The Emotional Roller Coaster of Responding to Requirements Changes in Software Engineering
Kashumi Madampe, Rashina Hoda, John Grundy
Abstract
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Background:</i> A preliminary study we conducted showed that software practitioners respond to requirements changes (RCs) with different emotions, and that their emotions vary at stages of the RC handling life cycle, such as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">receiving, developing,</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">delivering</i> RCs. Furthermore, such developer emotions have direct linkages to cognition, productivity, and decision making. Therefore, it is important to gain a comprehensive understanding the role of emotions in a critical scenarios like handling RCs. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Objective:</i> We wanted to study how practitioners <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">emotionally</i> respond to RCs. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Method:</i> We conducted a world-wide survey with the participation of 201 software practitioners. In our survey, we used the Job-related Affective Well-being Scale (JAWS) and open-ended questions to capture participants’ emotions when handling RCs in their work and query about the different circumstances when they feel these emotions. We used a combined approach of statistical analysis, JAWS, and Socio-Technical Grounded Theory (STGT) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">for Data Analysis</i> to analyse our survey data. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Findings:</i> We identified (1) emotional responses to RCs, i.e., the most common <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">emotions</i> felt by practitioners when handling RCs; (2) different <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">stimuli</i> – such as the RC, the practitioner, team, manager, customer – that trigger these emotions through their own different characteristics; (3) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">emotion dynamics</i> , i.e., the changes in emotions during the RC handling life cycle; (4) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RC stages</i> where particular emotions are triggered; and (5) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">time related aspects</i> that regulate the emotion dynamics. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Conclusion:</i> Practitioners are not pleased with receiving RCs all the time. Last minute RCs introduced closer to a deadline especially violate emotional well-being of practitioners. We present some practical recommendations for practitioners to follow, including a dual-purpose emotion-centric decision guide to help decide when to introduce or accept an RC, and some future key research directions.