Elsevier

Journal of School Psychology

Volume 42, Issue 3, May–June 2004, Pages 243-261
Journal of School Psychology

Changes in self-esteem during the middle school years: a latent growth curve study of individual and contextual influences

https://doi.org/10.1016/j.jsp.2004.04.001Get rights and content

Abstract

Hierarchical linear modeling (HLM) techniques were used to explore interactions among individual and contextual variables and their effects on initial self-esteem levels and changes in these levels across the three years of middle school in a large, diverse sample (N=1804 students attending 23 schools). Interactions among race, social class, school socioeconomic status (SES), and contextual congruity were found to be particularly influential. Low-income European-American youth consistently experienced the lowest self-esteem levels and the sharpest declines during middle school. Social incongruity appears to play a crucial role in predicting such losses. These results highlight the need for further research employing similar techniques to explore how individual and contextual factors affect the development of self-esteem over the middle school years.

Introduction

Adolescence is a period of dramatic change that often sets the stage for losses in positive feelings of self-worth (self-esteem). The identity struggles and egocentrism of adolescence can contribute to painful emotions, and a greater emphasis on peer relations often ignites youth's concerns about their own social skills and others' sincerity and allegiances (Way, 1998). These changes appear to be particularly difficult for young adolescents, who are often coping simultaneously with the onset of puberty and the transition to an unfamiliar and possibly stressful middle school setting (Seidman, Allen, Aber, Mitchell, & Feinman, 1994). Negative views of the self, in turn, constitute a risk factor for emotional difficulties (e.g., depression) and engagement in maladaptive behaviors (e.g., delinquency) (Harter, 1999).

Steep declines in self-esteem, however, are neither universal nor inevitable, and a range of individual and contextual factors appears to influence both the direction and magnitude of change (Hirsch & DuBois, 1991). Individual background variables such as gender, race, and social class have been shown to play important roles in determining trajectories of adolescent self-esteem (DuBois, Burk-Braxton, Swenson, Tevendale, & Hardesty, 2002). For example, although early studies suggested that African-American and ethnic minority adolescents suffered from steeper declines in self-image relative to European Americans, more recent work indicates that the self-esteem of African-American adolescents is comparable to or even higher than that of European-American peers Gray-Little & Hafdahl, 2000, Twenge & Crocker, 2002.

Gender can also affect changes in self-esteem during the early adolescent years. Girls consistently experience sharper declines than boys in their levels of self-esteem (Carlson et al., 2000, DuBois et al., 2002). Interestingly, African-American girls appear to be less vulnerable to disturbances in self-esteem than European-American girls Harter, 1999, Kling et al., 2000, McRae, 1991, Zimmerman et al., 1997. Several explanations have been given for these advantages, including the strong sense of ethnic identity among many African-American adolescents Gray-Little & Hafdahl, 2000, Twenge & Crocker, 2002, Ward, 2000.

Socioeconomic status has also been positively associated with self-esteem, and to a greater degree for European Americans than for African Americans (Demo & Savin-Williams, 1983). This finding has been attributed, at least in part, to differences in coping strategies and levels of support from extended family and community members around issues of class and discrimination Bowman & Howard, 1985, Fischer et al., 2000. The positive association between self-esteem and income level might be mediated by middle-class adolescents' relatively higher levels of academic achievement, as positive performance in school can translate into the development of a more positive self-image Campbell et al., 2002, Featherman, 1980.

Beyond these individual and family background characteristics, a range of contextual factors can also influence adolescent self-esteem. For example, researchers have noted the particular vulnerabilities of rural youth, who tend to be more isolated and to have fewer educational, recreational, and other public health resources Apostal & Bilden, 1991, Markstrom et al., 2000, Murray & Keller, 1991. In addition, relative to urban and suburban communities, rural settings are characterized by greater isolation, fewer educational and other public health resources, and higher levels of poverty Apostal & Bilden, 1991, Mayhew & Lempers, 1998, Murray & Keller, 1991. This, in turn, may lead to lower aspirations and fewer opportunities for success Housley et al., 1987, Markstrom et al., 2000.

