Examining Gender Differences in Writing Skill with Latent Factor Modeling

Abstract

Male students have historically under-performed in writing tasks compared to their female counterparts. This study examined writing skill differences between the male and female subjects of a sample of 4th and 5th grade students in the Pacific Northwest (n = 309) using an analysis of variance (ANOVA) to test for a possible gender effect on writing skill. A multiple-group confirmatory analysis (CFA) was used to examine possible structural differences between writing factors across gender groups, and a structural equation model (SEM) was used to examine writing factors predicting an essay task outcome for both gender groups. Despite evidence of a significant main effect of gender on writing within the sample, the multiple-group CFA indicated no significant structural differences between writing factors across the gender groups. The SEM results were not reported for this study because of the extremely high covariance between the writing factors in the model. The results of both the ANOVA and CFA are similar to other studies in the literature, and support a gender similarities hypothesis.