بهزیستی ذهنی جوانان با افزایش سن

Abstract This study explores the subjective well-being (SWB) of young people aging out of public care in Israel, identifying the individual, social support and institutional characteristics of young people on the verge of leaving care that predict their SWB one year later. The results were obtained from two waves of a longitudinal study with young people who aged out from residential settings in Israel. Shortly before they left care 272 young people completed a self-administered questionnaire and 234 of them were interviewed a year later. Overall, their SWB was fairly positive. Specifically, positive relationship with the mother contributes to a better SWB, and learning difficulties reduced SWB. Gender and country of origin also showed a significant effect. The findings highlight the importance of strengthening the young people's relationships with their biological parents while in care and emphasize the need to provide additional support for those with learning difficulties. The findings also help identify sub-groups of young people in care with greater need for support both while in care and afterwards.

1. Introduction The transition from childhood to adulthood is one of the more challenging life tasks in modern society (Arnett, 2000). This life task is even more difficult for young people growing up in foster families or in residential care. When the responsibility of the state ends due to age restrictions, many young people aging out of care experience abrupt move from a protective and supportive environment to independent living (Stein, 2005). They frequently lack the support received by most young people growing up in industrialized Western countries who continue to depend economically and emotionally on their parents (Arnett, 2000 and Schoeni and Ross, 2005). Instead, many young people who age out of care are compelled to face this complex life task on their own (Courtney et al., 2011 and Wade, 2008). Indeed, international research has presented a gloomy outlook in the major life domains for these young people. Their educational achievements are low and they are prone to becoming homeless, unemployed, showing anti-social behaviors, and lacking money (c.f. Cashmore and Paxman, 2007, Courtney et al., 2011, Pecora et al., 2003 and Reilly, 2003). There is a growing recognition that measures of nonmaterial subjective well-being (SWB) are also needed to fully assess the condition of a population or of complex social realities (Casas, 2011). Measuring SWB is particularly important when dealing with vulnerable populations like children in care, as these measurements can provide accurate information on the children's feelings and satisfaction (Altshuler & Gleeson, 1999). Although the SWB of young people aging out of care may be diminished by the challenges of leaving care and the transition to adulthood, this outcome has been a rather neglected area of policy, practice and research (Dixon, 2008) and hardly any research has been carried out on factors possibly contributing to better SWB. Furthermore, there are very few longitudinal studies identifying factors operating on young people on the verge of leaving care that predict their outcomes afterwards (Harder, Kongeter, Zeller, Knorth, & Knot-Dickscheit, 2011). The current study attempts to narrow this information gap. Individual, social support and institutional characteristics of young Israelis on the verge of leaving care and their SWB one year afterwards were examined, aiming to identify factors during care that predict SWB outcomes. 1.1. The SWB of young people who age out of care The attempt to better understand quality of life and informed policy decisions benefits from using objective indicators (e.g. income) along with nonmaterial subjective measurements (Casas, 2011 and Cummins, 2000). The importance of subjective indicators of well-being stems from the understanding that objective life conditions do not necessarily match the person's appraisal of these conditions, as individuals react differently to the same circumstances and also differ in what they enjoy and find as important (Cummins, 2000 and Diener et al., 1999). SWB is defined as a person's cognitive and affective evaluation of his/her life as a whole (Diener, Oishi, & Lucas, 2011). SWB comprises two main components, affect and life satisfaction (Diener, Suh, Lucas and Smith, 1999). The affective component is based on two dimensions of emotional experience: positive affect (e.g., happiness and joy) and negative affect (e.g., sadness and distress) (Ben-Zur, 2003). The life satisfaction component refers to satisfaction with specific life domains or to general satisfaction with life (Diener, 2000). Research on SWB is based on the understanding that the absence of negative aspects in a person's life does not necessarily indicate that s/he is doing well (Park, 2004). Furthermore, research on SWB in terms of life satisfaction and positive affect shifts attention away from psychopathology and risks and emphasizes the positive aspects of human development (Casas, 2011). Adults' evaluation of their life satisfaction and positive affect has long been in use for assessing individuals and making cross-national comparisons (Diener et al., 2011) and was recently recognized as an essential dimension for understanding children's and adolescents' well-being (U.N.I.C.E.F. Office of Research, 2013). Furthermore, studies on adolescents and youth, although lesser in scope, found that life satisfaction and positive affect are important indicators