ADOLESCENT SELF-ESTEEM AND ADULT DEPRESSION 32 In this line of reasonin with (formal ope tional thinkin and c ontextual change n-estcem e le be a erts.&Casp indiv s with quite lo self-e em can reduce thei lacing the overly positive a ight be eive them as a chall night give adol es as worth ble persons and,in tur enge estion remains as to the mec isms by which dec n (or in the bulk of the tial and several studies have e for an i c. dy weight sh teem durng ac enc and it is po that th trajectory fo nd a ral ch to a as part of be the in th might leam to nce of stre ng mals in the with initial high showing d nd s al cope and in self- s thoughts in turn.may 12 to cess of integrating positiv gn rch has doc self-estee of the numh luring very few asto cope with (a large or m n t 1997d6 that individ who decre ed in self and the subiective experienc of the chal ed ger ith u and Buswell (1999 hanges i self-esteer Finally nces such s a 24 racnderrole dat ch has appearance may steem on adult depression over a tim span wo d ne that 算0nyi vad of The Present Study ring the en 12 n deve when controlling for level of self-esteem in adolescence adult depressive symptoms two decades later.This study extend Normative Self-Esteem Development in Adolescence n addition to examining the relation hetwe n self-estee depression,the present study also provides further insights into the ither high or low level of self-esteem,largely ignoring the poten tSrpmduringad ial cor chces of change within these onstruct s the adol ears (Ro .2005:Rod nge f-est em during ado depr ssive ns,Trzesniewski,Tracy.Gosling.&Potter.2002:Zimmerman et at age 35. econd.no resea ch has tested the long-term effects o beginning of adolescence. Rather, we argue that an individual’s process through adolescence might be equally important for later life outcomes. In this line of reasoning, even adolescents with generally high self-esteem may be at risk for depression later in life if they decline to moderate levels in self-esteem, or conversely, even individuals with quite low self-esteem can reduce their risk for negative life outcomes if they increase to moderate levels in self-esteem. We emphasize change because it might be crucial to face difficulties, perceive them as a challenge, and then overcome these difficulties. Success experiences with developmental tasks might give adolescents a genuine trust in themselves as worthy and able persons and, in turn, enable them to approach later challenges in a more positive, constructive, and self-affirmative way. The question remains as to the mechanisms by which decreasing self-esteem predicts depression (or increasing self-esteem prevents depression). One potential pathway may be that decreasing self￾esteem leads to the deteriorating of positive beliefs about oneself, which, in turn, predict depression. For example, a girl may enter adolescence with high self-esteem, but due to pubertal changes in her body weight, she might experience insecurities with regard to her changing body image. Thus, she has to adapt to her feminine body image and accept these natural changes as part of becoming an adult. She might learn to define herself as a person of good qualities, although her body image may not look as typically idealized in the media. Another girl with initial high level of self-esteem might not be able to cope with her natural gain in body weight during adolescence in an adaptive way, and thus experience increases in self-conscious thoughts, which, in turn, may deterio￾rate her initially high self-esteem in childhood. Although previous research has documented that self-esteem is malleable during adolescence, very few studies investigated de￾velopmental change as a predictor of life outcomes. Zimmerman et al. (1997) demonstrated that individuals who decreased in self￾esteem during their adolescent years were more prone to peer pressure, alcohol misuse, and tolerance for deviance during the 4 years of the study. Consistent with the vulnerability model, Kim and Cicchetti (2006) found that initial levels of self-esteem pre￾dicted changes in depression, whereas initial levels of depression did not predict changes in self-esteem. Finally, Bolognini et al. (1996) demonstrated that decreasing self-esteem— global and do￾main specific—from age 12 to 14 was related to higher levels of depression at age 14. However, to date, no research has been conducted on the long-term consequences of adolescent self￾esteem on adult depression over a time span of two decades. In summary, we assume that not only initial level of self-esteem at the beginning of adolescence is relevant for later life outcomes but also how adolescents change during the age period between 12 and 16. Thus, adolescent trajectories of self-esteem development should provide prospective information for adult depression even when controlling for level of self-esteem in adolescence. Normative Self-Esteem Development in Adolescence In addition to examining the relation between self-esteem and depression, the present study also provides further insights into the normative development of self-esteem during adolescence. A large body of research suggests that self-esteem generally decreases across the adolescent years (Robins & Trzesniewski, 2005; Rob￾ins, Trzesniewski, Tracy, Gosling, & Potter, 2002; Zimmerman et al., 1997). This adolescent drop in self-esteem has been discussed as a consequence of maturational changes (puberty), cognitive changes (formal operational thinking), and contextual changes (school transitions) (Trzesniewski, Robins, Roberts, & Caspi, 2003). Adolescents typically begin to see themselves in a more critical and differentiated way, displacing the overly positive and holistic self-views (“I’m a good kid”) they maintained in child￾hood (Harter, 1999). This more differentiated view of the self can lead to a drop in self-esteem because individuals have to integrate undesirable aspects of themselves into their self-concept. This process of integrating positive and negative characteristics should be resolved by the end of adolescence, leading to an increase in self-esteem at the beginning of adulthood (Robins & Trzesniewski, 2005). Although the bulk of the evidence points to a drop in self-esteem during adolescence, there are some inconsistencies in the litera￾ture, and several studies have revealed evidence for an increase (e.g., Demo, 1992; McLeod & Owens, 2004). Moreover, we know little about the developmental trajectory of domain-specific self￾esteem during adolescence, and it is possible that the trajectory for these constructs diverges from that found for global self-esteem. Most important for the present study, the inconsistency in the findings suggests the presence of strong interindividual differences in self-esteem change, with some individuals showing increases, others showing decreases, and some showing no change at all. Thus, we expect to find interindividual differences in both the initial level and the developmental trajectory of self-esteem from age 12 to 16. The process of integrating positive and negative aspects of the self might differ largely between individuals. This difference could be a consequence of the number of challenges an adolescent has to cope with (i.e., a large or small increase in body weight, high or low academic prerequisites at school) and on the magnitude and the subjective experience of these challenges. Previous research revealed gender effects with respect to self￾esteem development. A meta-analytic review by Kling, Hyde, Showers, and Buswell (1999) revealed that males report higher self-esteem, on average, than females, with the largest gender gap observed during adolescence. The gender difference may be ex￾plained by divergent socialization experiences such as a tendency to give boys more autonomy than girls, different gender roles such as self-confidence being more valued in boys, or a stronger cultural emphasis on girls’ physical appearance, together with idealized body images portrayed in the media, which may lead to lower self-esteem in girls (Kling et al., 1999). The Present Study In the present study, we investigated adolescent level and change in self-esteem and examined their predictive effects on adult depressive symptoms two decades later. This study extends previous research on the relationship between self-esteem and depression in several ways. First, almost no research exists on change in self-esteem as a predictor of depression. Previous re￾search has often relied on examining potential consequences of either high or low level of self-esteem, largely ignoring the poten￾tial consequences of change within these constructs. We thus investigated the independent prospective effects of level and change in self-esteem during adolescence on depressive symptoms at age 35. Second, no research has tested the long-term effects of 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 327