A differentiated picture of student performance in introductory courses to theory of computation. Issue 3 (3rd July 2021)

• Record Type: Journal Article
• Authors: Frede, Christiane; Knobelsdorf, Maria
• Is Part Of: Computer science education. Volume 31:Issue 3(2021)
• Page Start: 315
• Page End: 339
• Publication Date: 2021-07-03
• Subjects: Computer science education -- theory of computation -- exploratory data analysis -- student performance; Computer science -- Study and teaching -- Periodicals; 004
• DOI: 10.1080/08993408.2020.1809946
• Languages: English
• ISSNs: 0899-3408

Abstract

ABSTRACT: Background and Context: Considerable numbers of Computer science (CS) undergraduate majors struggle in Theory of Computation (ToC) courses, which strengthen bimodality beliefs of student performance. Reasons for students struggling are assumed to be manifold but substantial ground is based on studies providing singular insights into this matter. Objective: To understand students' issues with ToC, we analysed their performance on homework and exams and replicated our study in two consecutive years. Method: We used exploratory data analysis to compare students' performances within single assignments. With one-way analyses of variances, we focused on significant differences between the years. Findings: Our results indicate that students' homework performance did not predict their exam performance and that all students, independent of their grade, had the most difficulty with formal proofs. Implications: Our results challenge potential beliefs about bimodal student performance indicating all CS undergraduates require thorough training in proof techniques to improve their overall performance in ToC.