General well-being and coping strategies of institutionalized male adolescents

(1)

General well-being and coping strategies of institutionalized male

adolescents

Masterscriptie Forensische Orthopedagogiek Pedagogische en Onderwijskundige Wetenschappen Universiteit van Amsterdam M.E. Kramer Amsterdam, augustus 2013 Begeleiding: Mw. A. R. van Beek MSc & Mw. dr. H. E. Creemers Catamaran, kliniek voor forensische jeugdpsychiatrie en orthopsychiatrie Geestelijke Gezondheidszorg Eindhoven en de Kempen (GGzE) Begeleiding: Mw. dr. I.L. Bongers & Mw. C.S. Barendregt MSc

(2)

Acknowledgments

I would like to express my gratitude to mw. Dr. I.L. Bongers and mw. C.S. Barendregt MSc for their support, guidance and engagement. I want to thank mw. A.R. van Beek Msc and mw. Dr. H.E. Creemers for their insightful comments and suggestions. Lastly, I would like to thank GGzE, centre for child & adolescent psychiatry, and Tilburg University, Scientific Centre for Care and Welfare (Tranzo). This master thesis would not have been possible without their cooperation.

(3)

Table of contents Abstract………...4 Introduction……….5 Method………...10 Results………...14 Discussion……….20 References……….25 Appendix………...29

(4)

Abstract

Objectives: The current study examines the prospective, reciprocal, longitudinal relationship

between general well-being and coping strategies of institutionalized male adolescents.

Methods: Data were used from 172 adolescents, participating in a three-wave, longitudinal

study in the Netherlands. General well-being, active coping and passive coping were assessed at admission (T1), after six months (T2) and after 12 months (T3). Participants were divided into two groups. The first (internal) group was institutionalized during the whole study, and the second (external) group was discharged at the time of T3. General well-being and coping were measured with the Lancashire Quality of Life Profile and the Utrechtse Coping Lijst voor Adolescenten. A multi-group cross-lagged model was used to analyse the data. Results: For the internal group, the use of passive coping predicted a higher general well-being after six months. For the external group, the use of passive coping strategies predicted a lower general well-being after six months. For both groups, a higher general well-being after six months, reduces the use of passive coping strategies after twelve months. No relationship was found between general well-being and active coping, and vice versa. Conclusions: The findings suggest a predictive relationship between general well-being and passive coping and between passive coping and general well-being. No predictive relationship was found between active coping and general well-being and vice versa. A different relationship was found between the adolescents who were discharged and the adolescents who were still admitted. Therefore, throughout the rehabilitation process, general well-being and passive coping are of interest in different ways.

Keywords: general well-being, active coping, passive coping, institutionalized

(5)

Introduction

Incarcerated male adolescents experience, compared to their peers, significantly lower general well-being (Forrest, Tambor, Riley, Ensminger, & Starfield, 2000). Adolescents who experience low well-being, are more likely to demonstrate persistent delinquent and aggressive behavior (Suldo & Huebner, 2004). Furthermore, a higher general sense of well-being is argued to be important to reduce the risk of persistent delinquent behavior and psychiatric relapse of adolescents (Barendregt, Van der Laan, Bongers, & Van Nieuwenhuizen, 2012; Bouman, 2009), and appears to function as a buffer against persistent delinquent behaviour and psychiatric relapse (Suldo & Huebner, 2004). This can be explained by Lazarus’s (1991) theory of coping. Adolescents with a high well-being, perceive stressful events in more positive ways, which increases the use of positive emotional reactions and effective coping strategies. In the light of Lazarus’s (1991) theory, a higher well-being contributes to effective coping, which decreases delinquent, and externalizing behaviour (Suldo & Huebner, 2004). Forensic adolescents prefer to use a passive coping strategy over other coping strategies, while this coping style is not efficient in protecting themselves against stress (Dumont & Provost, 1998; Eftekhari, Turner, & Larimer, 2004; Ireland, Boustead, & Ireland, 2005). Using this coping style contributes to a lower well-being (Ebata & Moos, 1991; Gullone, Jones, & Cummins, 2000), as experienced by incarcerated adolescents. It seems that both general well-being and active coping, can serve as a protective factor against persistent delinquent behavior and psychiatric relapse. The direction of this relationship is not yet clear. Insight in this relationship can lead to better understanding of how recidivism can be prevented, by focusing on general sense of well-being and coping strategies. Therefore, the purpose of this study is to examine the reciprocal relationship between general well-being and coping strategies of institutionalized adolescents.

General well-being refers to the individual’s cognitive and affective reaction to his or her entire life-situation, as well as to specific life domains (Diener & Suh, 1994). The importance of general well-being in forensic psychiatry can be explained by means of the Good Lives Model (Ward & Gannon, 2006). The Good Lives Model is a rehabilitation theory for offenders. It provides a framework to assist individuals to achieve their needs by enhancing the skills and capacities that are necessary to achieve these needs (Barendregt, et al., 2012; Ward & Gannon, 2006). According to the Good Lives Model, a high level of general well-being, attained by fulfilling primary needs, reduces the chance of committing a crime in the future (Ward & Gannon, 2006). When focusing on the general sense of well-being of institutionalized male adolescents, it can be stated that most research has been done

(6)

on institutionalized male adults. Incarcerated males experience lower general well-being and self-esteem, compared to the levels of general well-being in the general population, and more feelings of anxiety and depression. This results in a severely compromised general well-being (Gullone, et al., 2000). Very little research focuses on general well-being of institutionalized adolescents, but research findings for institutionalized adolescents are overall comparable to those of institutionalized or imprisoned adults. Incarcerated male adolescents experience, compared to their peers, significantly lower general well-being (Forrest, et al., 2000). They experience a higher level of discomfort, in particular more physical discomfort, they are more symptomatic, have a higher level of emotional discomfort, and have more limitations in activity (Forrest, et al., 2000).

After institutionalization and treatment in forensic care, the well-being of adolescents is increased (Listwan, Colvin, Hanley, & Flannery, 2010; Russell, 2005). Not only the level of general well-being is higher after one month after treatment, but this effect is still present after 24 months, which implies it is not post-treatment-euphoria. Especially when adolescents perceived social support in their environment after discharge, higher general well-being is maintained (Listwan, et al., 2010).

General well-being is also determined by how adolescents cope with problems (Park, 2004). Coping refers to the way people deal with developmental tasks and everyday problems (Bijstra, Jackson, & Bosma, 1994; Patterson & McCubbin, 1989). According to Lazarus (Stevens, 1989) coping behaviors are a result of how an individual approaches a stressful situation in relation to threat, challenge, or benefit to general well-being. During adolescence, coping is important because this is a period of life which is characterized by many stressors, such as physical changes, school transitions, and evolving relationships with friends and parents (Shulman & Cauffman, 2011). Adolescents frequently face new problems which they have to solve independently, for which they previously received the support of their parents. Solving the problems and tasks they face, requires them to develop adequate coping strategies (Bijstra, et al., 1994). Most adolescents are able to successfully cope with these challenges and problems, thanks to advancement in coping skills (Shulman & Cauffman, 2011).

