Building an intuitive and friendly data collection framework to analyze human factors involved in software development

Abstract

Despite the most important aspect of software products being the people who design them, the research literature concerning the people in question is somewhat sparse and limited. The research that currently exists, while promising, only studies for small, controlled sample sizes. These studies are also condensed and focused down, often focusing on a very specific factor, at times in great detail. In this research, we analyze past literature concerning some of the interpersonal factors involved in software engineering. We specifically delve into the impact of team dynamics, personality, communication and skill level on the quality of the final product, often measured as a function of code quality, using a non-invasive testing framework. In contrast to other researchers who have used the popular, albeit controversial MBTI framework as a way to assess the impact of developer personality on the output of the software quality, we propose using data about the moment to moment emotional responses that occur during different stages of software development as a means of capturing this information. These data are captured from a survey that has a short enough response time and little or no technical jargon. We present and discuss our results to date from the survey, along with challenges that we faced and advice to future researchers in this area.