Innovations and Opportunities

SIGCSE 2022 Affiliated Event by the SIGCSE Committee on Computing Education in Liberal Arts Colleges

Helping Non-CS Faculty Teach Core CS Concepts

Mount Holyoke College, Colby College, Union College

Contributed by Valerie Barr, vbarr@mtholyoke.edu

Institutional and departmental context

Description of Curricular Innovation

This project was designed to address two key concerns about the student experience in liberal arts settings: diversity concerns about who is learning about computing, and concern about the computing capabilities gained by non-CS students over the course of their studies. It takes advantage of the liberal arts setting to motivate learning computer science by embedding it within courses in each student’s discipline.

Increasingly, students in non-CS disciplines are expected to incorporate computing into their class work. Not all faculty, despite being able to use computing in their own scholarly work, are able to effectively teach core concepts of computing to their students. As a result, we run the risk that non-CS students are largely being equipped to do only “plug and chug” kinds of computing work without true comprehension. They are not being taught the computing concepts that will enable them to transfer their computing skills to new settings, new languages, and new application venues as they move on to new courses and to jobs and graduate study after college.

At the same time, many non-CS disciplines are far more diverse than is computer science. While we have seen an explosion in CS enrollments, and considerable work on improving diversity in the field, change in overall diversity of the student population is slow. In addition, we know that no computer science department, especially at liberal arts colleges, is able to teach every student at their institution. The goal of this project is to equip a more diverse student body with understanding of core computing concepts, and to do so by reaching them in non-CS courses that utilize computing.

The key curricular innovation of this project is the development of an extensive video suite (currently 19 videos) that can be incorporated by non-CS faculty into their course activities. This shifts off of non-CS faculty the burden of teaching those concepts, and makes it possible for students to learn about these concepts from CS faculty (in particular, CS faculty who are already engaged in teaching to students from a range of disciplines in a liberal arts college setting). The concepts addressed start with true basics, such as the importance of computing and why one would bother with core concepts, and move on to material that students will encounter as they start to use computing. Example topics include:

The core concepts were synthesized after extensive review of course materials provided by a core group of participating faculty across three campuses and results from a survey of a larger number of faculty. This process enabled us to determine the concepts most relevant in the non-CS courses and how the level of mastery sought by the non-CS faculty differed across different disciplines based on the type of computing being carried out in their courses.

Challenges/Limitations

This project has been carried out, so far, completely during the Covid pandemic. Even though all three campuses were in-person in Fall 2021, it is hard to say what impact, if any, there was on the ability of the non-CS faculty to modify their courses. By the time of the SIGCSE Liberal Arts event we will, however, have evaluation results from the Fall 2021 participating faculty and from the students in their classes.

The project team faced the reality that non-CS faculty are using different computing environments. Primarily, the non-CS faculty associated with this project were using Python or R in their courses. Because of that, the project team generated videos that present concepts in three ways: generic concepts, the concepts presented in Python, and the same concepts presented in R. This approach helps the students understand that the core concepts are relevant across applications of computing and can be applied in the language or computing environment that is most appropriate.

Another impact of Covid on this project is that it took much longer than anticipated for the project to get started, and we have not been able to produce the full set of videos originally imagined. In addition, despite the participating CS faculty doing considerable online teaching ourselves, we are not experts at video production.

Acknowledgements

This work is supported by the National Science Foundation under Grant Numbers CUE Ethics 1935115, 1935099, and 1935061.