Coping is conceptualized as active versus passive coping strategies (Ebata & Moos, 1991; Stevens, 1989). Active, or approach strategies, include attempts to change the ways of thinking about the problem, and to resolve events by dealing directly with the problem and its consequences. Another way to cope actively with everyday situations is by seeking support (Bijstra & Jackson, 1998; Bijstra, et al., 1994; Stevens, 1989). Passive, or emotion-focused strategies include cognitive attempts to deny threat and attempts to get away from, or avoid

(7)

confrontation, (Ebata & Moos, 1991), depressive reactions (Bijstra & Jackson, 1998; Bijstra, et al., 1994), hoping for a miracle, accepting one’s destiny, acting as if nothing really happened, and trying to forget (Dumont & Provost, 1998). These actions do not change the objective circumstances of the situation, but make the individual feel better (Stevens, 1989).

Previous research shows that active coping strategies are more effective in dealing with problems (Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001). Also, an active coping style may help to reduce violent behavior (Shulman & Cauffman, 2011) and problem behavior (Recklitis & Noam, 1999) while incarcerated, and is negatively related to stress and distress (Dumont & Provost, 1998). On the other hand, passive coping strategies are positively related to stress and distress (Dumont & Provost, 1998). As mentioned, this type of coping is used to lower psychological discomfort, but is not opportune in resolving the problem. It seems evident that an adolescent who uses this type of coping will not be efficient in protecting himself against stress (Dumont & Provost, 1998). In institutionalized adolescents, passive coping strategies are more prevalent than other coping strategies (Eftekhari, et al., 2004; Ireland, et al., 2005). In psychiatrically hospitalized adolescents, passive coping strategies are related to an increase in problem behavior and lower ego development. Particularly negative and avoidant coping strategies are associated with problem behavior (Recklitis & Noam, 1999).

During treatment in forensic care, coping strategies can develop (Prinz, Blechman, & Dumas, 1994; Rohde, Jorgensen, Seeley, & Mace, 2004). During their stay in an institution, different coping strategies are used. Time spent in forensic care is a contributing factor in the use of coping strategies (Mohino, Kirchner, & Forns, 2004). In the first months after admission, adolescents make greater use of passive coping strategies, and less use of active coping styles. The use of passive coping strategies reduces during their stay. After treatment, the coping strategies of the adolescents were improved (i.e. more use of active, problem-solving coping, less use of passive coping) (Mohino, et al., 2004), which implies that the coping strategies that are used by adolescents change over time, during en after their stay in forensic care.

Research to examine the relationship between general well-being and coping strategies has mostly been cross-sectional. According to previous research, individuals who use active coping strategies report a higher perceived general well-being, compared to those individuals who use passive coping strategies (Diener, Eunkook, Lucas, & Smith, 1999). When examining the coping strategies in adolescents, it appears that more use of active problem-solving, an active coping strategy, and less use of acceptance, a passive coping strategy, was

(8)

related to higher levels of general well-being and lower levels of distress (Ebata & Moos, 1991). Additionally, people who are positive in life, and experience high general well-being, (Marriage & Cummins, 2004) use a problem-focused coping style more frequently, especially when they experienced controllable stressful situations (Wrosch & Scheier, 2003). Positivity appears to promote active coping strategies and problem-focused coping mechanisms (Ayyashj-Abdo & Alamuddin, 2007).

To investigate the relationship between general well-being and coping strategies, two pathways are examined. First, the relationship between active and passive coping strategies and general being, in which active and passive coping are predictive of general well-being. Institutionalized adolescents represent a group vulnerable for psychological distress (Brown & Ireland, 2006). During institutionalisation, the adolescents are exposed to several stressors. Which coping strategies these adolescents use, has been found to determine how they deal with these stressors, and to what extent they suffer from these stressors. That defines and contributes to the level of their experienced general well-being (Brown & Ireland, 2006). This way, the coping strategy used contributes to the general well-being of the adolescent. Several prospective studies support this theory. The use of an active coping strategy leads to a higher general well-being, while the use of passive coping strategies leads to a significantly lower general well-being (Herman-Stahl, Stemmler, & Petersen, 1995; Seiffge-Krenke & Klessinger, 2000). Adolescents who changed over time from an active coping strategy to an passive coping strategy, evidenced a significant decrease of general well-being. In institutionalized adolescents, a similar result occurs. In institutionalized adolescents, the use of a passive coping strategy predicts lower general well-being (Brown & Ireland, 2006).

Second, an existing theory describing the relationship between general well-being and coping strategies, is Lazarus’ theory of coping (Park, 2004). According to Lazarus (1991), people with a high general well-being are more likely to appraise stressful life events in more positive ways. This leads to the use of more effective coping behaviors (Lazarus, 1991; Park, 2004). This can only be supported by the cross-sectional study of Mavroveli, Petrides, Rieffe & Bakker (2007). General well-being seems to contribute to the strategies individuals use to deal with everyday problems. Adolescents with a higher general well-being have an advantage in skills related to active coping, because positive emotions are conducive to the development of the physical, intellectual and social resources necessary for successful coping. Therefore, general well-being is particularly relevant to develop adequate coping skills (Mavroveli, et al., 2007). There is no research available that confirms that this also applies to institutionalized adolescents.

(9)

The aim of the present study is to examine the longitudinal relationship between general well-being and coping strategies among institutionalized male adolescents. The primary purpose of this study is to examine the direction of the relationship between general well-being and active and passive coping. Little research has been done to clarify the direction of this relationship, and results are inconclusive. The second aim of this study is to investigate whether this relationship is different when adolescents are discharged. Based on previous research, it seems evident that there is a difference in general well-being for institutionalized adolescents, and the general population (Boxer, Middlemass, & Delorenzo, 2009; Forrest, et al., 2000), and that general well-being is increased during and after their in an institution (Russell, 2005). Also, institutionalized adolescents use more passive coping strategies than their counterparts in the general population (Eftekhari, et al., 2004; Ireland, et al., 2005). Coping strategies develop during treatment, and were improved after discharge (Mohino, et al., 2004; Rohde, et al., 2004). Therefore, it is worthwhile to examine two groups of adolescents: a group who are institutionalized during the whole study, and a group of adolescents who are released after 12 months.

Based on extant findings, it is hypothesized that passive coping predicts a lower general well-being (Brown & Ireland, 2006; Byrne, 2000; Diener, et al., 1999; Ebata & Moos, 1991; Herman-Stahl, et al., 1995; Seiffge-Krenke & Klessinger, 2000). It is expected that active coping will improve general well-being (Diener, et al., 1999; Ebata & Moos, 1991; Herman-Stahl, et al., 1995; Seiffge-Krenke & Klessinger, 2000).

