higher palliative emotion regulation in the cross-sectional sample (with small effect size). In line with the overall use of electronic media in adolescence, that boys spend more time on television and game consoles (Robert Koch Institute, 2015), as well as the result that boys reported more indirect efforts to cope (i.e., avoidance, palliative emotion regulation), media use as a way of coping was reported more frequently from boys (with small effect size; see also Eschenbeck et al., 2010).

Limitations and Strengths There are limitations in the present study that must be noted. First, our study focused on changes in coping within small gaps of 6 months. Thus, the 6-month versus 1-year gaps differed between the longitudinal and cross-sectional approach. However, the included age range of third to seventh graders was comparable. Second, the dropout rate of fifth graders was high after the school transition from primary to secondary school at the last time point (i.e., third graders at initial assessment). However, within the cohort of third graders (at baseline), analyses did not document a bias between those who dropped out and those who did not. There were no differences regarding gender and migration background. For coping strategies, the only difference found was for palliative emotion regulation, for which no age differences or developmental changes were shown in our study. Finally, as is common in coping research (Compas et al., 2001), the findings are based on self-reported data, which may not be consistent with the child’s actual behavior in the stressful situations. However,

European Journal of Health Psychology (2018), 25(1), 18–30

H. Eschenbeck et al., Coping in Children and Adolescents

major strengths of this study are the longitudinal design (in contrast to the literature reporting mostly cross-sectional findings), which allows studying developmental trends in coping over four points in time in three cohorts covering ages between approximately 9 and 13 years, as well as the simultaneous cross-sectional approach, which allows comparisons between longitudinal and cross-sectional data in children’s coping. Additional strengths of this study include the focus on middle/late childhood as a relevant age period for developmental changes in coping, the realization of small gaps between the four assessments to locate potential changes in coping, and the assessment of more central strategies (see Compas et al., 2001; Skinner et al., 2003) used to cope with two specific stressful situations (contrary to more general coping styles).

Conclusions and Future Directions To summarize, this study provides a detailed description of developmental changes in coping during middle/late childhood and early adolescence. Within the studied coping strategies, developmental changes were especially clear for problem solving and avoidant coping in the youngest cohort of children between 9 and 11 years, representing an age period of high relevance for changes in coping. Moreover, age changes in coping differed for the two studied normative stressful situations (social, academic) but were largely independent of the child’s gender and migration background. Thus, increases in problem solving were especially evident in the academic homework situation, whereas declines in avoidant coping were especially evident in the social argument situation. Besides documenting age differences and developmental changes in coping in children and young adolescents and identifying influencing factors, future studies should explore the (possibly agerelated) association between developmental changes in coping with psychological adjustment and health. Thereby, the stress process is complex and dynamic and involves the person, the environmental context, and the relationship between them (e.g., Folkman & Moskowitz, 2004) taking changes over time into consideration in terms of a developmental life-span perspective. Even though, it is conceptually useful to differentiate between different coping strategies, the strategies are also associated and intertwined with each other. Thus, multiple coping strategies or patterns of coping show relations with adjustment and health (e.g., Russell et al., 2015). Regarding developmental changes in coping in children and young adolescents, future studies might also examine changes in coping patterns. Possibly a pattern that is characterized by increased use of engagement, problem-focused strategies and decreased use of avoidant coping might be effective and adaptive

Ó 2018 Hogrefe Publishing


H. Eschenbeck et al., Coping in Children and Adolescents

for managing a broad range of social and academic challenges in the course of life. Acknowledgment We thank Faith Simpson for providing language help. This work was supported by the Ministry of Science, Research and the Arts of the State of Baden-Württemberg, Germany [Grant #43-871.98/200].

