Abstract:

Translating courses and programs in data science education while fostering diverse perspectives can present significant challenges in today's skill-demand-driven landscape. Educators are increasingly realizing the value of adapting and translating successful courses and programs and sharing insights gained in promoting diversity within data science education. In this context, we detail our experiences adapting a lower-division data science course from the University of California, Berkeley, a large public R1 university, to Mills College, a small liberal arts college for female-identified students.

While both institutions share the goal of offering an introductory data science course to students from various majors, they differ markedly in terms of student populations, institutional environments, and class sizes. We outline the key modifications made in course infrastructure and content, and share lessons learned. These lessons, drawn from the balance of structure and flexibility, as well as experiences across different scales and institutional contexts, contribute to discussions on promoting diversity within the data science field. Throughout our adaptation process, we encountered challenges and, in this paper, we discuss some strategies to overcome them.

Our findings underscore the importance of adapting courses to align with current curricula, student demographics, technological infrastructure, and the resources available to faculty. Additionally, smaller class sizes provide the opportunity to design tailored assignments that resonate with students' academic majors, career aspirations, and passion for driving social change.

Keywords:

Education, data science, adoption.