Additionally, no predictive relationship is expected between general well-being and active and passive coping. Little empirical evidence for this relationship exists. Firstly, there is no prospective research to substantiate this hypothesis. Second, the existing theory focuses mainly the relationship between general well-being and active coping. Because institutionalized adolescents are different the level of general well-being, and in the use of coping styles (Eftekhari, et al., 2004; Ireland, et al., 2005), i.e. they use more passive coping styles and less active coping styles than the general population, a different outcome is expected.

Furthermore, a difference between the internal and the external group is expected. Institutionalisation is associated with a lower general well-being in adolescents (Boxer, et al., 2009). Peers in the general population experience a higher general well-being than institutionalized adolescents (Forrest, et al., 2000). After discharge, the level of well-being is increased(Russell, 2005). This effect is still present after 24 months (Listwan, et al., 2010). Also, coping styles that are used by institutionalized adolescents are different from youths

(10)

who are not institutionalized (Eftekhari, et al., 2004; Ireland, et al., 2005), and coping styles can change, or improve during treatment (Prinz, et al., 1994). The coping styles that are used after admission, differ from coping strategies that are used after being admitted for a longer period of time and after discharge. In the months after admission, more passive coping strategies are used, while after discharge, more active coping styles are used (Mohino, et al., 2004; Rohde, et al., 2004). It is expected that the participants who were discharged after 12 months experience a different general well-being and use different coping styles than the adolescents who were still institutionalized after twelve months, therefore this relationship is expected to be different for both groups. It is expected that the external group experiences a higher level of general well-being, due to the use of more effective coping strategies, than the internal group.

Method Participants

Data were collected from a three-wave, longitudinal study in the Netherlands. Participants were male adolescents, admitted to a youth detention centre or a youth forensic psychiatric hospital in the Netherlands. In this study, a total of 228 institutionalized adolescents were approached to participate, of which 56 declined to participate, their parents did not sign informed consent, or were not admitted anymore. At T1, 172 institutionalized male adolescents were included (mean age T1: 17.2 years, SD = .9, range = 15-19). At T2, 119 adolescents participated (mean age T2: 17.7 years, SD = .9 range = 15-20). Fifty-three participants refused to participate, or were transferred or released. At T3, 116 adolescents participated in the study. Several participants were discharged within 12 months, therefore they joined in T1 and T3 (mean age T3: 18.0 years, SD = 1.0, range 16-20). Of the 172 participants at T1, 98 were detained under juvenile civil law, and 74 under juvenile criminal law. All participants have completed primary school in the Netherlands. The participants were divided into two groups: the first group consisted of participants who were admitted during the whole study (N=76), the participants in the second group were institutionalized at T1 and T2, but were discharged at the time of T3 (N=44). Table 1 represents the baseline characteristics of the participants of both groups (internal and external). Mean age was 17.35 (SD = 0.95) years for the internal group and 17.2 (SD= 0.75) years for the external group. In the internal group, 52% was placed under criminal law, while 48% was placed under civil law. In the external group, 25% was placed under criminal law, and 75% under civil law. A disruptive behavior disorder was the most common DSM-IV diagnosis in both groups.

(11)

Analyses revealed two significant differences between the internal group and the external group at T1. First, the mean age of the participants in the internal group was significantly higher than the mean age in the external group at T1 (t(170)=2.23, p= .027). Second, in the external group, significantly more participants were placed under juvenile civil law, and in the internal group, significantly more participants were placed under juvenile criminal law (χ ² (3) = 11.12, p <.05).

Measures

General well-being: The Dutch Youth version of the Lancashire Quality of Life Profile (LQoLP) was used to measure the general sense of well-being of the participants (Van Nieuwenhuizen, Schene, Boevink, & Wolf, 1998; Van Nieuwenhuizen, Schene, & Koeter, 2002; Van Nieuwenhuizen, Schene, Koeter, & Huxley, 2001). To determine the general sense of well-being of the participants, two items were used belonging to the Life Satisfaction Scale (LSS), which together produce an average life satisfaction score (mean-score). This life satisfaction score is determined by one question asked two times, at the beginning and the end of the interview, to assess the participants’ general well-being at the moment (‘How would you rate your life at this moment?’). The participants can rate their satisfaction on a seven-point scale (from `can´t be worse´ to ´can´t be better´) (Van Nieuwenhuizen, et al., 1998; Van Nieuwenhuizen, et al., 2001).

Coping: to measure coping, the Utrecht Coping List for Adolescents (UCL-A) was used (Bijstra, et al., 1994; Schreurs, Van De Willige, Broschot, Tellegen, & Graus, 1993). The UCL-A is a questionnaire that can be used to measure and classify the different types of coping skills. The UCL-A consists of 47 items, which are classified in 7 different scales: confrontation (7 items; e.g. thinking of several solutions to a problem), distraction (8 items; e.g. trying to relax), avoidance (8 items; e.g. trying to duck out of a situation), social support seeking (6 items; e.g. asking somebody else for help), depressive reactions (7 items; e.g. feeling incapable of doing something), expressing emotions (3 items; e.g. working off your strain), and comforting thoughts (5 items; e.g. thinking that after a storm comes a calm.). All of these items are scored on a 4-point Likert scale, from ‘1=seldom or never’, ‘2=sometimes’, ‘3= often’ and ‘4 = very often’, with higher scores indicating a more frequent use of a given coping strategy (Schreurs, et al., 1993). The UCL-A was used to assess two broad domains of coping: active (mean-score of the 15 items of the scales confrontation and seeking social support: α=.82) and passive (mean-score of the 13 items of the scales avoidant and depressive reactions: α=.73) (Bijstra & Jackson, 1998; Meijer, Sinnema, Bijstra, Mellenbergh, &

(12)

Wolters, 2002). The UCL-A has comparable psychometric characteristics as the original UCL (Bijstra et al., 1994). The UCL has sufficient to good psychometric qualities (Bijstra, et al., 1994).

Procedure

General well-being and coping were assessed at admission (T1), and after that every six months. The participants were assessed three times in total (T2, T3). After explanation of the study, written informed consent was acquired from all participants, and from their parents if participants were under 18.

Table 1

Sample Characteristics; age, Dutch juvenile law and DSM diagnosis for both groups

Variable Group internal (N=115) Group external (N=57)

M (SD) M (SD) Age at T1 17.4 (.95) 17.2 (.75) Duration institutionalization until T1 (months) 7.8 (7.1) Range = 0-33 8.6 (8.3) Range = 0-46 DSM-IV ASD ADD DBD RAD 26 (22.6%) 28 (24.3%) 70 (60.9%) 9 (7.8%) 15 (26.3%) 10 (17.5%) 31 (54.4%) 4 (7.5%)

Dutch juvenile law Civil Criminal 55 (48%) 60 (52%) 43 (75%) 14 (25%)

Note. ASD = autism spectrum disorder; ADHD = attention deficit disorder; DBD = disruptive

behavior disorder; RAD = reactive attachment disorder; Because of comorbidity, DSM-IV N

(13)

Data analysis

To analyze the data, first descriptive analyses were performed to assess group differences in the means of general well-being, active coping and passive coping. Second, bivariate correlation analyses were performed to describe the strength and direction of the relationship between coping and general well-being. Third, multi-group cross-lagged regression analyses were performed to analyze the relation between general well-being and active and passive coping (Figure 1).