References Ben-Eliyahu, A., & Kaplan, A. (2015). Growth curve modeling analysis of social and academic coping during elementary school. Journal of Applied Developmental Psychology, 41, 99–109. https://doi.org/10.1016/j.appdev.2015.09.001 Byrne, D. G., Thomas, K. A., Burchell, J. L., Olive, L. S., & Mirabito, N. S. (2011). Stressor experience in primary school-aged children: Development of a scale to assess profiles of exposure and effects on psychological well-being. International Journal of Stress Management, 18, 88–111. https://doi.org/10.1037/ a0021577 Causey, D. L., & Dubow, E. F. (1992). Development of a self-report coping measure for elementary school children. Journal of Clinical Child Psychology, 21, 47–59. https://doi.org/10.1207/ s15374424jccp2101_8 Compas, B. E., Connor-Smith, J. K., Saltzman, H., Thomsen, A., & Wadsworth, M. E. (2001). Coping with stress during childhood and adolescence: Problems, progress, and potential in theory and research. Psychological Bulletin, 127, 87–127. https://doi. org/10.1037/0033-2909.127.1.87 Eschenbeck, H., Heim-Dreger, U., Tasdaban, E., Lohaus, A., & Kohlmann, C.-W. (2012). A Turkish adaptation of the coping scales from the German Stress and Coping Questionnaire for Children and Adolescents. European Journal of Psychological Assessment, 28, 32–40. https://doi.org/10.1027/1015-5759/ a000088 Eschenbeck, H., Kohlmann, C.-W., Lohaus, A., & Klein-Heßling, J. (2006). Die Diagnostik von Stressbewältigung mit dem ,,Fragebogen zur Erhebung von Stress und Stressbewältigung im Kindes- und Jugendalter“ (SSKJ 3–8): Faktorielle und psychometrische Analysen [The assessment of coping with the “Questionnaire for the Measurement of Stress and Coping in Children and Adolescents (SSKJ 3–8)”: Factorial and psychometric analyses]. Diagnostica, 52, 131–142. https://doi.org/ 10.1026/0012-1924.52.3.131 Eschenbeck, H., Kohlmann, C.-W., & Lohaus, A. (2007). Gender differences in coping strategies in children and adolescents. Journal of Individual Differences, 28, 18–26. https://doi.org/ 10.1027/1614-0001.28.1.18 Eschenbeck, H., Kohlmann, C.-W., & Meier, S. (2010). Mediennutzung als Bewältigungsstrategie von Kindern und Jugendlichen: Eine Erweiterung des SSKJ 3–8 [Media use as a coping strategy for children and adolescents: An addition to the SSKJ 3–8]. Zeitschrift für Gesundheitspsychologie, 18, 183–189. https://doi.org/10.1026/0943-8149/a000019 Folkman, S., & Moskowitz, J. T. (2004). Coping: Pitfalls and promise. Annual Review of Psychology, 55, 745–774. https:// doi.org/10.1146/annurev.psych.55.090902.141456 Hampel, P., & Petermann, F. (2005). Age and gender effects on coping in children and adolescents. Journal of Youth and Adolescence, 34, 73–83. https://doi.org/10.1007/s10964-0053207-9

