Participatory Design of Visual Analytics Tools for Different Target Groups

Abstract

This paper reflects on the software engineering process behind the development of data visualization and analytics technologies tailored to the needs of diverse user groups. These considerations, introduced in our earlier work, are briefly revisited here. We focused on two use cases: one tailored to the needs and preferences of practitioners (data analysts), and the other directed towards meeting the requirements of nonprofessional, volunteer-based participants engaged in participatory citizen science. In both scenarios, we employed participatory methods, actively involving the target users in conceptualization and implementation phases. We observed diverse requirements and preferences concerning data visualization choices, additional functionalities, and analytical measures. To assess the effectiveness of these tools, in the current paper, we conducted a taskbased evaluation with selected participants, asking them to perform specific tasks such as identifying faults in the data, patterns, or detecting outliers. This was supplemented with qualitative feedback gathered through interviews and surveys, providing insights into user satisfaction, perceived challenges, and suggestions for improvement. The evaluation process revealed several areas for improvement from non-practitioners, particularly in the visual clarity of visualizations and the explanations regarding their usage, while practitioners responded more positively, noting no critical issues in software design and function.