Figure 1

Cross-lagged model of general well-being and active/ passive coping for both groups

Note. T1= Time 1; T2 = Time 2; T3= Time 3; GWB= General well-being; AC= Active

coping; PC= Passive coping

A cross-lagged model is used to test whether variable X at time 1 is a significant predictor of variable Y at time 2, while controlling for the influence of variable Y itself (Bateman & Strasser, 1983). A multi-group approach was used to study if the structural paths between general well-being and self-esteem differ across subpopulations of institutionalized adolescents. The cross-lagged model gives an estimate of: the correlations between coping and general well-being at all measurements (T1, T2 and T3), the stability coefficients and the cross-lagged regression coefficients (β). To evaluate the model fit of the cross-lagged model, the Chi-square test, TLI (Tucker Lewis Index), CFI (Comparative Fit Index) and RMSEA

GWB

AC/PC

GWB

6 months

(14)

(Root Mean Square Error of Approximation) were used. The following values indicate a good model fit: χ ²> .05, TLI> .95, CFI> .90 and RMSEA <.05 (Bollen & Long, 1993). The current cross-lagged model is presented in figure 1, and was used to test the relations between passive coping and general well-being, active coping and general well-being and vice versa.

Several models are tested. In each model, the stability coefficients between T1, T2 and T3, and the correlations between general well-being and active/ passive coping at T1, T2 and T3 are tested. Model 1 (M1) is the model in which all cross-lagged paths are included. In model 2 (M2), only the cross-lagged paths from general well-being T1 to active/passive coping T2, and from general well-being T2 to active/passive coping T3 are included. Model 3 (M3) is the opposite of Model 2. In Model 3, only the cross-lagged paths from active/passive coping to general well-being are included. Finally, Model 4 (M4) includes no cross-lagged paths and represents the stability model. All four model were tested for general well-being and active coping, and general well-being and passive coping.

Model 1, 2, 3, and 4 are compared by means of a Chi-square difference test. The difference in the Chi-square value, along with the difference in the degrees of freedom between the models was tested against the critical values of the chi-square distribution. This is accomplished to determine whether the addition or removal of paths in the model yielded a significant improvement or deterioration. The Chi-square difference test gives information which model was the best fit to the data. To study whether the relationship is different for the internal and the external group, a multi-group analysis was performed for each model. This was achieved by equating the estimates for both groups, the constrained models, and by releasing this equation, the unconstrained models. Comparing the best fitting constrained and unconstrained model with a Chi-square difference test, gives information about which model best describes the reality. Data analyses were performed using SPSS 19.0 (Statistical Package for the Social Sciences) and MPlus version 5.2 (Muthen & Muthen, 1998-2013).

Results

Table 2 presents the means and standard deviations for general well-being and active and passive coping for the internal and external group. There were no significant differences for the means of both groups (all p’s > .05).

(15)

Table 2

Means and standard deviations of general well-being, active coping and passive coping for both groups

Variable T1 T2 T3 Group internal M(SD) M(SD) M(SD) General well-being 4.6(1.3) 4.9(1.3) 5.0(1.2) Active coping 2.2(.5) 2.3(.5) 2.2(.5) Passive coping 1.9(.4) 1.9(.4) 1.9(.4) Group external General well-being 4.6(1.1) 5.0(1.2) 5.4(.8) Active coping 2.3(.5) 2.3(.5) 2.4(.6) Passive coping 1.9(.4) 1.9(.4) 1.9(.4)

Note. Response scales ranged from 1 to 7 for the measure of general well-being; For coping response scales ranged from 1 to 4.

Table 3 presents the correlations between the variables. The correlation analysis shows a significant negative correlation between general well-being and passive coping at T1, and T3 for participants in the internal group. For the external group, there is also a significant negative correlation between general well-being and passive coping at T1, at T2, and T3.

Between general well-being and active coping, no significant correlations exist for both groups. Additionally, for both groups, significant positive associations exist between general well-being at T1 and T2, T1 and T3, and T2 and T3. For both groups, significant positive associations exist between active coping at T1 and T2, T1 and T3, and T2 and T3. Furthermore, for both groups, significant positive associations exist between passive coping at T1 and T2, T1 and T3, and T2 and T3.

(16)

Table 3

Correlations between general well-being, and active coping and passive coping for both groups

Variables N

Variables 1 2 3 4 5 6 7 8 9 Internal External 1 GWB T1 - .65** .31** .14 .23 .10 -.44** -.34 -.44** 120 52 2 GWB T2 .52** - .57** -.16 .36 .26 -.65** -.56** -.62** 96 23 3 GWB T3 .48** .64** - .22 .24 .27 -.24 -.25 -.37** 64 52 4 AC T1 .00 -.05 .11 - .33 .45** -.12 -.33 -.06 120 52 5 AC T2 .07 .00 .03 .58** - .65** -.26 -.24 -.21 96 23 6 AC T3 .11 .05 .21 .49** .65** - .26 -.33 -.02 64 52 7 PC T1 -.33** .01 -.02 -.18 .08 -.03 - .77** .68** 120 52 8 PC T2 -.15 -.03 -.05 -.03 .13 -.08 .48** - .65** 96 23 9 PC T3 -.30** .36** -.33 .15 -.06 .11 .23 .61** - 64 52 Note. * = p < .05, ** = p < .01, *** = p < .001; GWB = general well-being; AC= Active Coping; PC = Passive Coping; T1 =

Time 1; T2 = Time 2; T3 = Time 3; Correlations below the diagonal are for the internal group and the top right correlations are for the external group.

Measurement models Active coping

In the implementation of the cross-lagged analysis, several models were tested. Table 4 and 5 present the results of the Chi-square difference test for the constrained (a) and unconstrained (b) models for active coping. Based on the fit indices, the model in which none of the cross-lagged effects are included, the stability model (M4a, M4b) was the model that best described the reality. None of the other models had a significantly better fit. A subsequent Chi-square difference test showed no significant difference between the constrained stability model (M4a) and the unconstrained stability model (M4b) (M4a vs. M4b: Δ χ ² (7) = 6.07, p >.05). This means that there is no significant difference between the internal and external group, and separation of the groups does not provide additional improvement to the model, thus the constrained stability model (M4a) will be further described.

In the stability model for both groups (M4a) (Figure 2, Appendix) significant stability effects were found between general well-being at T1 and T2 (β = .54, p < .001) and between general well-being at T2 and T3 (β = .60, p < .001). This also applies for active coping. Significant stability effects were found between active coping T1 and T2 (β = .56, p < .001),

(17)

and T2 and T3 (β = .66, p < .001). Also, a significant correlation was found between general well-being T3 and active coping at T3 (β = .25, p < .01).

