Students' Test Anxiety and Performance in Introductory Programming: Do Exam and Assessment Modalities Play a Role?

Abstract

Many students encounter programming for the first time during undergraduate studies. Programming, being different from conventional elementary and middle school subjects, maybe a source of anxiety and frustration for these students. It is already known that heightened anxiety can negatively impact test performance, that females are more test anxious than males, and that many introductory computer science courses suffer from high dropout rates. This is the second part of our two-fold study, which aims to answer whether exam and assessment modalities influence students' test anxiety and the (possible) correlation between test anxiety and performance in an introductory programming course. The research questions and most important findings are as follows: (i) female versus male students' test anxiety regarding a computer-based automatically assessed exam (female students exhibited higher levels of test anxiety than male students), (ii) test anxiety regarding a paper-based manually assessed exam versus test anxiety regarding a computer-based automatically assessed exam (test anxiety during a paper-based manually assessed programming exam was lower than the latter), (iii) general test anxiety versus test anxiety regarding a computer-based automatically assessed exam (only female students were more test anxious on a computer-based automatically assessed exam than in general), (iv) test anxiety regarding a paper-based manually assessed exam and exam score (no correlation found), and (v) test anxiety regarding a computer-based automatically assessed exam and exam score (statistically significant, moderately negative correlation found only for female students). Our findings suggest that efforts should be made toward addressing female students' test anxiety in introductory programming as their well-being and exam performance are affected the most.