Addressing the Gender Gap in STEM MOOCs: How Brief, In-Course Messages Can Increase Females' Motivation and Online Learning Success

Women are routinely underrepresented in higher education science, technology, engineering, and mathematics (STEM). Despite lower admission barriers and more flexible schedules, massive open online courses (MOOCs) are not exempt from this trend, as female learners are less likely than males to enroll in and complete them. Using the ecological systems framework, this researcher explored factors contributing to the STEM gender gap across home country, course, and learner levels. During the mixed-methods needs assessment, female learners’ self-efficacy, time constraints, and home country’s level of gender inequality emerged as the three strongest factors related to the gender gap in retention and completion. The self-determination theory of intrinsic motivation was used to align the novel intervention prompts with females’ sense of competency, autonomy, and relatedness. The intervention study deployed text-based messages to students in 150 STEM MOOCs to tackle the primary identified needs: boost confidence, improve planning, and emphasize individuals’ values to counteract gender inequality. The Coursera platform was used to conduct a randomized controlled trial (RCT) experiment, allowing causal quantitative data analysis to assess the impact of these intervention groups on learners’ persistence, skill development, and continued learning. Females’ self-reported reasons for stopping before completion were also qualitatively coded by theme. This explanatory mixed-methods RCT study included 324,457 total active learners with identified gender. The four treatment groups (three variant types plus the combined treatment) each resulted in a significant increase in first-week completion rates for female learners compared with the control. The value relevance treatment group retained this significant increase, successfully eliminating the gender gap in STEM MOOC course completion. The self-efficacy treatment significantly raised the number of female course completers by 50% in the youngest age tier. Moving all active learners in this RCT from the control group to the value relevance treatment would result in approximately 1,400 additional female STEM course completers. Implications for future research and practice are explored, including the personalized deployment of the messages given differences in impact by age, gender, and geography