Table 4

Goodness of fit indices and chi-square difference tests of general well-being and active coping for the constrained models (a)

Coefficients

Models χ² Df TLI CFI RMSEA Models Δ χ² Δ df

M1 22.24 19 .97 .98 .05 M2 22.69 21 .99 .99 .03 M1 vs. M2 0.45 2 M3 23.68 21 .98 .96 .04 M1 vs. M3 1.44 2 M2 vs M3 .99 0 M4 24.15 23 .99 .99 .02 M1 vs. M4 1.91 4 M2 vs. M4 1.46 2 M3 vs. M4 0.47 2

Note. TLI = Tucker-Lewis Index; CFI = comparative fit index; RMSEA = root-mean-square error of

approximation; * = p < .01, ** = p < .005, *** = p < .001; M1 = both cross-lagged paths; M2 = General

(18)

Table 5

Goodness of fit indices and chi-square difference tests of general well-being and active coping for the unconstrained models (b)

Coefficients

M1 13.82 8 .89 .97 .09 M2 16.06 12 .95 .98 .06 M1 vs. M2 2.23 4 M3 15.95 12 .95 .98 .06 M1 vs. M3 2.13 4 M2 vs M3 .11 0 M4 18.09 18 .98 .99 .04 M1 vs. M4 4.27 8 M2 vs. M4 2.03 4 M3 vs. M4 2.14 4

well-being to active coping; M3 = active coping to general well-being; M4 = no cross-lagged paths.

Passive coping

Table 6 and 7 present the results of the Chi-square difference test of the constrained (a) and unconstrained (b) models for passive coping.

The first series of Chi-square difference tests showed that the difference between the stability model for the constrained model (M4a) and the model with cross-lagged effects from general well-being to passive coping (M2a) (M2a vs. M4a: Δ χ ² (2) = .15.26, p <.01), was significant. This means that the model with cross-lagged effects (M2a) better accounted for the data than de model with no cross-lagged effects (M4a). Also, the Chi-square difference tests showed that the difference between the model with cross-lagged effects from passive coping to general well-being (M3a) and the model with all cross-lagged effects (M1a) (M1b vs. M3a: Δ χ ² (2) = 15.25, p <.01), and the difference between the stability model (M4a) and the model with all cross-lagged effects (M1a) (M1a vs. M4a: Δ χ ² (4) = 15.77, p <.01), was also significant. Therefore, the unconstrained model with all cross-lagged effects (M1a) accounted best for the data

The second series of Chi-square difference tests, for the unconstrained models, showed similar results. The model with all cross-lagged effects (M1a) had a significantly better fit than the other models (M2b, M3b, M4b) (M3b vs. M4b: Δ χ ² (4) = 14.23, p <.01;

(19)

M2b vs. M4b: Δ χ ² (4) = 15.47, p <.01; M1b vs. M4b: Δ χ ² (8) = 30.75, p <.01). Therefore, for both groups, the model with all cross-lagged effects, accounted best for the data.

The third Chi-square difference test showed that there was a significant difference between the constrained model with all cross-lagged effects (M1a) and the unconstrained model with all cross-lagged effects (M1b) (M1a vs. M1b: Δ χ ² (11) = 29.79, p <.01). Thus, there is a significant difference between the internal and external group, and separation of the groups provides an additional improvement to the model. The unconstrained model with all cross-lagged effects (M1b) will be further described for both groups.

In figure 3 (Appendix) the estimates for the internal group of participants are depicted. Significant stability coefficients (β) were found between general well-being at T1 and T2 (β = .58, p < .05), general well-being at T2 and T3 (β = .64, p < .05), passive coping at T1 and T2 (β = .63, p < .05), and passive coping at T2 and T3 (β = .48, p < .05). In this model, a significant cross-lagged regression coefficient (β) was found between passive coping at T1, and general well-being at T2 (β = .20, p < .05), and a negative significant cross-lagged regression coefficient was found between general well-being at T2 and passive coping at T3 (β = -.31, p < .05). This indicates that the level of passive coping at T1 is a significant positive predictor of general well-being at T2, and the level of general well-being at T2 a significant negative predictor of the level of passive coping at T3.

For the external group (Figure 4, Appendix), different results were found. All stability coefficients were significant, i.e. between general well-being at T1 and T2 (β = .47, p < .05), general well-being at T2 and T3 (β = .55, p < .05), passive coping at T1 and T2 (β = .67, p < .05), and passive coping at T2 and T3 (β = .61, p < .05). Also, negative correlations were found between general well-being at T1 and passive coping at T1 (β = -.42, p < .05). Significant negative cross-lagged regression coefficients (β) were found between passive coping at T1 and general well-being at T2 (β = -.45, p < .05), and general well-being at T2 and passive coping at T3 (β = -.34, p < .05). Thus, the level of passive coping at T1 is predictive of the level of general well-being at T2, and the level of general well-being at T2 is predictive of the level of passive coping at T3.

(20)

Table 6

Goodness of fit indices and chi-square difference tests of general well-being and passive coping for the constrained models (a)

Coefficients

M1 38.33 19 .87 .91 .11 M2 38.84 21 .89 .92 .10 M1 vs. M2 0.51 2 M3 53.58 21 .80 .85 .13 M1 vs. M3 15.25* 2 M2 vs. M3 14.74 0 M4 54.10 23 .83 .86 .13 M1 vs. M4 15.77* 4 M2 vs. M4 15.26* 2 M3 vs. M4 0.52 2 Table 7

Goodness of fit indices and chi-square difference tests of general well-being and passive coping for the unconstrained models (b)

Coefficients

Models χ² df TLI CFI RMSEA Models Δ χ² Δ df

M1 8.56 8 .99 .99 .03 M2 23.83 12 .87 .95 .11 M1 vs. M2 15.28* 4 M3 25.08 12 .86 .94 .11 M1 vs. M3 16.53* 4 M2 vs M3 1.25 0 M4 39.90 16 .81 .89 .13 M1 vs. M4 30.75* 8 M2 vs. M4 15.47* 4 M3 vs. M4 14.23* 4

(21)

Discussion

The aim of the present study was to gain insight into the prospective, reciprocal relationship between general well-being and coping strategies among institutionalized adolescents. A three-wave longitudinal design was used. The participants were divided into two groups. The first group, the internal group, was institutionalized during the whole study. The second group, the external group, was institutionalized at T1 and T2, but discharged at T3. The first hypothesis stated that, based on previous findings (Brown & Ireland, 2006; Byrne, 2000; Diener, et al., 1999; Ebata & Moos, 1991; Herman-Stahl, et al., 1995; Seiffge-Krenke & Klessinger, 2000), it was expected that the use of passive coping strategies predicted a lower general well-being. This could only be confirmed for the external group in the first six months (T1-T2). In the external group, the use of passive coping strategies predicted a lower general well-being six months later. In the relationship between passive coping and general well-being, a difference was shown between the internal group and external group, which was stated in the last hypothesis. This implies that the rehabilitation phase and the amount of time until release of the participants affects the impact of the use of passive coping on general well-being. In the internal group, the use of passive coping predicted a higher general well-being after six months, and a higher general well-being after six months predicted less use of passive coping strategies after twelve months. For the external group, the use of passive coping strategies predicted a lower general well-being after six months, and a higher general well-being predicted less use of passive coping strategies after twelve months. The difference between the two groups will be further discussed in the last hypothesis.