of a healthy development (Park, 2004). These components are also related to a broad range of other important outcomes, such as low risk-taking behavior (for review see Proctor, Linley, & Maltby, 2009) and adaptive functioning (for review see Huebner, 2004). However, research concerning young people who age out of care focus mostly on objective indicators such as income, living conditions and employment (Dixon, 2008) with almost no research on their SWB. The few studies that have examined SWB among young people who age out of care report mixed results. Some studies show that SWB decreased over time while others found the reverse. For example, in the large-scale longitudinal Midwest Study of young people who age out of care in the United States, 75.0% of the young people at the age of 21 reported feeling satisfied or very satisfied with their lives as a whole (Courtney et al., 2007). This rate decreased to 62.8% at the age of 26 (Courtney et al., 2011). Findings from a study on 47 young people who aged out of care in Australia showed somewhat different findings as their overall happiness declined a year after leaving care in comparison to their happiness a few months before discharge (Cashmore & Paxman, 1996) but slightly increased four to five years later (Cashmore & Paxman, 2007). 1.2. Factors associated with the SWB of young people who age out of care The most consistent finding in the few investigations of factors contributing to the SWB of young people who age out of care is that a strong social network, before or after leaving, is positively linked to SWB (Cashmore and Paxman, 2007, Dixon, 2008 and Reilly, 2003). Being in trouble while in care (e.g., due to offences or truancy) is related to poor SWB a year after aging out of care (Dixon, 2008). In addition, self-esteem has long been recognized as an important contributor to SWB in the general population (e.g. Diener & Diener, 1995). Consequently, its effect on the SWB of young people aging out of care was also explored. 1.3. The Israeli context Out-of-home placement in Israel differs somewhat from that in other Western countries. Due to historical and social processes, the vast majority of placements are to residential settings and not to foster families (Zeira, 2004). There are two systems of out-of-home placements in Israel: welfare settings supervised by the Ministry of Welfare and Social Services and educational settings overseen by the Ministry of Education. Young people are placed in welfare settings based on professional discretion and/or court decisions due to a range of family and child problems. Of the 9000 children removed annually from home by the welfare system about 80% are placed in residential settings and only about 20% are placed with foster families (National Council for the Child, 2011). According to the intensity of care needed, residential settings comprise remedial, rehabilitation or post-hospitalization institutional settings (Attar-Schwartz, 2009). In this study only young people from remedial residential settings were included. The other type of placement, overseen by the Ministry of Education, is voluntary. This system places children to residential settings only and host about 19,000 children annually (National Council for the Child, 2011). These settings were created to respond to the challenges of massive waves of immigration to Israel, aiming to support and enhance assimilation of mostly adolescent immigrants into Israeli society. Since the 1990s, these settings are receiving adolescent immigrants mostly from the Former Soviet Union (FSU) and Ethiopia (Zeira & Benbenishty, 2011). Today, in addition to immigrants, these facilities host adolescents who come from underprivileged families, mostly from the geographical or social periphery of Israel (Mash, 2001). Some of the facilities are called ‘mixed’ settings because they host adolescent entering from both the educational and the welfare placement systems. 1.4. The research goals Based on the differential influence of diverse ecological environments on the development of a child or a young person (Bronfenbrenner, 1979), factors predicting the SWB of young people aging out of care relate to three ecological levels: individual, social support, and institutional characteristics, each level showing a distinct proximity to the individual. This perspective is consistent with Stein's (2005) suggestion that in identifying the factors that contribute to the outcomes of young people who age out of care, it is not enough to simply focus on factors at the individual level, but rather look for factors in different environments. Such perspective also allows shifting from a narrow point of view of personality-based factors to a broader understanding that environmental factors are also critical to the functioning of adolescents and youth (Fergus & Zimmerman, 2005). In the current research the individual level comprises the young people's personal characteristics and qualities; social support refers to relationships with peers, family and carers; and institutional characteristics involve the type of setting. Because so little is known about factors predicting SWB in young people who age out of care, only exploratory goals can be set. This study aims, firstly, to explore the SWB of young people one year after aging out of care. Secondly, it aims to identify characteristics found at the three ecological levels—individual, social support, and institutional—when the young person is on the verge of leaving care that predict his/her SWB one year later.