Ó 2018 Hogrefe Publishing

29

Havighurst, R. J. (1972). Developmental tasks and education (3rd ed.). New York, NY: McKay. Knobloch, S. (2003). Mood adjustment via mass communication. Journal of Communication, 53, 233–250. https://doi.org/ 10.1111/j.1460-2466.2003.tb02588.x Krohne, H. W. (1993). Vigilance and cognitive avoidance as concepts in coping research. In H. W. Krohne (Ed.), Attention and avoidance: Strategies in coping with aversiveness (pp. 19–50). Seattle, WA: Hogrefe & Huber. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY: Springer. Leiner, M., Argus-Calvo, B., Peinado, J., Keller, L., & Blunk, D. I. (2014). Is there a need to modify existing coping scales to include using electronic media for coping in young people? Frontiers in Pediatrics, 2, 127. https://doi.org/10.3389/fped.2014.00127 Lerner R. M. (Ed.). (2015). Handbook of child psychology and developmental science (7th ed.). New York, NY: Wiley. Lohaus, A., Ball, J., Klein-Heßling, J., & Wild, M. (2005). Relations between media use and self-reported symptomatology in young adolescents. Anxiety, Stress, and Coping, 18, 333–341. https://doi.org/10.1080/10615800500258123 Lohaus, A., Eschenbeck, H., Kohlmann, C.-W., & Klein-Heßling, J. (2006). Fragebogen zur Erhebung von Stress und Stressbewältigung im Kindes- und Jugendalter (SSKJ 3–8) [German Stress and Coping Questionnaire for Children and Adolescents (SSKJ 3–8)]. Göttingen, Germany: Hogrefe. Meier, S., Kohlmann, C.-W., Eschenbeck, H., & Gross, C. (2010). Coping in children and adolescents with obesity: The costs and benefits of realistic versus unrealistic weight-evaluations. Applied Psychology: Health and Well-Being, 2, 222–240. https://doi.org/10.1111/j.1758-0854.2010.01034.x MPFS Pedagogical Media Research Centre Southwest Medienpädagogischer Forschungsverbund Südwest. (2016). JIM Study 2016: Basis data on the media use of youths in Germany. Retrieved from https://www.mpfs.de/fileadmin/files/Studien/ JIM/2016/JIM_Studie_2016.pdf MPFS Pedagogical Media Research Centre Southwest Medienpädagogischer Forschungsverbund Südwest. (2017). KIM Study 2016: Media behaviour of children between the ages of six and 13 years in Germany. Retrieved from https://www.mpfs. de/fileadmin/files/Studien/KIM/2016/KIM_2016_Web-PDF.pdf Ofcom. (2016). Children and parents: Media use and attitudes report. Retrieved from https://www.ofcom.org.uk/__data/assets/pdf_ file/0034/93976/Children-Parents-Media-Use-Attitudes-Report2016.pdf Robert Koch Institute. (Eds.). (2015). Mediennutzung. Faktenblatt zu KiGGS Welle 1: Studie zur Gesundheit von Kindern und Jugendlichen in Deutschland – Erste Folgebefragung 2009–2012 [Media use. Factsheet on KiGGS Wave 1: German Health Interview and Examination Survey for Children and Adolescents – First follow-up interview 2009–2012]. Berlin, Germany: RKI. Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with stress. The American Psychologist, 41, 813–819. https:// doi.org/10.1037/0003-066X.41.7.813 Russell, H. F., January, A. M., Kelly, E. H., Mulcahey, M. J., Betz, R. R., & Vogel, L. C. (2015). Patterns of coping strategy use and relationships with psychosocial health in adolescents with spinal cord injury. Journal of Pediatric Psychology, 40, 535–543. https://doi.org/10.1093/jpepsy/jsu159 Skinner, E. A., Edge, K., Altman, J., & Sherwood, H. (2003). Searching for the structure of coping: A review and critique of category systems for classifying ways of coping. Psychological Bulletin, 129, 216–269. https://doi.org/10.1037/0033-2909.129.2.216 State Statistics Office of Baden-Württemberg. (2011). Zensus 2011. Bevölkerung und Haushalte – Landkreis Ostalbkreis [Population and Housing Censuses 2011 – Administrative

European Journal of Health Psychology (2018), 25(1), 18–30


30

District Ostalb]. Retrieved from www.destatis.de/GPStatistik/ servlets/MCRFileNodeServlet/BWHeft_derivate_00003600/ 08136_Ostalbkreis_Bev.pdf;jsessionid= 879181224B6383EE91088073E0800521 van Buuren, S., & Groothuis-Oudshoorn, K. (2011). mice: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45, 1–67. https://doi.org/10.18637/jss.v045.i03 Vierhaus, M., & Lohaus, A. (2009). Children’s perception of relations between anger or anxiety and coping: Continuity and discontinuity of relational structures. Social Development, 18, 747–763. https://doi.org/10.1111/j.1467-9507.2008.00504.x Vierhaus, M., Lohaus, A., & Ball, J. (2007). Developmental changes in coping: Situational and methodological influences. Anxiety, Stress, and Coping, 20, 267–282. https://doi.org/10.1080/ 10615800701330242 Yasui, M., & Dishion, T. J. (2007). The ethnic context of child and adolescent problem behavior: Implications for child and family interventions. Clinical Child and Family Psychology Review, 10, 137–179. https://doi.org/10.18637/jss.v045. i0310.1007/s10567-007-0021-9 Zimmer-Gembeck, M. J., & Skinner, E. A. (2011). The development of coping across childhood and adolescence: An integrative

European Journal of Health Psychology (2018), 25(1), 18–30

H. Eschenbeck et al., Coping in Children and Adolescents

review and critique of research. International Journal of Behavioral Development, 35, 1–17. https://doi.org/10.1177/ 0165025410384923 Zimmer-Gembeck, M. J., & Skinner, E. A. (2016). The development of coping: Implications for psychopathology and resilience. In D. Cicchetti (Ed.), Developmental psychopathology (pp. 485–545). New York, NY: Wiley. Received March 31, 2017 Revision received November 15, 2017 Accepted November 17, 2017 Published online January 15, 2018 Heike Eschenbeck Department of Psychology University of Education Schwäbisch Gmünd Oberbettringer Str. 200 73525 Schwäbisch Gmünd Germany heike.eschenbeck@ph-gmuend.de

Ó 2018 Hogrefe Publishing


News and Announcements Call for Papers

The European Journal of Health Psychology was founded to provide a platform for research in Health Psychology, and for its application in a wide range of contexts. Health Psychology is the scientific discipline within Psychology that aims to promote and preserve health, to prevent diseases and contribute to their treatment by identifying diseaserelevant aetiological processes, and to improve health provision. The European Journal of Health Psychology strives to promote theory and practice in the analysis of psychological approaches to health and disease. Its aim is, therefore, to publish high quality empirical or experimental research as well as sound practice-oriented articles, current methodological developments, and comprehensive critical reviews of the scientific literature. The journal invites you and/or your working group to submit papers on all aspects of the field!