In line with findings of Diener and colleagues (1999) and Ebata and Moos (1991), the second hypothesis stated that the use of an active coping strategy would improve sense of general well-being. This could not be confirmed in the present study. In contrast to most previous findings, this study showed that the use of active coping did not predict the level of general being. A few studies had similar findings to this study, in which general well-being was not predicted by any of the coping strategies which they examined, including active and passive coping strategies (Masthoff, et al., 2007; Penley, Tomaka, & Wiebe, 2002). To date, research concerning general well-being and coping strategies has not focused on a population of institutionalized adolescents. This may explain why in this study, the expected relationship was not found. In addition, in previous studies, coping strategies and general

(22)

well-being were assessed with different measures, which may influence the results that were found.

In the third hypothesis, no predictive relationship between general well-being and passive and active coping was expected. The findings in this study support this hypothesis partially. Regarding active coping, this hypothesis can be confirmed. No predictive relationship was found between general well-being and active coping. Three explanations could clarify this finding. First, adolescents have more difficulty learning an active coping style when they are institutionalized (Ireland, et al., 2005), because the sources of learning coping strategies are now restricted to other institutionalized adolescents. Second, an environment with a lot of stressors complicates the process of learning active coping strategies. Third, institutionalized adolescents have a limited range of coping strategies available to them. It is assumed that the restrictive environment may reduce the coping styles that are accessible to offenders (Ireland, et al., 2005).

Regarding passive coping, this hypothesis could not be confirmed. For the internal group and external group, a higher level of general well-being after six months predicted less use of passive coping strategies after twelve months. This finding can not be supported by previous research, but may be explained by means of Lazarus’ theory (1991) of coping. Lazarus (1991) stated that a higher general well-being contributes to effective coping. In addition to Lazarus’ theory, another theory also supports the pathway of passive coping (Folkman, 1997). This theory states that a higher general well-being may contribute to more use of active coping strategies, and less use of passive coping strategies, as it appeared in this study.

Furthermore, an unexpected result was found regarding the relationship between passive coping and general well-being. For the internal group, passive coping strategies contributed to a higher well-being in the first six months. After six months, a high general well-being reduced the use of passive coping after twelve months. For the external group, passive coping contributed to a lower general well-being in the first six months, and a high general well-being reduced the use of passive coping strategies after twelve months. Previous research shows that the coping strategies used by adolescents, change during admission. Passive coping strategies are more frequently used the first six months after admission, and seem to be effective in the early stage of admission (Mohino, et al., 2004). Several explanations could be mentioned. Adolescents, when in an institution, are actively encouraged to use an coping style that would be less effective outside an institution (Ireland, et al., 2005). For example, an adolescent walks out of a situation to prevent escalation, or an adolescent

(23)

provokes the situation by demonstrating self-destructive behavior, ensuring that he is removed from the unit and away from the situation (Ireland, et al., 2005). The treatment that adolescents receive during their stay in a forensic setting may also contribute to this relationship. A study of social skills training shows that adolescents who receive social skills training, use more passive coping strategies after the training, while the group adolescents who received no social skills training during their admission use less passive coping strategies. Both groups showed increase of general well-being (Bijstra & Jackson, 1998). Thus, passive coping strategies may be encouraged by the setting and treatment, and prove to be more effective during admission, which may lead to a higher general well-being. However, previous research shows that these strategies are not effective in solving the problem in the long-term (Dumont & Provost, 1998).

In the last hypothesis, the difference between the two groups was assessed. It was expected that the external group had a higher level of general well-being and more use of active coping strategies than the internal group. In this study, this could not be confirmed. There was no significant difference between the internal and external group in active coping strategies and general well-being. Also, it was expected that the relationship between general well-being and passive and active coping would be different for the internal group and external group. As previously mentioned, there is a different relationship between general well-being and passive coping for the internal group and external group. In the internal group, the use of passive coping predicted a higher general well-being after six months, and a higher general well-being after six months predicted less use of passive coping strategies after twelve months. For the external group, the use of passive coping strategies predicted a lower general well-being after six months, and a higher general well-being predicted less use of passive coping strategies after twelve months.

The difference between the internal group and external group in the relationship between general well-being and passive coping in the first six months may be explained by the difference between the two groups. First, the group who was institutionalized during the whole study was significantly older at admission, and were admitted for a longer period of time than the group who were discharged after twelve months at the time of the first measurement. This is due to the fact that in the internal group, more participants were detained under juvenile criminal law. They were admitted for a longer period of time, as determined in advance, while the admission of the participants who were detained under juvenile civil law was not defined in advance. During admission, with less focus on rehabilitation and discharge, passive coping strategies seem to be more effective to general

(24)

well-being. When the focus of the treatment is on rehabilitation and discharge, passive coping strategies seem to be less effective, and predict a lower level of general well-being. Difference in treatment trajectory, rehabilitation and perspective may cause a difference in the effectiveness of passive coping strategies.

Four limitations should be noted concerning this study. First, findings suggest that most adolescents do not use only one coping strategy. In fact, most adolescents display a more flexible and dynamic approach to coping with stress (Ebata & Moos, 1991; Herman-Stahl, et al., 1995). Also, which coping strategy is used, depends on a number of factors. First, the way an individual assesses a situation. Second, the situation itself, and third, the preference of an individual for a particular coping strategy in a given situation (Lazarus, as cited in Stevens, 1989). Therefore, a distinction between passive and active coping may not be made. An individual has a preference for a particular strategy in a particular situation. Which strategy is preferred, may vary by situation. An individual can use different coping strategies in different situations and is not tied to one strategy. (Herman-Stahl, et al., 1995; Shulman & Cauffman, 2011).

Second, the duration of institutionalisation of the participants at the time of the first measurement is very divergent. This results in participants who were in various stages in their rehabilitation process. Changes in the environment of an adolescent, like the admission at an institution, or the transition from an institution to their homes, can lead to changes in their risk- and protective factors concerning delinquency and recidivism (Van der Laan, Blom, & Bogaerts, 2007).This may lead to differences in results. Third, coping was assessed through a self-report questionnaire. Adolescents were asked to report on the behaviours they usually display when dealing with stressful events. This self-report questionnaire may not reflect an adolescents’ coping efforts with actual problems (Herman-Stahl, et al., 1995; Ireland, et al., 2005). To generalize the findings, data from other informants would be necessary. Lastly, no distinction was made between the type of treatment the youngsters received. Youngsters who were detained under juvenile criminal law and juvenile civil law from different types of institutions participated in the study. As shown in this study, there is a difference between youngsters who were admitted during the whole study, and youngsters who were discharged after twelve months. Both groups also differ in age, and whether they are detained under juvenile criminal law or juvenile civil law. Differences in treatment may be present, due to the type of measure that is received. The differences within the groups may cause differences in perceived general well-being or self-reported coping strategies. In future research, these groups can be separately examined, in order to minimize these differences.

