Self-concept: differences among adolescents by gender.

Abstract

This study compared adolescents by gender on self-concept. Participants consisted of sixth, seventh, and eighth graders placed in general education classes or special day classes and designated by their teachers as either high-or low-achieving. Responses on the Student Self-Concept Scale were examined using a between groups MANOVA design. Findings were not significant when male and female adolescents were divided by class placement and rank within placement. However, one significant effect was found when analyzed without the division of class placement or rank within placement. Males scored significantly lower on social self-concept. The results suggest that there are differences between adolescent males and females on social self-concept.

Adolescent Self-Concept by Gender: A Review of the Literature

Students with mild disabilities such as learning disabilities, mild mental retardation, behavior disorders, and attention deficit disorder have frequently been shown in the special education literature to be poorly accepted or rejected by peers without disabilities (Gresham & MacMillan, 1997). The majority of these studies have been conducted with elementary-age samples. Adolescent samples tend to be less frequently utilized in self-concept studies and are sparse, inconsistent, and largely substantive in nature (Byrne & Shavelson, 1987). A major hindrance to adolescent self-concept research has been that the majority of studies have used inadequately defined constructs which relate only to either general self-concept or academic self-concept whereas self-concept research has lent credence to the notion that self-concept is a multidimensional construct.

Differences in self-concept among adolescent males and females varies throughout the literature. Three studies found no differences between males and females in their rating of global or general self-concept (Crain & Bracken, 1994; Marsh, 1993; Osborne & LeGette, 1982). Others found large and consistent differences among adolescents (Byrne & Shavelson, 1987; Keltikangas-Jarvinen, 1990; Markstrom-Adams & Adams, 1995; Marsh, 1994; Stoner & Kaiser, 1978; Watkins, 1995). Statistically significant gender differences were found in a study of 901 Australian adolescents (Bryne & Shavelson, 1987) and replicated by other researchers: girls had higher English self-concept and boys had higher mathematics self-concept (Bryne & Shavelson, 1986a; Marsh, 1993; Meece, Parsons, Kaczala, Goff, & Futterman, 1982). Overall, males tend to have higher self-concept scores on dimensions such as math, emotions, physical abilities, physical appearance, and general self-concept. Females are more likely to score higher in areas such as verbal self-concept, honesty, parental relations, and same-sex peer relations (Widaman, MacMillan, Hemsley, Little, & Balow, 1992).

Whereas few gender differences were found among elementary school children, significant differences were found in certain studies among adolescents (Byrne & Shavelson, 1987). For instance, boys exhibited higher mathematics self-concept than girls, whereas girls have consistently higher English self-concept than boys. In addition, differences were found in the subscale scores of males and females on both the Piers-Harris Children's Self-Concept Scale and the Coopersmith Self-Esteem Inventory (Osborne & LeGette, 1982). Females scored significantly higher on the behavior and social self scales, while males, by comparison, revealed significantly higher levels of anxiety. Males also tended to express higher levels of satisfaction with their physical selves than did their female classmates. Hattie (1992) summarized gender differences in self-concept as the following: males had higher general, physical, and math self-concept and females had higher verbal self-concept. Wigfield, Eccles, MacIver, Rueman, and Midgely (1991) reported that this pattern of gender differences was maintained before, during, and after the transition from sixth grade to junior high school.

Primary Objective

The primary objective of this research was to systematically determine differences in self-concept among four groups of adolescents by gender: low-achieving adolescents in special day classes, high-achieving adolescents in special day classes, low-achieving adolescents in general education, and high-achieving adolescents in general education.

The four groups of adolescents were specifically created to first capture the differences between males and females for those who were labeled as special education students and those who were in the general education program. Second, there was interest in the effects of rank within placement (low-achieving and high-achieving) as an important variable because it clarifies the relative standing of the students in both programs. By identifying a student's class placement and rank within that placement, it was anticipated that it would reveal how the experiences of being either the "best" or "worst" student in a particular setting can influence self-concept.

Methods and Procedures

This study took place in two middle schools in one of the largest elementary districts in the state of California. This district is located in a suburban area with a large number of low-income families of minority descent.

Participants

Teachers

Thirteen teachers readily agreed to participate by allowing the principal investigator to utilize two class periods over a two-week period of time. Overall, the study utilized seven teachers of special education (mild/moderate special day class programs), two teachers from the lowest track of general education, and two teachers from the highest track of general education.

Students

The entire sample (N = 145) consisted of 72 males (50%) and 73 females (50%). The sample was also divided evenly by high and low-achieving groups. The high-achieving group consisted of 73 participants (50%) while the low-achieving group consisted of 72 participants (50%). There were 37 sixth grade students (26%), 64 seventh grade students (44%), and 44 eighth grade students (30%). Nineteen percent of this sample was White (n = 28) while just six percent (n = 8) of the sample was Black. The majority of the adolescents in this sample (n = 109) were Hispanic (75%)(See Table I). See issue's website <http://rapidintellect.com/AEQweb/sum02.htm>

Sampling Procedures

An attempt was made to select equal numbers of participants from both general and special education. Stratified random sampling was also used in order to secure equal numbers of males and females in both settings.

