The Influence of Human Factors on Programming Performance: Personality, Programming Styles and Programming Attitudes
Type of DegreePhD Dissertation
Computer Science and Software Engineering
MetadataShow full item record
The programming performance has been studied over several years. Researchers and scientists utilized various optimization technologies on algorithms and computer architectures to improve the performance. But, to date, few studies focus on the impact of human factors on the programming performance. In this study, we investigate the influence of human factors on the programming performance based on Mayer-Briggs Type Indicator (MBTI) personality, programming attitude and programming styles. Although some researchers have investigated the effects of personality based on the Five-Factor model on programming styles, two problems are not resolved: (1) Five-factor personality model does not theorize what goes inside people’s heads and focuses on actual people’s behaviors instead of the cognitive theory; (2) the programming styles were not validated and are out of date. To improve this research work, a theoretical personality model-- Myers–Briggs Type Indicator – is adopted. In addition, the programming styles have been updated since 2006 and validated using statistical metrics such as Cronbach’s Alpha. Finally, a new programming factor-- programming styles-- are added into our investigation. The objective of this proposal is: (1) to identify which human factors play a positive/negative role in programming performance; (2) to study the relationship among personality, programming styles and programming attitudes. The author firstly distributes three questionnaires on personality, programming attitudes and programming styles to students in department of computer science and software engineering at Auburn University. Three surveys towards programming will be measured via the self-assessed method. The programming performance consists of: (1) run time from participants’ code; (2) grades of projects. The analysis, such as Pearson Correlation analysis and linear regression analysis, will be applied to investigate the links among personality, programming styles and programming attitudes.