(25)

The relationship between general well-being and passive coping provides insight in the development of institutionalized adolescents. According to the Good Lives Model, a high well-being can reduce the risk of reoffending and relapse (Barendregt, et al., 2012). Insight in the reciprocity between general well-being and active and passive coping might provide a manner to shape treatment programs in order to prevent reoffending and relapse. There seem to be indications that being admitted for a shorter period of time and focusing more on rehabilitation, reduces the use of passive coping during the admission and after discharge and therefore enhances general well-being, and vice versa. When youngsters are incarcerated for a longer period of time, passive coping strategies are effective, because passive coping strategies lead to a higher perceived general well-being. Having perspective seems to be of interest. Adolescents who were detained under juvenile civil law may have more perspective on early discharge and may perhaps have more influence on the course of their treatment, since they are not convicted to a certain duration of treatment.

Therefore, it is important in the rehabilitation process to focus on reducing passive coping strategies to increase general well-being. When perceived general well-being is higher, fewer passive coping strategies are used. Previous research shows that coping strategies can improve during treatment (Prinz, et al., 1994). Because coping skills can be improved among these youngsters, and a relationship was found between general well-being and passive coping strategies, it is important to focus on coping during the rehabilitation process.

Little research had been done in institutionalized male adolescents on general well-being and coping strategies. To prevent recidivism and psychiatric relapse in the future, more research is needed. In this study, general well-being and coping strategies were examined over a period of twelve months. Future research is needed to examine the course of coping strategies over a longer period of time after discharge. Then, the relationship between general well-being and coping strategies and the effects on reoffending and relapse can be examined.

Furthermore, most institutionalized adolescents are diagnosed with (comorbid) psychiatric disorders (Vermeiren, Jespers, & Moffitt, 2006). Therefore, the relationship between psychiatric disorders, general well-being and coping strategies may be of interest. In adults, previous research has shown that adults diagnosed with a psychiatric disorder report a lower general well-being than the general population (Atkinson, Zibin, & Chuang, 1997), and have poorer coping strategies (Aldwin & Revenson, 1987). In institutionalized adolescents, there may be a similar relationship which can provide options for developing treatment programs in the future.

(26)

The main findings of this study suggest that there is a relationship between general well-being and passive coping of institutionalized adolescents and vice versa. It seems that the level of rehabilitation and the time until discharge, has an impact on the effectiveness of the use of passive coping strategies. When institutionalized adolescents have less perspective on discharge, the use of passive coping strategies seem to be effective and contributory to general well-being. Additionally, when institutionalized adolescents are close to discharge, the use of passive coping strategies reduce the level of general well-being. For both groups, a higher level of general well-being decreases the use of passive coping strategies. This finding is of interest because there is little research on general well-being and coping strategies of institutionalized adolescents. This study may gain insight in the contribution of general well-being and coping strategies in treatment programs to enhance general well-well-being and reduce the risk of recidivism and psychiatric relapse.

(27)

References

Aldwin, C., & Revenson, T. (1987). Does coping help?: a reexamination of the relation between coping and mental health. Journal of Personality and Social Psychology,

53(2), 337-348.

Atkinson, M., Zibin, S., & Chuang, H. (1997). Characterizing quality of life among patients with chronic mental illness: a critical examination of the self-report methodology.

American Journal of Psychiatry, 154(1), 99-105.

Ayyashj-Abdo, H., & Alamuddin, R. (2007). Predictors of subjective well-being among college youth in Lebanon. Journal of Social Psychology, 147(3), 265-284.

Barendregt, C. S., Van der Laan, A. M., Bongers, I. L., & Van Nieuwenhuizen, Ch. (2012). Explaining reoffending and psychiatric relapse in youth forensic psychiatry from a good lives model perspective. In T. I. Oei & M. S. Groenhuijsen (Eds.), Progression

in forensic psychiatry: Kluwer.

Bateman, T. S., & Strasser, S. (1983). A cross-lagged regression test of the relationship between job tension and employee satisfaction. Journal of Applied Psychology, 68(3), 439-445.

Bijstra, J. O., & Jackson, S. (1998). Social skills trainig with early adolescents: Effects on social skills, well-being, self-esteem and coping. European Journal of Psychology of

Education, 13(4), 569-583.

Bijstra, J. O., Jackson, S., & Bosma, H. A. (1994). De Utrechtse Coping Lijst voor Adolescenten. Kind en Adolescent, 15(2), 98- 109

Bollen, K. A., & Long, J. S. (1993). Testing structural eqation models. Newbury Park, CA: Sage.

Bouman, Y. H. A. (2009). Quality of life and criminal recidivism in forensic outpatients with

personality disorders: A Good Lives approach. Maastricht: Universiteit Maastricht.

Boxer, P., Middlemass, K., & Delorenzo, T. (2009). Exposure to violent crime during

incarceration: effects on psychological adjustment following release. Criminal Justice

and Behavior, 36(8), 793-807.

Brown, S. L., & Ireland, C. A. (2006). Coping style and distress in newly incarcerated male adolescents. Journal of Adolescent Health, 38, 656-661.

Byrne, B. (2000). Relationships between anxiety, fear, self-esteem, and coping strategies in adolescence. Adolescence, 35(137), 201-215.

(28)

Compas, B. E., Connor-Smith, J. K., Saltzman, H., Thomsen, A. H., & Wadsworth, M. E. (2001). Coping with stress during childhood and adolescence: Problems, progress, and potential in theory and research. Psychological Bulletin, 127, 87-127.

Diener, E., Eunkook, M. S., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being: three decades of progress. Psychological Bulletin, 125(2), 276-302.

Diener, E., & Suh, E. (1994). Measuring Quality of Life: economic, social, and subjective indicators. Social Indicators Research, 40, 189-216.

Dumont, M., & Provost, M. A. (1998). Resilience in Adolescents: Protective role of social support, coping strategies, self-esteem, and social activities on experience of stress and depression. Journal of Youth and Adolescence, 28(3), 343-363.

Ebata, A. T., & Moos, R. H. (1991). Coping and adjustment in distressed and healthy adolescents. Journal of Applied Developmental Psychology, 2, 33-54.

Eftekhari, A., Turner, A. P., & Larimer, M. E. (2004). Anger expression, coping, and substance use in adolescent offenders. Addictive Behaviors, 29, 1001-1008. Folkman, S. (1997). Positive psychological states and coping with severe stress. Social

Science & Medicine, 45(8), 1207-1221.