Ă&#x201C; 2018 Hogrefe Publishing

EJHP publishes the following types of article: Original Articles, Review Articles, Short Reports, and Commentaries. Manuscripts, including Electronic Supplementary Material (ESM), are to be submitted via the online submission system: http://www.editorialmanager.com/zgp EJHP appears in 4 issues per year, in print as well as online at econtent.hogrefe.com/loi/zgp and in PsycARTICLES, where the journal is available to up to 44 million institutional users per year! For additional information, please contact the Editor-inChief, Claus VĂśgele (zfgesundheitspsychologie@uni.lu).

European Journal of Health Psychology (2018), 25(1), 31 https://doi.org/10.1027/2512-8442/a000006


News and Announcements Meeting Calendar March 16–17, 2018 PUBLIC HEALTH '18 / II. International Conference on Public Health Istanbul, Turkey Contact: Ozgur Ozturk, Istanbul, Turkey, E-mail conference@dakam. org, Web https://www.dakamconferences.org/public-health

July 4–6, 2018 4th International Conference on Health and Health Psychology Viseu, Portugal Contact: Carlos Pereira, Conference Chair. E-mail essvgeral@essv.ipv.pt, Web http://www.futureacademy.org. uk/conference/icH&Hpsy/

April 5–6, 2018 2018 Spring Global Symposium for Psychology Professionals Los Angeles, CA, USA Contact: Conference Coordinator, University of Riverside, Riverside, CA, USA, E-mail psychconference@ uofriverside.com, Web http://www. uofriverside.com/conferences/ globalpsychologysymposium/2018spring-psychology-symposium/

August 21–22, 2018 Promoting Scientific Integrity in Health Psychology Research and Publishing – Synergy Expert Meeting Galway, Ireland Contact: European Health Psychology Society, E-mail secretary@ehps.net, Web https://www.ehps.net/synergy/ news

June 11–12, 2018 2nd ICHLSR Italy – International Conference on Healthcare & LifeScience Research Rome, Italy Contact: International Association for Promotion of Healthcare and LifeScience Research, Jaipur, India, E-mail info@iaphlsr.org, Web https://iaphlsr.org/2nd-ichlsr-italy— international-conference-on-healthcare-and-life-science-research-11june—12-june-2018-about-77

September 5–7, 2018 2018 DHP Annual Conference Newcastle-upon-Tyne, UK Contact: British Psychological Society, Division of Health Psychology, Leicester, UK, E-mail enquiries@bps. org.uk, Web http://www.bps.org.uk/ networks-and-communities/membermicrosite/division-health-psychology/ conferences September 3–7, 2019 33rd Annual Conference of the European Health Psychology Society Dubrovnik, Croatia Contact: European Health Psychology Society, E-mail secretary@ehps.net, Web http://www.ehps.net/conferences

August 21–25, 2018 32nd Annual Conference of the European Health Psychology Society Galway, Ireland Contact: European Health Psychology Society, E-mail secretary@ehps.net, Web http://www.ehps2018.net/

European Journal of Health Psychology (2018), 25(1), 32 https://doi.org/10.1027/2512-8442/a000004

Ó 2018 Hogrefe Publishing


The most effective medical, psychological and neuropsychological methods and interventions for multiple sclerosis “I just wish such a resource had been available when I was a young psychologist starting to see MS patients.” Rosalind Kalb, PhD, Vice President, Healthcare Information & Resources, Advocacy, Services and Research Department, National Multiple Sclerosis Society, New York, NY

Pearl B. Werfel / Ron E. Franco Durán / Linda J. Trettin

Multiple Sclerosis (Series: Advances in Psychotherapy – Evidence-Based Practice – Vol. 36) 2016, viii + 128 pp. US $29.80 / € 24.95 ISBN 978-0-88937-409-6 Also available as eBook This innovative book will help both mental health and medical professionals empower patients or clients to live well with multiple sclerosis (MS). It is a practical, evidence-based, culturally relevant guide to the most effective current medical, psychological, and neuropsychological diagnostic methods and interventions. The book describes a biopsychosocial,

www.hogrefe.com

multidisciplinary, and integrative approach to treatment and provides information on psychological, mindbody, and complementary interventions for symptom management and to increase quality of life. Both seasoned practitioners and students will find this volume useful in helping clients cope with this complex, unpredictable, and chronic neurological disorder.


