Font Size: a A A

Gender differences in mathematics ability: A meta-analysis of aptitude and achievement test scores

Posted on:1990-07-18Degree:Ph.DType:Dissertation
University:The University of IowaCandidate:Hengkietisak, SuneeFull Text:PDF
GTID:1477390017953222Subject:Education
Abstract/Summary:
Meta-analysis techniques were used to investigate the variation of effect sizes of gender differences in mathematics aptitude and achievement test scores. Unlike previous meta-analyses of gender differences, this study employed a random effects model to study the variation of effect sizes with respect to three variables: age of the subjects, selectivity of the sample, and calendar year of testing. This study investigated not only main effects but also interaction effects. In addition, a repeated measures analysis was used to study the gender differences with respect to three types of mathematics skills: computation, concepts, and problem solving. Where the data were available, effect sizes at the 10th, 75th, and 90th percentile levels were also examined in addition to the effect sizes at the level of mean.;The data sets in this study were large samples of subjects either from national norms published in test technical manuals or reports of research projects using national data. In addition, data from state assessments were included.;The results showed that 88 percent of the variation in the effect sizes was accounted for by the three variables: age of the subjects, selectivity of the sample, and the calendar year of testing. Effect sizes increased with age and selectivity. Age accounted for most of the variation in effect sizes (71.1%); selectivity accounted for an additional 11.7 percent of the variation in the effect sizes; and there was little variation (5.2%) in the calendar year when age and selectivity were controlled. At 14 years and younger, females' mathematics performance was slightly better than that of males. However, during the high school years and beyond, males outperformed females especially in the selected samples.;Tests for the three mathematical skills were available only for children 14 years and younger. Girls were advantaged in computation, but the gender differences for concepts and problem solving were almost negligible. Effect sizes increased with percentile levels; the greatest male advantage was found at the 90th percentile level.
Keywords/Search Tags:Effect sizes, Test, Gender, Mathematics, Variation
Related items