Forrest, C. B., Tambor, E., Riley, A. W., Ensminger, M. E., & Starfield, B. (2000). The Health Profile of Incarcerated Male Youths. Pediatrics, 105, 286 -291.

Gullone, E., Jones, T., & Cummins, R. A. (2000). Coping styles and prison experience as predictors of psychological well-being in male prisoners. Psychiatry, Psychology and

Law, 7(2), 170-181.

Herman-Stahl, M. A., Stemmler, M., & Petersen, A. C. (1995). Approach and avoidant

coping: implicantions for adolescent mental health. Journal of Youth and Adolescence,

24(6), 649-665.

Ireland, J. L., Boustead, R., & Ireland, C. A. (2005). Coping style and psychological health among adolescent prisoners: a study of young and juvenile offenders. Journal of

Adolescence, 28, 411-423.

Lazarus, R. S. (1991). Emotion and adaption. New York: Oxford University Press.

Listwan, S. J., Colvin, M., Hanley, D., & Flannery, D. (2010). Victimization, Social Support, and Psychological Well-Being: A Study of Recently Released Prisoners. Criminal

Justice and Behavior, 37, 1140-1159.

Marriage, K., & Cummins, R. A. (2004). Subjective Quality of Life and Self-Esteem in Children: The Role of Primary and Secondary Control in Coping with Everyday Stress. Social Indicators Research, 66(1/2), 107-122.

(29)

Masthoff, E. D., Trompenaars, F. J., Van Heck, G. L., Michielsen, H. J., Hodiamont, P. P., & DeVries, J. (2007). Predictors of quality of life: A model based study. Quality of Life

Research, 16, 309-320.

Mavroveli, S., Petrides, K. V., Rieffe, C., & Bakker, F. (2007). Trait emotional intelligence, psychological well-being and peer-rated social competence in adolescence. British

Journal of Developmental Psychology, 25, 263-275.

Meijer, S. A., Sinnema, G., Bijstra, J. O., Mellenbergh, G. J., & Wolters, W. G. W. (2002). Coping styles and locus of control as predictors for psychological judgement of adolescents with a chronic illness. Social Science & Medicine, 54, 1453-1461. Mohino, S., Kirchner, T., & Forns, M. (2004). Coping strategies in Young Male Prisoners.

Journal of Youth and Adolescence, 33(1), 41-49.

Park, N. (2004). The role of subjective well-being in positive youth development. The

ANNALS of the American Academy of Political and Sociale Science, 591, 25-39.

Patterson, J. M., & McCubbin, H. I. (1989). Adolescent coping style and behaviors: conceptualization and measurement. Journal of Adolescence, 10, 163-186.

Penley, J. A., Tomaka, J., & Wiebe, J. S. (2002). The association of coping to physical and health outcomes: a meta-analytic review. Journal of Behavioral Medicine, 25(6), 551-603.

Prinz, R. J., Blechman, E. A., & Dumas, J. E. (1994). An evaluation of peer coping-skills training for childhood aggression. Journal of Clinical Child Psychology, 23, 193-203. Recklitis, C. J., & Noam, G. G. (1999). Clinical and developmental perspectives on

adolescent coping. Child Psychiatry and Human Development, 30(2), 87-101.

Rohde, P., Jorgensen, J., Seeley, J. R., & Mace, D. E. (2004). Pilot Evaluation of the Coping Course: A Cognitive-Behavioral Intervention to Enhance Coping Skills in Incarcerated Youth. Journal of the American Academy of Child and Adolescent Psychiatry, 43(6), 669-676.

Russell, K. C. (2005). Two Years Later: A Qualitative Assessment of Youth Well-Being and the Role of Aftercare in Outdoor Behavioral Healthcare Treatment. Child & Youth

Care Forum, 34(4), 209-239.

Schreurs, P. J. G., Van De Willige, G., Broschot, J. F., Tellegen, B., & Graus, G. M. H. (1993). Utrechtse Coping Lijst (UCL) [Utrecht Coping List]. Amsterdam: Swets & Zeitlinger.

Seiffge-Krenke, I., & Klessinger, N. (2000). Long-term effects of avoidant coping on

(30)

Shulman, E. P., & Cauffman, E. (2011). Coping while incarcerated: a study of male juvenile offenders. Journal of research on Adolescence, 21(4), 818-826.

Stevens, M. (1989). Coping strategies of hospitalized adolescents. Children's Health Care,

18(3), 163-169.

Suldo, S. M., & Huebner, E. S. (2004). Does life satisfaction moderate the effects of stressful life events on psychopathological behavior in adolescence? School Psychology

Quarterly, 19(2), 93-105.

Van der Laan, A. M., Blom, M., & Bogaerts, S. (2007). Zelfgerapporteerde

jeugddelinquentie: Risico's en bescherming Bevindingen uit de WODC monitor Zelfgerapporteerde Jeugdcriminaliteit 2005. Den Haag.

Van Nieuwenhuizen, Ch., Schene, A. H., Boevink, W. A., & Wolf, J. R. L. M. (1998). The Lancashire Quality of Life Profile: First Experiences in the Netherlands. Community

Mental Health Journal, 34(5), 513-524.

Van Nieuwenhuizen, Ch., Schene, A. H., & Koeter, M. W. J. (2002). Quality of life in forensic psychiatry: an unreclaimed territory? International Review of Psychiatry, 14, 198-202.

Van Nieuwenhuizen, Ch., Schene, A. H., Koeter, M. W. J., & Huxley, P. J. (2001). The Lancashire Quality of Life Profile: modification and psychometric evaluation. Social

Psychiatry and Epidemiology, 36, 36-44.

Vermeiren, R., Jespers, I., & Moffitt, T. (2006). Mental health problems in juvenile justice populations. Child and Adolescents Psychiatric Clinics, 15(2), 333-351.

Ward, T., & Gannon, T. A. (2006). Rehabilitation, etiology, and self-regulation: The comprehensive good lives model of treatment for sexual offenders. Aggression and

Violent Behavior, 11, 77-94.

Wrosch, C., & Scheier, M. F. (2003). Personality and quality of life: The importance of optimism and goal adjustment. Quality of Life Research, 12(1), 59-72.

(31)

Appendix

Figure 2

Cross-lagged regression model (M4a) of general well-being and active coping for both groups

Note. * = p < .01; Model fit: χ² (df = 23, N = 119) = 24.15; TLI = .99; CFI = .99; RMSEA = .02; T1 =

Time 1; T2 = Time 2; Time 3 = Time 3.

Figure 3

Cross-lagged regression model (M1b) of general well-being and passive coping for the internal group

GWB

PC

GWB

6 months T1 T2 6 months T3 .58* .64* .48* .63* -.31* .20* -.33*

GWB

AC

GWB

6 months T1 T2 6 months _T3 .54* .60* .56* .66* .25*

(32)

Figure 4

Cross-lagged regression model (M1b) of general well-being and passive coping for the external group