Bewegung ist Leben!

Annette Höhmann-Kost (Hrsg.)

Integrative Leib- und Bewegungstherapie (IBT) Theorie und Praxis 3., akt. u. erg. Aufl. 2018. 240 S., 22 Abb., Kt € 29,95 / CHF 39.90 ISBN 978-3-456-85760-2 Auch als eBook erhältlich

17.09.17 18:02

www.hogrefe.com

Die Integrative Leib- und Bewegungstherapie (IBT) verbindet körpertherapeutische und psychotherapeutische Methoden und wird in vielen psychosomatisch-psychotherapeutischen, psychiatrischen Kliniken, Suchtkliniken, ambulanten Praxen und Beratungsstellen eingesetzt. Erlebniszentrierte sowie konfliktaufdeckende Wahrnehmungs- und Bewegungsübungen fördern den Zugang zum Unbewussten und helfen bei heilsamen Neuorientierungsprozessen.

Die 3. Auflage wurde komplett überarbeitet und um aktuelle Themen ergänzt: • Umgang mit aggressiven Impulsen und schwierigen Gefühlen • Natur- und Landschaftstherapie mit schwer traumatisierten Patienten • Bewegungstherapie auf der Basis der Kampfkunst (Budotherapie) • Ressourcenorientierte Arbeit mit „Jungen Erwachsenen“ Das Werk fokussiert insbesondere auf die Behandlung von Erwachsenen mit psychosomatischen Erkrankungen und erläutert in verständlicher Sprache die Grundlagen der Methode. Der ausführliche Praxisteil ermöglicht durch zahlreiche Beispiele auch die Übertragung auf andere Arbeitsbereiche.


A fully updated new edition of the popular text that separates the facts from the myths about drug and substance use 2nd n editio

Mitch Earleywine

Substance Use Problems (Series: Advances in Psychotherapy – Evidence-Based Practice - Volume 15) 2nd edition 2016, viii + 104 pages US $29.80 / € 24.95 ISBN 978-0-88937-416-4 Also available as eBook The literature on diagnosis and treatment of drug and substance abuse is filled with successful, empirically based approaches, but also with controversy and hearsay. Health professionals in a range of settings are bound to meet clients with troubles related to drugs – and this text helps them separate the myths from the facts. It provides trainees and professionals with a handy, concise guide for helping problem drug users build enjoyable, multifaceted lives using approaches based on decades

www.hogrefe.com

of research. Readers will improve their intuitions and clinical skills by adding an overarching understanding of drug use and the development of problems that translates into appropriate techniques for encouraging clients to change behavior themselves. This highly readable text explains not only what to do, but when and how to do it. Seasoned experts and those new to the field will welcome the chance to review the latest developments in guiding self-change for this intriguing, prevalent set of problems.


Instructions to Authors The European Journal of Health Psychology strives to promote theory and practice in the analysis of psychological approaches to health and disease. Its aim is, therefore, to publish high quality empirical or experimental research as well as sound practiceoriented articles, current methodological developments, and comprehensive critical reviews of the scientific literature. European Journal of Health Psychology publishes the following types of article: Original Articles, Review Articles, Short Reports, Commentaries, and Book Reviews.

Manuscript submission: All manuscripts should in the first instance be submitted electronically at http://www.editorialmanager. com/zgp. Detailed instructions to authors are provided at http:// www.hogrefe.com/periodicals/zgp/advice-for-authors/ Copyright Agreement: By submitting an article, the author confirms and guarantees on behalf of him-/herself and any coauthors that the manuscript has not been submitted or published elsewhere, and that he or she holds all copyright in and titles to the submitted contribution, including any figures, photographs, line drawings, plans, maps, sketches, and tables, and that the article and its contents do not infringe in any way on the rights of third parties. The author indemnifies and holds harmless the publisher from any third party claims. The author agrees, upon acceptance of the article for publication, to transfer to the publisher the exclusive right to reproduce and distribute the article and its contents, both physically and in nonphysical, electronic, or other form, in the journal to which it has been submitted and in other independent publications, with no limitations on the number of copies or on the form or the extent of distribution. These rights are transferred for the duration of copyright as defined by international law. Furthermore, the author

