Attributes Contributing to Students' Use of Quality Software Development Practices

Abstract

In 2001 the "McCracken group", through a multi-institutional study, concluded that many students finishing their introductory programming courses could not program due to a lack of problem solving skills. In 2004 Lister established that students have a fragile grasp of skills to read and interpret code. Humphrey suggests that educators must shift their focus from the programs that the students create to the data of the processes the students use. This paper addresses the problem of poor performing students through an investigation of their quality appraisal techniques (QATs) and development processes. Firstly, a survey was conducted to determine the current software development practices used by a group of undergraduate Computer Science students. Numeric data collected revealed that the current practices used by the majority of students would not be sufficient to produce quality programs. Secondly, a case study was conducted to gain a deeper understanding of the various factors that are likely to influence students’ intention to use QATs. Analysis of numeric data collected through a survey revealed that students' intentions to use QATs are driven by ease of use, compatibility, usefulness, result demonstrability, subjective norm and career consequences. Thirdly, an experiment was conducted to determine students' perceptions on the use of process measurement data to improve their current software development practices. Analysis of numeric and narrative data revealed that performance measurement data could provide students with useful information to adopt proper development practices.