正在加载图片...
ADOLESCENT SELF-ESTEEM AND ADULT DEPRESSION 333 summary.these findings demons rate the power of positive er for adult depression Furthermore we contributed to the by examining the self-esteen Discussion ve otoms two decades later isi ctive data tim ple in of self-esteem contribut global selt em and in tw self-es the vulnerability effect and ror whom owsef-esteem isa risk inte .For exa nle de ng self- f-si changes during ado ompetence might call for a specific interv d ceived academic Although the presc t study included a large sample ncie hould pron incre ses in self- pils sed on self-reports ol/high school e adolesce e often consic nd 0rh.2013) other ific self-esteem change p meters,indi ng th d econ we found gender diffe with (e.g.. ian)mea ontrol for possible lf-repor This finding with veen is un nd poor healti ly to the mo adu dep d.This finding anters Baume 2005)conc 001 et al stdies have span of and self-esteem durine adol cence exhibited more depressive s 5(6 oms two de des later,in adulth Prio r studi ented a icant link be also evidenced virtually identical findings as to our original analysis.4 In summary, these findings demonstrate the power of positive self-esteem development to reduce risk for depressive symptoms both in adolescence and two decades later in adulthood. Discussion The primary aim of this study was to test the hypothesis that both level and change in self-esteem during adolescence predict depressive symptoms two decades later. Using prospective data over a 23-year time span collected from a large sample in Germany, three important findings emerged. First, we found significant increases in global self-esteem and in two self-esteem domains (i.e., perceived physical appearance and academic competence) across the adolescent years. The question of how, and whether, self-esteem changes during adolescence has been discussed controversially. According to maturation theories (cf. Gecas, 2004; Harter, 1999), gains in physical, social, and cognitive competencies should promote increases in self-esteem. Another reason for an increase of self-esteem in this sample could be that— unlike in the United States—German pupils usually experience the transition to secondary school/high school earlier, at the age of 10 or 11. German pupils therefore might not experience the same academic stress during adolescence as American students do because this transitional period has already been processed. More important, we found significant variances for global and domainspecific self-esteem change parameters, indicating that adolescent self-esteem trajectories vary in shape and directions between individuals. Second, we found gender differences in self-esteem levels, with males reporting more positive self-evaluations for all three constructs. This finding is consistent with current research on the relation between gender and self-esteem (e.g., Kling et al., 1999). For our third research aim, we tested the predictive effect of self-esteem level and change on adult depressive symptoms. In line with longitudinal research showing that low adolescent self-esteem level predicts a number of adult life outcomes, such as limited economic prospects and poor health (e.g., Trzesniewski et al., 2006), we found small to medium-sized effects of self-esteem level on depressive symptoms two decades later. Our findings support the notion that both adolescent global and domain-specific selfesteem level have predictive validity for important mental health outcomes in adulthood. This finding counters Baumeister, Campbell, Krueger, and Vohs’ (2005) conclusion that self-esteem is not useful in predicting life outcomes. However, as pointed out by Baumeister et al., most studies have methodological problems such as small sample sizes or cross-sectional designs. We used both a large sample size and a long time span of 23 years to test these effects. Most importantly, compared with earlier studies that focused on self-esteem level, we further found significant effects of selfesteem change on depressive symptoms at age 35— over and above the effect of self-esteem level. Individuals who decreased in self-esteem during adolescence exhibited more depressive symptoms two decades later, in adulthood. To date, change estimates have been underrepresented in studies of long-term life development. Prior studies have documented a significant link between low self-esteem level and depression (e.g., Orth et al., 2009). This study clearly advanced previous work by further revealing that it is not just level but also change in self-esteem that has predictive power for adult depression. Furthermore, we contributed to the literature by examining the self-esteem domains of perceived physical appearance and academic competence. These domains represent typical developmental tasks to the adolescent years, as individuals have to adapt to physical and hormonal changes and choosing among career possibilities during that age period. We wanted to differentiate between facets of adolescent challenges in order to detect what exactly drives the association between low or decreasing self-esteem and adult depression. This knowledge helps us to better understand which domains of self-esteem contribute to the vulnerability effect and for whom low self-esteem is a risk factor for depression. Furthermore, such knowledge facilitates to design interventions. For example, decreasing self-esteem in perceived academic competence might call for a specific intervention program at school, targeting that particular domain of self-esteem which is about to develop into a risk factor for later depression. Although the present study included a large sample size and a long time span to study the effects of adolescent development on adult life outcomes, several limitations should be noted. First, the measures were all based on self-reports. However, because selfesteem is by definition a subjective construct, it is impossible to measure it by objective criteria, and consequently, well-validated self-report scales are often considered the gold standard for assessing self-esteem (Robins, Trzesniewski, & Schriber, 2008; Sowislo & Orth, 2013). Furthermore, other methods of assessing self-esteem, such as implicit measures, have been criticized due to lack of validity (Buhrmester, Blanton, & Swann, 2011). In contrast, for depressive symptoms, it would be useful to include informantbased (e.g., clinician) measures to control for possible self-report biases (e.g., an unwillingness to acknowledge the symptoms of depression) and to account for the effects of shared method variance on the associations between self-esteem and depression. Note, however, that shared method variance is unlikely to account for the effects of self-esteem on age 35 depression, because some 4 Additionally to the model of global self-esteem on adult depression using only the seven corresponding depression items, we also conducted new models for the two self-esteem domains. Model fit of the self-esteem of physical appearance with only the seven corresponding depression items at age 16 and 35 was well, too, 2 (355) 809.90, p .001, CFI .953, RMSEA .029 (90% CI [.026, .032]), and also our findings remained virtually identical. Again, we found small to medium-sized effects of self-esteem level on depressive symptoms at age 16 (.27, p .001) and at age 35 (.21, p .001). Additionally and equal to our original analysis, self-esteem change revealed significant effects on depressive symptoms at age 16 (.18, p .01) and at age 35 (.13, p .05). In all our models, we controlled for depression at age 16 on depression at age 35, again revealing a small effect from age 16 to 35 ( .14, p .01). The same was true for the adapted model of self-esteem of academic competence. We again only included the seven corresponding depression items at age 16 and 35. Model fit was well, 2 (375) 847.01, p .001, CFI .955, RMSEA .029 (90% CI [.026, .031]), and our results remained virtually identical compared with our original model. We found small to medium-sized effects of self-esteem level on depressive symptoms at age 16 (.43, p .001) and at age 35 (.13, p .01). Additionally and equal to our original analysis, also self-esteem change revealed significant effects on depressive symptoms at age 16 (.30, p .001) and at age 35 (.16, p .01). Again, depression at age 16 has a small effect on depression at age 35 ( .12, p .05). This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. ADOLESCENT SELF-ESTEEM AND ADULT DEPRESSION 333