Academic achievement--Evaluation

For 306 twelfth grade students, secondary science
achievement by sex, race, and socioeconomic status (SES) was
compared using analysis of variance (ANOVA), and the
relationship of a set of 22 predictor variables to science
achievement was investigated using multiple linear
regression. That the reduced regression model with its primarily
cognitive characteristics exhibited the highest significant
correlation coefficients in predicting science achievement
for low SES, black students is a finding that is at odds with
much of the current literature, as is also the finding that
females outperform males in science achievement. These
anomalies, along with the implications of differences in
science achievement based on sex and race, are discussed.

This study explored the impact of peer reputations for academic ability and school-related affect upon fourth- through eighth-graders' academic outcomes. In light of the prevailing stereotypes regarding the differential academic abilities of girls and boys (favoring girls in English, and boys in math and science), it was hypothesized that peer reputations in English would be more predictive of outcomes for girls than for boys, while math and science reputations would be most predictive of boys' outcomes. Peer reputations were found to be predictive of school grades, but not of standardized test scores. Although ability reputations in most areas were predictive of grades for both sexes, modest gender differences were observed which were consistent with hypotheses. Results suggest that children's peer reputations may play an important role in their academic achievement, especially within domains most central to their gender identities. Implications and suggestions for future directions are discussed.

This study explored the extent and nature of academic peer reputation upon children's academic self-concept. Peer and self-perceptions of academic ability and affect were assessed for the subject domains of English, math, and science in order to investigate the generality of peer reputation influences across academic subject areas, and determine the extent to which gender differences might be evident. Gender differences were hypothesized, and peer reputation was expected to have the most influence on academic self-concept in school domains viewed as gender-normative and thus central to self-concept. MR analyses provided some support for this gender-congruency hypothesis, as it was primarily children's academic ability reputation in gender-congruent areas that was predictive of ability self-concept, and, influential with respect to perceived ability in normative domains. Contrary to expectations, peer affective reputation was more predictive with respect to gender-incongruent domains. Results were interpreted to suggest the dominance of gender-schema consistency concerns in relation to ability perceptions, but contrast effects in relation to school affect.