The “goodness of fit” or congruity between the adolescent and his or her overall context can also influence self-esteem. Early research on the development of self-concept indicates that adolescents who live in a social environment or go to a school in which their religious, racial or socioeconomic group is in the minority are more likely than those who do not experience such dissonance with their immediate contexts to have self-image problems Rosenberg, 1965, Rosenberg, 1979.

This claim has been borne out in more recent work suggesting that racial and socioeconomic dissonance can contribute to low self-esteem and poor academic performance (Gray-Little & Carels, 1997). Some studies have found African-American teenagers, for example, to have a higher opinion of themselves when they go to schools in which African-American students are a majority than when they attend predominantly European-American schools, where they may feel out of place and under pressure to play down their cultural heritage (Ward, 2000). Hispanic–Americans and European Americans also appear to be at higher risk for maladjustment when they are the racial minority in their schools Kaufman et al., 1990, Twenge & Crocker, 2002. Research on consonance and dissonance at the neighborhood level has indicated a self-esteem advantage for youth living in communities where they are part of the ethnic and/or religious majority Duncan, 1994, Gerken et al., 1984, Rosenberg, 1975. Youth experiencing racial, socioeconomic, or religious dissonance in their schools and communities can sense sharp differences between their own self-images and the ways they are perceived in their immediate surroundings. Once internalized, negative reflected appraisals can precipitate declines in self-esteem Arunkumar et al., 1999, DuBois et al., 1998. In this sense, a dissonant school setting represents a “developmental mismatch” (Eccles & Midgley, 1989) between adolescents' heightened needs for positive peer appraisals and their surrounding indifferent or hostile school context (Clements & Seidman, 2002).

In summary, adolescence presents both challenges and opportunities, which interact with a range of individual and contextual factors to produce different pathways of self-esteem. Examining these pathways across youth of varying ethnicity, social class, gender, and social contexts may help to illuminate the extent to which changes in self-esteem are a function of individual and contextual factors.

To date, however, most of the studies cited above have relied on relatively small, homogeneous samples of adolescents. Additionally, the techniques that have been used to understand the nature of these variables have generally employed cross-sectional rather than longitudinal techniques, and have focused on correlational to mean level analyses of subgroups. Similarly, typical multivariate designs reflect tacit assumptions that adolescents who represent a wide range of adjustment and who follow multiple trajectories can be meaningfully grouped. Since young adolescents are adjusting to and coping with different challenges across middle school, such assumptions may be misleading. Although many researchers have examined one or two of these background variables at a time, the need for a comprehensive study including age, gender, race, sexual orientation and socioeconomic status, and the degree of dissonance or incongruity that youth experience in their middle-school settings, remains. Finally, although self-esteem as a construct is acknowledged to be fluid, previous studies have treated it as static, thereby failing to observe changes in self-esteem over time. By using growth curve modeling to isolate the effects of various individual and contextual variables, we hope to provide a more comprehensive, contextual understanding of trends in adolescent self-esteem.

Based on the literature reviewed above, we expected that the middle school years would be marked by overall declines in students' self-esteem. Within this context, we expected that trajectories of self-esteem would vary according to students' gender, race, SES, and congruity with their social context. Females, European-Americans, lower-income adolescents, and adolescents who attend schools in which they are the racial or socioeconomic minority were expected to be at greatest risk for drops in self-esteem over the three years of middle school.