European Journal of Health Psychology (2018), 25(1)

transfers to the publisher the following exclusive rights to the article and its contents: 1. The rights to produce advance copies, reprints, or offprints of the article, in full or in part, to undertake or allow translations into other languages, to distribute other forms or modified versions of the article, and to produce and distribute summaries or abstracts. 2. The rights to microfilm and microfiche editions or similar, to the use of the article and its contents in videotext, teletext, and similar systems, to recordings or reproduction using other media, digital or analog, including electronic, magnetic, and optical media, and in multimedia form, as well as for public broadcasting in radio, television, or other forms of broadcast. 3. The rights to store the article and its content in machine readable or electronic form on all media (such as computer disks, compact disks, magnetic tape), to store the article and its contents in online databases belonging to the publisher or third parties for viewing or downloading by third parties, and to present or reproduce the article or its contents on visual display screens, monitors, and similar devices, either directly or via data transmission. 4. The rights to reproduce and distribute the article and its contents by all other means, including photomechanical and similar processes (such as photocopying or facsimile), and as part of so-called document delivery services. 5. The right to transfer any or all rights mentioned in this agreement, as well as rights retained by the relevant copyright clearing centers, including royalty rights to third parties. Online Rights for Journal Articles: Guidelines on authors’ rights to archive electronic versions of their manuscripts online are given in the ’’Guidelines on sharing and use of articles in Hogrefe journals’’ on the journal’s web page at http://www.hogrefe.com/j/zgp. December 2017

Ó 2018 Hogrefe Publishing


Effective, evidence-based approaches to treating alcohol misuse in young adults “Given the significant public health burden of alcohol misuse in our society, this book is a must-have on the shelves of both college and public libraries.” Cecile A. Marczinski, writing in PsycCRITIQUES (2015), 60

Rachel P. Winograd / Kenneth J. Sher

Binge Drinking and Alcohol Misuse Among College Students and Young Adults (Series: Advances in Psychotherapy – Evidence-Based Practice – Vol. 32) 2015, vi + 92 pp. US $29.80 / € 24.95 ISBN 978-0-88937-403-4 Also available as eBook Heavy drinking – and its associated problems – are an integral part of many college students’ and other young adults’ lives. Though some young drinkers are able to consume alcohol without incident, many face significant negative fallout from their excessive consumption. This volume in the series Advances in Psychotherapy: Evidence-Based Practice describes the nature of alcohol misuse, its epidemiology, its causes, and methods for treatment, specifically as they

www.hogrefe.com

pertain to college students and other young adults. It provides practitioners and trainees with a range of evidencebased treatment approaches to help clients change their alcohol use habits. The information presented is both thorough and concise and will help readers with varied backgrounds and experience improve their understanding of the many nuanced factors involved in assessing and treating problematic drinking in young adults.


Behavioral sciences for the next generation of health care providers

Free al tion Addi rces for Resouachers Te

“This latest edition continues the editors’ outstanding success in bridging the exciting worlds of the physiological and behavioral-social contributions to quality health care.” Patrick DeLeon, PhD, MPH, JD, former President of the American Psychological Association

Olle Jane Z. Sahler / John E. Carr / Julia B. Frank / João V. Nunes (Editors)

The Behavioral Sciences and Health Care 4th edition 2018, xiv + 578 pp. US $69.00 / € 54.95 ISBN 978-0-88937-486-7 Also available as eBook

www.hogrefe.com

A clearly structured, compact text on the behavioral sciences for medical and other health professions, including practical features such as chapter review questions and an annotated practice exam.

pulsive and related disorders, and theories of social relations and interprofessional collaboration as well as a new appendix on psychological testing as well as integrating the new classification of the DSM-5.

This edition of this easy-to-use text presents succinct information about the wide variety of social and psychological sciences that interact with the biological sciences in contributing to health and illness. Based around but expanding on the Integrated Sciences Model, and focusing on the Institute of Medicine’s key themes for medical training, this latest edition includes new chapters on pain, obsessive com-

Each chapter begins with guidance questions and ends with current recommended readings and review questions. A complete 335 questionand-answer multiple choice-type review section not only allows readers to assess how well they have learned the material, but also highlights important points and adds additional specific information to supplement the text.

Zgp 2018 25 issue 1  
Zgp 2018 25 issue 1