Development of a Data-Driven Pathway into STEM

Denise Hum
Skyline College

STEM classes, particularly introductory computer science and engineering classes, traditionally have few women (Kahn et al., 2017) and under-represented minorities enrolled (Kricorian et al., 2020). Skyline College, a Hispanic Serving Institution (HSI) in the San Francisco Bay Area, is no exception. A 2019 state-mandated shift in assessment and placement practices has resulted in more students who are decidedly non-STEM majors being placed directly into Introduction to Statistics. Historically, this has been a terminal math course designed for non-majors; however, Skyline College is using this new opportunity to increase the number and diversity of STEM students and improve STEM learning and teaching by bringing coding and project-based learning to statistics students.

Guiding Question
Instead of Introduction to Statistics being the last math class for students, can it serve as an onramp to data science, computer science, engineering, and other STEM disciplines? Can faculty and staff recruit more students – including women and underrepresented minorities – into STEM through engaging them in a redesigned, project-based Introduction to Statistics course where they also learn to code?

By teaching students to code in the context of analyzing data and conducting original research, a much more diverse student group is learning how to code in project-based Introduction to Statistics compared to Introduction to Computer Programming and Introduction to Engineering classes, the traditional introductory STEM classes. About 1/3 of Introduction to Statistics students want to learn more coding, and almost 35% of them also wanted to learn more statistics. An increase in students in Introduction to Data Science and Introduction to Computer Programming in Python, a new brand-new course created for the data science pathway, report to have taken Introduction to Statistics class. Over 50% of students in the Data Path courses are female compared to the traditional introductory STEM courses. In terms of ethnicity, over 35% of students in the Data Path courses are underrepresented minorities compared to less than 25% of students in the traditional introductory STEM courses. Furthermore, faculty have expressed much more satisfaction in teaching the project-based statistics curriculum and have been actively collaborating and developing new curriculum for the course through the Statistics Community of Practice.

Broader Impacts
Data science, an emerging and continually evolving interdisciplinary STEM field, provides an opportunity rethink curriculum and pedagogy to increase access and remove barriers to STEM degrees and careers (Cobb, 2015). Introduction to Statistics has historically had a much more diverse student body – in terms of both gender and race and ethnicities — compared to Introduction to Computer Programming and Introduction to Engineering classes. By meeting students where they are at, having them work on a project of their choosing while learning how to code, work with data, and develop their quantitative and problem-solving skills, we can use Introduction to Statistics and a Data Science pathway to train and recruit a much more diverse group of students into STEM.