Results 3.1. Descriptive results Table 1 presents the descriptive results of the two SWB measures. A year after leaving the care system the young people showed quite positive scores on both SLSS (a mean of 2.8 on a scale of 1–4) and MHI-5 (a mean of 21.96 on a scale of 5–30). Table 1 also presents the independent variables. Starting with the individual characteristics, length of stay in care was almost six years, the young people's self-esteem was fairly high and one half of them (50.0%) reported learning difficulties. Table 1. Youth characteristics and SWB: descriptive statistics and Pearson's correlation with MHI-5 and SLSS. Variables Descriptive statistics Pearson's correlation n M SD MHI-5 SLSS MHI-5a 225 21.96 4.53 – .53⁎⁎⁎ SLSSb 225 2.80 .67 .53⁎⁎⁎ – Overall length in care (in years) 264 5.83 2.90 − .02 − .06 Academic achievements (standard scores) 271 − .004 .71 .11 .11 Self-esteemb 267 3.37 0.51 .20⁎⁎ .21⁎⁎ Positive relationship with motherc 245 3.83 1.02 .35⁎⁎⁎ .33⁎⁎⁎ Positive relationship with fatherc 208 3.46 1.14 .29⁎⁎⁎ .11 Difficulty in the relationship with parentsc 259 2.53 .97 − .06 − .04 Tangible peer supportc 268 4.17 .77 .30⁎⁎⁎ .18⁎⁎ Emotional peer supportc 268 4.18 .82 .25⁎⁎⁎ .21⁎⁎ Expression of affection by peerc 268 4.21 .88 .20⁎⁎ .18⁎⁎ Informational peer supportc 268 4.10 .85 .28⁎⁎⁎ .24⁎⁎⁎ Positive social interaction with peerc 268 4.41 .72 .17⁎ .16⁎ Emotional staff supportc 268 3.57 1.12 − .11 .12 Expression of affection by staff c 268 4.34 .77 .15⁎ .17⁎ Informational staff supportc 268 3.72 1.11 − .07 .11 Note. Scale scores: a5–30; b1–4; c1–5. ⁎ p < 0.05. ⁎⁎ p < 0.01. ⁎⁎⁎ p < 0.001. Table options At the level of the social support characteristics, most reported a better relationship with their mother than with their father, but difficulty in the relationship with both parents was low. Furthermore, the young people reported very good support from peers in all five domains. The support from the residential staff was also perceived as high in all the three domains measured, but somewhat lower than support from peers. At the institutional level, as mentioned above, the sample was divided almost equally among the three types of residential settings. 3.2. Factors on the verge of leaving care which predict SWB a year afterwards Bivariate analyses were performed between the individual, social support, and institutional characteristics and the dependent variables. Table 1 shows that most of the significant correlations were similar for both outcomes. Among the individual characteristics, self-esteem was positively correlated with both SLSS and MHI-5. While there were no significant differences in life satisfaction between young people from different countries of origin (F[3,219] = .08, ns), the MHI-5 was significantly different (F[3,219] = 2.70, p = 0.047, η2 = 0.04). Young people from the FSU scored lower than young people of Ethiopian origin (M = 20.62, SD = 4.78; M = 22.82, SD = 4.17 respectively), with no differences from Israeli and other origin (M = 21.58, SD = 4.99; M = 22.11, SD = 3.81 respectively). In addition, MHI-5 was significantly different between genders (t[223] = 2.53, p = 0.012, η2 = 0.02), as young women scored lower than young men (M = 20.99, SD = 4.88; M = 22.55, SD = 4.21 respectively). No difference was found between young men and young women in life satisfaction (t[223] = 0.34, ns). Significant differences were found in both SLSS and MHI-5 between young people with learning difficulties and those without (t[223] = 2.29, p = 0.023, η2 = 0.02; t[223] = 2.46, p = 0.015, η2 = 0.03 respectively), as young people with learning difficulties scored lower than young people without learning difficulties in SLSS (M = 2.70, SD = .68; M = 2.90, SD = .64 respectively) and MHI-5 (M = 21.26, SD = 4.95; M = 22.72, SD = 3.90 respectively). Many of the social support characteristics were positively related to both dependent variables, except for positive relationship with the father, which correlated positively only with MHI-5. For the institutional characteristics, MHI-5 