Publications and Research
Document Type
Article
Publication Date
Summer 7-21-2026
Abstract
Background and Context: Motivational beliefs influence academic choice and persistence and are an important factor in understanding the underrepresentation of women in undergraduate computer science courses and the field more broadly.
Objective: This study examines the relations between expectancy, value, and cost, and two distinct measures of self-reported behavioral engagement (i.e., weekly lecture engagement and end-of-semester engagement) and investigates gender differences in an undergraduate computer science course.
Method: Self-reported expectancy, value, and cost were collected from 346 undergraduate CS students at the beginning of the semester. Self-reported behavioral engagement data was collected both weekly in lectures throughout the semester for 12 weeks at the end of each class and retrospectively at the end of the semester. Latent growth curve modeling was used to exam how expectancy, value, and cost (and interactions with gender) predict initial and trajectory of weekly engagement, and end-of-semester engagement.
Findings: Results indicated that initial perceived value positively predicted self-reported weekly lecture behavioral engagement, that men reported higher levels of expectancy for success than women, and that gender moderated the relationship between expectancy for success as well as cost and self-reported weekly lecture engagement, but not end-of-semester engagement. Furthermore, results suggest that men’s self-reported weekly lecture engagement levels were more variable and context-dependent across weeks, whereas women’s engagement tended to be more consistent throughout the semester.
Implications: Findings discuss situated expectancy-value theory, highlight important measurement considerations, and advance understanding of why women are underrepresented in computer science, such as gender differences in costs and weekly lecture engagement variability.