Section snippets

Source of data

Data were drawn from a larger ongoing longitudinal evaluation study of young adolescents in middle schools.1

Unconditional model

The unconditional individual growth model for global self-esteem was estimated first (see Table 1). The estimated overall mean initial status across schools was 3.10 and was significantly different from zero (p<.001). The overall mean growth rate across schools was β=.07 and p<.001. Students across all schools report declines in self-esteem at the rate of .07 units per academic year. The significant variance components reveal that there is significant heterogeneity among students within

Discussion

The results of this study suggest that it is not only race, social class, or the material advantages of the school or community per se that influence developing self-esteem among adolescents, but also the students' congruence with the racial or socioeconomic environment of their schools. As predicted, social incongruity and poverty were both associated with lower self-esteem among youth. Two complex findings in the present study, concerning grade level and race–class interaction respectively,

References (44)

  • P. Clements et al.

    The ecology of middle grades schools and possible selves

  • M.G. Constantine et al.

    Black adolescents' racial socialization experiences: Their relations to home, school, and peer self-esteem

    Journal of Black Studies

    (2002)
  • R.M. Demo et al.

    Early adolescent self-esteem as a function of social class: Rosenberg and Pearlin revisited

    American Journal of Sociology

    (1983)
  • D. DuBois et al.

    Early adolescent self-esteem: A developmental–ecological framework and assessment strategy

    Journal of Research on Adolescence

    (1996)
  • D.L. DuBois et al.

    Self-esteem and adjustment in early adolescence: A social-contextual perspective

    Journal of Youth and Adolescence

    (1998)
  • D.L. DuBois et al.

    Race and gender influences on adjustment in early adolescence: Investigation of an integrative model

    Child Development

    (2002)
  • DuBois, D. L., Felner, R. D. (1991). Reliability and validity data for the self-esteem questionnaire. Unpublished raw...
  • G.J. Duncan

    Families and neighbors as sources of disadvantage in the schooling decisions of white and black adolescents

    American Journal of Education

    (1994)
  • J.S. Eccles et al.

    Stage-environment fit: Developmentally appropriate classrooms for young adolescents

  • S. Erkut et al.

    Puerto Rican early adolescents' self-esteem patterns

    Journal of Research on Adolescence

    (2000)
  • D. Featherman

    Schooling and occupational careers: Constancy and change in worldly success

  • C.B. Fischer et al.

    Discrimination distress during adolescence

    Journal of Youth and Adolescence

    (2000)
  • Cited by (50)

    • Part-time special education predicts students’ reading self-concept development

      2018, Learning and Individual Differences
      Citation Excerpt :

      We expect that the levels of academic self-concepts of students receiving part-time special education support while otherwise studying in mainstream classes will be lower than that of their peers (Allodi, 2000; Bear et al., 1991; Bear et al., 2002; Chapman, 1988b; Pijl & Frostad, 2010; Zeleke, 2004); and We expect that there will be no differences in the trends of self-concept development between girls and boys (Bear et al., 2002; Cole et al., 2001; Rhodes et al., 2004; Wei & Marder, 2012). Since there are no longitudinal studies on the effect of part-time special education support on self-concepts for students studying in the mainstream setting, and findings on the relationship between special education support and development of academic self-concepts are inconclusive and contradictory, no hypothesis was developed on part-time special education's effect on the trend of academic self-concepts.

    • Calibration of self-evaluations of mathematical ability for students in England aged 13 and 15, and their intentions to study non-compulsory mathematics after age 16

      2014, International Journal of Educational Research
      Citation Excerpt :

      Variations have been observed, however, with mathematics self-concept decreasing from Grades 7 to 9, but then increasing again in Grades 10 and 11 (Marsh, 1989). Gender differences over time have varied across studies, with the gap sometimes increasing (De Fraine et al., 2007), sometimes narrowing (Fredricks & Eccles, 2002), and sometimes remaining constant (Marsh, 1989; Rhodes et al., 2004; Watt, 2004). Gender differences have also emerged in other attitudes, such as girls exhibiting greater declines over time in the intrinsic value they associate with mathematics (Fredricks & Eccles, 2002).

    View all citing articles on Scopus
    View full text