Gail Crombie

All-Female Classes in High School Computer Science: Positive Effects in Three Years of Data.

In a three-year study, female students from all-female computer science (CS) classes were compared to male and female students from mixed-gender CS classes. Participants were 250 students enrolled in an elective Grade 11 CS course (63 females from three all-female classes and 155 males and 32 females from nine mixed-gender classes). Participants completed a questionnaire assessing perceived support from teachers and parents, computer-related attitudes, and future academic and occupational intentions. Females from all-female classes reported higher levels of perceived teacher support, confidence, and future academic and occupational intentions than did females from mixed-gender classes. Females from all-female classes reported levels as high as those reported by males on perceived teacher support, whereas males reported higher levels than did females from mixed-gender classes on perceived teacher support, confidence, intrinsic value, and future intentions. The present study provides some initial empirical evidence supporting the positive effects of all-female classes in CS at the high school level.

Effects of classroom gender composition on adolescents' computer-related attitudes and future intentions.

Differences in computer-related attitudes and future intentions between females in an all-female class and males and females in mixed-gender classes were examined. Participants were sixty-seven students enrolled in a Grade 11 computer science course which was offered in both all-female and mixed-gender classes. Students completed a questionnaire assessing perceived support of teachers and parents, computer-related attitudes, and future intentions. Females from the all-female class reported greater perceived teacher support than either males or females from the mixed-gender classes and were similar to males in confidence, intrinsic motivation, and future academic intentions. To the extent that differences can be attributed to classroom gender-composition, obtained results indicate that an all-female computer science classroom environment may enhance the learning experiences of female students.

Self-Perceptions of Ability and Achievement Across the High School Transition: Investigation of a State-Trait Model

The authors investigated the relationship between self-perceptions of ability and achievement in math, science, and English from Grades 8 to 11 (N = 342). A state-trait model that included an association between stable (i.e., trait-like) components of self-perceptions and achievement as well as time-specific (i.e., state-like) effects during the transition to high school (i.e., Grade 8 to Grade 9) demonstrated superior fit to alternative models that did not incorporate these features. Stable components of self-perceptions of ability and achievement exhibited a substantial association in this model. In most instances, however, there also was evidence of a positive effect of self-perceptions in Grade 8 on achievement in Grade 9.

Bridging the Gender Gap in High-Technology Education

Low enrollment rates in high school computer science (CS) courses suggest that students may be planning ineffectively for a future labor market. Female students enroll in these courses at far lower rates than their male peers and represent a small proportion of the high-technology workforce. To increase participation, all-female and mixed-gender CS classes were offered in grade 11 at one high school in Canada. Enrollment patterns were observed for three years. Results show increased participation, which increased the number of CS courses offered at the school. Implications are discussed.

Children's Problem-Solving Performance: The Effects of Compliance and Cognitive Factors

Abstract This study was designed to examine the robustness of Crombie and Golds finding (1989) that a high degree of compliance is negatively related to childrens problem-solving performance. The effect of compliance on childrens problem-solving performance was examined by extending Crombie and Golds investigation to a different task, assessing the effects of a training session in which childrens nonreliance on an adult model was reinforced, and measuring the contribution of relevant cognitive factors. Childrens problem-solving competence was examined with a standardized measure of general problem-solving performance and a task-specific measure. Age, verbal intelligence, and task-oriented memory were entered as covariates. High-compliant children did not perform as well as their low-compliant peers on the general problem-solving measure, and this group difference remained significant after the effects of the covariates were taken into consideration. On the task-specific measure, which requires respo...