Over the last years, there have been great efforts to improve the diversity of engineering and computing education. Our work contributes to these efforts by following a data-driven approach to analyze easily accessible data and provide insights that could potentially lead to meaningful and impactful interventions. In particular, we focus on course selection and how we can balance the percentage of women in courses across engineering and computing courses. Deciding which courses to take every semester can be challenging for students. One of the factors influencing their decisions is the descriptions of the available courses. By reading these, the students get a first impression of the type and content of a course. This study reviews course descriptions offered by the college of engineering and computing. We employ natural language processing (NLP) approaches to identify patterns in the language used in course descriptions and how this relates to the student enrollment and descriptive characteristics of the different departments and courses. Our ultimate goal is to identify and quantify how different course descriptions are from different majors as these relate to the student gender distribution. Our language analysis indicates that verbs, adjectives, and adverbs have the most significant impact on differentiating course descriptions and highlighting differences across the different programs and across the different variables of focus. Implications of this work and the impactful dissemination include sharing results with faculty and staff within the college during departmental and college-wide meetings to encourage meaningful course description changes for their courses. This research adds significantly to the literature as there is very little research on the impact of course descriptions on students' course selection process.