Special Education Sample Selection A total of thirty-nine students in special day classes, using the SSRS-T, were rated by their teachers as low-achievers whereas thirty-one students were rated by their teachers as high-achievers. Therefore, this group of seventy special day class males and females were divided according to their relative standing within the special day class as two groups differentiated by their rank within placement as either low-achieving or high-achieving.

General Education Sample Selection Teachers from four classes in the lowest track and highest track of general education were asked to rate the males and females in their classes on the academic competence portion of the SSRS-T.

Lowest Track Within the lowest track of students in general education, fifty-five out of one hundred eight students were rated by two teachers as the low-achieving general education group. Of fifty-five students in the general education program who had been rated by their teachers as low-achieving, only thirty-three parents (60%) submitted signed consents.

Highest Track Teachers who teach in the highest track rated their students on the academic competence portion of the SSRS-T. Forty-two students were rated as high-achieving. As a result, the sample group of students included, in effect, the lowest-achieving adolescents from the entire general education population as well as the highest-achieving students in general education. Additionally, the majority of the adolescents in the high- achieving general education sample participated in the Gifted and Talented Education (GATE) program during the 1998-1999 school year. In summary, a total of 196 students were selected to participate in this study. The investigators received a 74% rate of cooperation overall. See issue's website <http://rapidintellect.com/AEQweb/sum02.htm>

Assessment Tool Developed by Gresham, Elliott, and Evans in 1992, the Student Self-Concept Scale (SSCS) is a 72-item multidimensional measure of self-concept. It is norm-referenced and provides a reliable method of measuring the self-concept of children and adolescents in grades 3-12. The SSCS documents perceived confidence in performing, perceived importance of performing, and perceived confidence in the likelihood of outcomes from performing specific behaviors, influencing the development of self-concept (Gresham, Elliott, & Evans, 1992).

Experimental Design

The experimental design for this study was a between-group 2 X 2 X 2 (class placement X rank within placement X gender) mixed model. This design allows for measurement of the dependent variables between groups based on class placement, rank within placement and among gender.

Results

Self-concept of the adolescent groups was analyzed to determine if there were differences between the groups based on their class placement and rank within placement by gender on the dependent variables of academic self-concept and social self-concept. All analyses were performed using the SAS package of computer programs for statistical analysis.

First, the multivariate test for the interaction of class placement X rank within placement X gender on the two confidence subscales of self-concept (academic self-concept and social self-concept) was nonsignificant (approximate F (2,136) = 0.54, p> .05). Likewise, the multivariate test for the two-way interaction of class placement X gender was also found to be nonsignificant when testing for significant differences on the academic self-concept confidence and social self-concept confidence subscales of the SSCS (approximate F (2,136) = 1.63, p>.05). Similarly, the interaction of rank within placement X gender was nonsignificant on the same two confidence subscales of the SSCS (approximate F (2,136) = 0.97, p>.05).

However, the multivariate test of differences on the two subscales (academic self-concept confidence and social self-concept confidence) of the SSCS for the effects of gender was significant, approximate F (2,140) = 6.40, p<.001. Therefore, it was appropriate to analyze the univariate effects. The main effects of gender were nonsignificant for academic self-concept confidence, approximate F (1,144) = 3.58, p>.05, yet significant for social self-concept confidence (approximate F (1, 144) = 12.09, p<.001; see Tables III and IV). See issue's website <http://rapidintellect.com/AEQweb/sum02.htm>

On the social self-concept confidence subscale of the SSCS males had an average score of 26.66 (SD = 6.20) which were significantly lower scores than those found for the female participants (M = 30.20, SD = 5.52; see Table V). Therefore, these findings validate the conjecture that males will score lower on social self-concept than females. However, the previously held belief that males will have higher scores than females on academic self-concept was not validated since no significant differences were found. See issue's website <http://rapidintellect.com/AEQweb/sum02.htm>

Discussion

This investigation resembles the findings of Dusek and Flaherty (1981), Marsh, Barnes, Cairns, and Tidman (1984), and Marsh, Parker, and Barnes (1985). These studies found that females had higher social self-concept than did males. In addition, females in the norm sample were more likely to have higher ratings on the social self-concept subscale on the SSCS across grade levels (Gresham, Elliott, & Evans-Fernandez, 1993). These findings support substantive research that females have higher social self-concepts than males of all ages.

One reason why females were found to have higher social self-concepts could be the higher verbal skills that are developed in females. Marsh (1993) discussed the fact that "sex-linked differences in socialization patterns may fail to reinforce adequately boys' positive attitudes, expectations, and performance in verbal areas" (p. 844). Likewise, he suggests that "stereotypic gender differences in mathematical and verbal areas emerge during early adolescence and grow larger during the adolescent years" (Marsh, 1993, p. 844). According to the current investigation, young males as well as adolescent males need more training in verbal areas to reinforce their social self-concept.

