APSIPA Transactions on Signal and Information Processing > Vol 11 > Issue 1

Is Self-Rated Confidence a Predictor for Performance in Programming Comprehension Tasks?

Zubair Ahsan, Dept of Artificial Intelligence, Faculty of Computer Science and Information Technology, Universiti Malaya, Malaysia, Unaizah Obaidellah, Dept of Artificial Intelligence, Faculty of Computer Science and Information Technology, Universiti Malaya, Malaysia, unaizah@um.edu.my , Mahmoud Danaee, Dept of Social and Preventive Medicine, Faculty of Medicine, Universiti Malaya, Malaysia
 
Suggested Citation
Zubair Ahsan, Unaizah Obaidellah and Mahmoud Danaee (2022), "Is Self-Rated Confidence a Predictor for Performance in Programming Comprehension Tasks?", APSIPA Transactions on Signal and Information Processing: Vol. 11: No. 1, e5. http://dx.doi.org/10.1561/116.00000041

Publication Date: 21 Feb 2022
© 2022 Z. Ahsan, U. Obaidellah and M. Danaee
 
Subjects
 
Keywords
Machine learningexpertiseconfidenceprogramming comprehensioncomputer education
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 211 times

In this article:
1. Introduction 
2. Related Work 
3. Methodology 
4. Results 
5. Discussion 
6. Conclusion 
References 

Abstract

Studies on programming comprehension have focused largely on the type of reading strategies individuals employ. However, quite few programming comprehension studies have focused on the relationship between the self-rated confidence levels and the performance levels of the participants. In this study, our aim was to identify the effect of confidence levels among the participants as they attempt familiar programming questions. Our results indicate that due to familiarity, all participants generally show high confidence levels. High performers demonstrated self-rated high confidence levels as compared to low performers. However, the difference in confidence levels of high and low performers was found non significant. Furthermore, the confidence levels and the performance levels are weakly correlated indicating that confidence levels do not affect the performance levels of this set of participants on the types of questions tested. Moreover, the machine learning algorithms utilized to classify the participants in this study showed potential based on their performance and confidence levels.

DOI:10.1561/116.00000041

Companion

APSIPA Transactions on Signal and Information Processing Special Issue - Information Processing for Understanding Human Attentional and Affective States: Articles Overview
See the other articles that are part of this special issue.