Requirements Elicitation in Government Projects: A Preliminary Empirical Study
Anqi Ren, Lin Liu, Yi Wang, Xiao Liu, Hailong Wang, Kaijia Xu, Xishuo Zhang, Chetan Arora
TL;DR
The paper addresses the gap in empirical requirements elicitation (RE) studies for government software by conducting semi-structured interviews with 12 practitioners involved in government data visualization projects in Inner Mongolia, China. It identifies key differences between government and private-sector RE, highlights dominant human and social factors (culture, team dynamics, policy implications), and outlines challenges such as ambiguity, changing requirements, and data-security concerns. The findings emphasize policy-driven and stakeholder-oligarchic dynamics, plus regional disparities, and suggest directions for extending RE research to broader technologies and regions. Overall, the work lays groundwork for more human-centric RE practices in government software and proposes future studies to bridge identified gaps.
Abstract
Government development projects vary significantly from private sector initiatives in scope, stakeholder complexity, and regulatory requirements. There is a lack of empirical studies focusing on requirements engineering (RE) activities specifically for government projects. We addressed this gap by conducting a series of semi-structured interviews with 12 professional software practitioners working on government projects. These interviewees are employed by two types of companies, each serving different government departments. Our findings uncover differences in the requirements elicitation phase between government projects, particularly for data visualization aspects, and other software projects, such as stakeholders and policy requirements. Additionally, we explore the coverage of human and social aspects in requirements elicitation, finding that culture, team dynamics, and policy implications are critical considerations. Our findings also pinpoint the main challenges encountered during the requirements elicitation phase for government projects. Our findings highlight future research work that is important to bridge the gap in RE activities for government software projects.