Overall, there has been limited research on gender when analyzing adolescent self-concept. However, one study found that eighth grade females had higher social self-concepts when looking at the dimension of same-sex peer relations (Widaman et al., 1992). In the same study, eighth grade males scored higher than females on the opposite-sex social self-concept dimension. In the current investigation, the primary self-concept tool used to measure social self-concept did not include any statements pertaining directly to same-sex relations nor to opposite-sex relations. The statements were written to include words such as "other kids my age" or "classmates." Therefore, it was not determined whether students used same-sex, opposite-sex, or both types of social comparisons.

In contrast, no differences were found between males and females on academic self-concept confidence. These findings are similar to the Widaman et al. (1992) study which did not find significant differences for gender on academic self-concept for eighth grade students. However, this investigation conflicts with several studies that found gender differences in self-concept areas related to academics (Byrne & Shavelson, 1987; Harter, 1982; Marsh, Barnes, Cairns, & Tidman, 1984; Marsh, Parker, & Barnes, 1985). Since this study did not analyze specific subjects within academic self-concept such as English self-concept or mathematics self-concept, differentiation between males and females in these areas was not possible. This could be one explanation for the lack of significance among gender when looking at academic self-concept confidence.

In summary, gender differences in self-concept research are conflicting. However, the findings in the current investigation revealed higher social self-concepts for adolescent females which is possibly related to their superior verbal skills. Continued research in the area of self-concept should focus on adolescent gender differences as well as interventions to improve the social self-concept of males.

References

Byrne, B.M., & Shavelson, R.J. (1986a, April). On gender differences in the structure of adolescent self-concept. Paper presented at the annual meeting of the American Educational Research Association, San Francisco.

Byrne, B.M., & Shavelson, R.J. (1987). Adolescent self-concept: Testing the assumption of equivalent structure across gender. American Educational Research Journal, 24, 365-385.

Crain, R.M., & Bracken, B.A. (1994). Age, race, and gender differences in child and adolescent self-concept: Evidence from a behavioral-acquisition, context-dependent model. School Psychology Review, 23, 496-511.

Dusek, J.B., & Flaherty, J.F. (1981). The development of self-concept during the adolescent years. Monographs of the Society for Research in Child Development, 46, (4, Serial No. 191).

Eccles, J.S. (1987). Gender roles and achievement patterns: An expectancy value perspective. In J.M. Reinish, L.A. Rosenblum, & S.A. Sanders (Eds.), Masculinity/femininity: Basic perspectives (pp. 240-280). New York: Oxford University Press.

Gresham, F.M., Elliott, S.N. (1990). Social Skills Rating System. Circle Pines, MN: American Guidance Service, Inc.

Gresham, F.M., & MacMillan, D.L. (1997). Social competence and affective characteristics of students with mild disabilities. Review of Educational Research, 67, 377-415.

Keltikangas-Jarvinen, L. (1990). The stability of self-concept during adolescence and early adulthood: A six-year follow-up study. The Journal of General Psychology, 117, 361-368.

Markstrom-Adams, C., & Adams, G.R. (1995). Gender, ethnic group, and grade differences in psychosocial functioning during middle adolescence? Journal of Youth and Adolescence, 24, 397-417.

Marsh, H.W. (1993). The multidimensional structure of academic self-concept: Invariance over gender and age. American Educational Research Journal, 30, 841-860.

Marsh, H.W. (1994). Using the national longitudinal study of 1988 to evaluate theoretic models of self-concept: The Self-Description Questionnaire. Journal of Educational Psychology, 86, 439-456.

Marsh, H.W., Barnes, J., Cairns, L., & Tidman, M. (1984). Self-description questionnaire: Age and sex effects in the structure and level of self-concept for preadolescent children. Journal of Educational Psychology, 76, 940-956.

Marsh, H.W., Parker, J., & Barnes, J. (1985). Multidimensional adolescent self-concepts: Their relationship to age, sex, and academic measures. American Educational Research Journal, 22, 422-444.

Meece, J.L., Parsons, J.E., Kaczala, C.M., Goff, S.B., & Futterman, R. (1982). Sex differences in math achievement: Toward a model of academic choice. Psychological Bulletin, 91, 324-348.

Osborne, W.L., & LeGette, H.R. (1982). Sex, race, grade level, and social class differences in self-concept. Measurement and Evaluation in Guidance, 14, 195-201.

Stoner, S., & Kaiser, L. (1978). Sex differences in self-concepts of adolescents. Psychological Reports, 43, 305-306.

Widaman, K. F., MacMillan, D.L., Hemsley, R.E., Little, T.D., & Balow, I.H. (1992). Differences in adolescents' self-concept as a function of academic level, ethnicity, and gender. American Journal on Mental Retardation, 96, 387-404.

Wigfield, A., Eccles, J.S., MacIver, D., Reuman, D.A., & Midgely, C. (1991). Transitions during early adolescence: Changes in children's domain-specific self-perceptions and general self-esteem across the transition to junior high school. Developmental Psychology, 27, 552-565.

Pierson, Ph.D., is an associate professor and was a special education teacher for seven years. Her research interests focus on affective characteristics. Dr. Glaeser is an assistant professor and was a special education teacher for 13 years. Her research interests are in the area of reading.