Abstract

Previous research supported the significance of gender differences in writing performance and also acknowledged the influences of cognitive (e.g., metacognition, critical thinking, verbal working memory) and motivational variables (e.g., writing beliefs) upon the quality of written composition. This study was conducted to examine whether the influences of writers' gender on writing performance would be significant above and beyond the influences of cognitive and motivational variables. One-hundred twenty-three 10th-grade students (53 males, 70 females) participated in this study. Four writing tasks, differentiated by text mode (i.e., narrative, descriptive, expository, persuasive), the verbal working memory test, the critical thinking test, the metacognitive knowledge about writing processes questionnaire, and the writing beliefs inventory were administered. First, the data were analyzed by a repeated-measures analysis of variance to test differences in writing performance by gender and writing task. The results showed that female students performed better on all of the writing tasks than did male students. In the male group, the performance level of narrative writing was higher than that of descriptive, expository, and persuasive writing. In the female group, the performance level of descriptive writing was higher than that of narrative, expository, and persuasive writing. Second, hierarchical regression analyses were conducted to test whether the influences of writers' gender upon writing performance on each writing task would be significant above and beyond those of cognitive and motivational variables. The results revealed that the influences of gender were no longer significant after writing performance on the narrative text was explained by metacognitive knowledge and critical thinking, and that they were no longer significant after writing performance on the expository text was explained by transactional writing beliefs. However, the influences of gender upon writing performance on descriptive and persuasive texts were significant above and beyond those of metacognitive knowledge. These results imply that it be necessary to consider not only students' gender but also their metacognitive knowledge about writing processes, critical thinking skills, and transactional writing beliefs when planning and implementing writing instruction programs or writing assessment. Finally, the suggestions for future studies were discussed.