Student affect in CS1

Abstract

Recent research, and our own experience as educators, has highlighted the need for an approach to CS1 that includes consideration of students' emotional state. Unfortunately, collection of affect data usually requires a large investment of time and resources. In this article, we describe a simple and easy tool for collection of student affect data. We illustrate how these data can provide detailed insight into the quality of curricular materials and make accurate predictions of student performance. Based on our first year using the tool, we identify specific response patterns that can identify a student at risk of CS1 failure. Students who find classroom exercises both difficult and boring, or who recognize a programming problem as involving familiar material but have no clear plan as to how to solve the problem, are likely to struggle as the course proceeds.