was significantly different among types of residential care (F[2,222] = 3.99, p = 0.02, η2 = 0.04). Young people from educational settings scored significantly higher (M = 23.00, SD = 3.18) than mixed settings (M = 21.01, SD = 5.14), but with no differences from welfare settings (M = 21.68, SD = 5.03). SLSS showed no differences (F[2,222] = .08, ns). A multivariate hierarchical regression analysis was performed separately for each of the dependent variables. According to the recommendation for working with a small maximum model and the desire for parsimony (Kleinbaum, Kupper, Nizam, & Muller, 2008) only variables which were significantly associated with SWB were included in the analysis, according to the level of characteristics (i.e., individual, social support and institutional). Table 2 presents the regression conducted for MHI-5. Among the individual characteristics, which were added in the first step, gender was most strongly associated with MHI-5, followed by self-esteem and learning difficulties. However, when the social support level variables were introduced in the second step, origin from the FSU was also associated with MHI-5, and self-esteem was no longer associated. Additionally, only a positive relationship with the mother was related to MHI-5. The type of setting, added in the third step as the institutional variable, was not associated with MHI-5. In sum, four factors affecting young people on the verge of leaving care were found to contribute to MHI-5 a year later: origin in the FSU was most strongly associated, followed by a positive relationship with mother, learning difficulties, and gender. Altogether, these variables accounted for 26% of the variance in MHI-5. Table 2. Factors on the verge of leaving care and MHI-5: hierarchical linear regression. Model 1 Model 2 Model 3 Variables B β B β B β Israeli origin − .25 − .02 − .36 − .04 − .24 − .02 Ethiopian origin .73 .08 − .08 − .01 − .86 − .09 FSU origin − 2.06 − .17 − 2.69 − .23⁎ − 3.17 − .27⁎⁎ Gender − 2.56 − .29⁎⁎⁎ − 1.88 − .20⁎ − 1.60 − .17⁎ Learning difficulties − 1.71 − .19⁎ − 1.78 − .20⁎⁎ − 1.56 − .17⁎ Self-esteem 1.82 .21⁎⁎ 1.01 .12 1.20 .14 Positive relationship with mother .99 .22⁎⁎ .91 .20⁎ Positive relationship with father .24 .06 .15 .04 Tangible peer support 1.25 .22 1.30 .23 Emotional peer support 1.03 .19 .93 .17 Expression of affection by peer .09 .02 .00 .00 Informational peer support − .05 − .01 − .02 − .00 Positive social interaction with peer .00 .00 .16 .03 Expression of affection by staff − 1.42 − .26 − 1.33 − .24 Educational settings 1.43 .15 Welfare settings − .37 − .04 F(df) 6.70(6,153)⁎⁎⁎ 4.93(14,145)⁎⁎⁎ 4.58(16,143)⁎⁎⁎ Adjusted R2 .18 .26 .26 Δ R2 – .12⁎⁎ .02 ⁎ p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001. Table options Table 3 shows the results of the regression for SSLS. In the first step the individual variables were added and, again, learning difficulties and self-esteem were associated with SLSS. However, when the social support characteristics were introduced in the second step, self-esteem was no longer associated. Here, too, the only social support characteristic related to SLSS was a positive relationship with the mother. Overall, two factors affecting young people on the verge of leaving care were found to contribute to SLSS a year later: a positive relationship with the mother was most strongly associated, followed by learning difficulties. These variables accounted for 15% of the variance in SLSS. Table 3. Factors on the verge of leaving care and SLSS: hierarchical linear regression. Model 1 Model 2 Variables B β B β Learning difficulties − .24 − .18⁎⁎ − .25 − .19⁎⁎ Self-esteem .26 .20⁎⁎ .15 .11 Positive relationship with mother .21 .31⁎⁎⁎ Tangible peer support − .09 − .11 Emotional peer support − .04 − .05 Expression of affection by peer − .04 − .06 Informational peer support .15 .19 Positive social interaction with peer .04 .04 Expression of affection by staff .09 .11 F(df) 8.33(2,202)⁎⁎⁎ 4.94(9,195)⁎⁎⁎ Adjusted R2 .07 .15 Δ R2 – .11⁎⁎ ⁎⁎ p < .01. ⁎⁎⁎ p < .001.