Interaction Patterns: A Visual Approach to Mixed Methods Research

In this paper we describe an inventory of visual interaction patterns – repeatable combinations of visualizations and accompanying questions aimed at extending quantitative studies with qualitative data. More precisely, a visual interaction pattern is a systematic approach by which quantitative research results are visualized, shown to the research participants and interpreted together with the research participants. After a quantitative research phase (like a survey, experiment or card sorting), we produce visualizations (bar charts, matrices, etc.) which display the quantitative results. We then show these visualizations to our research participants in follow-up group conversations and ask questions in order to spur an exchange of interpretations and insights. We involve our research participants in the data interpretation process. We use the visualizations as centers around which we facilitate the group conversations by referring to what the visualizations are “saying” and asking what this means in the eyes of our research participants. The visual interaction patterns are elicitation mechanisms which enable researchers and research participants to jointly produce value-adding research insights. For example, the disagreement pattern provides value-adding insights by visualizing the spread of opinions and using constructive conflict as a way of collective thinking. The outlier pattern visualizes extreme dissenting views and gives voice to them. The benefits of applying the patterns include improving the correspondence of correlational models with empirical data, providing explanations for unexpected findings and alternative explanations for expected findings, enabling the formulation of post-hoc hypotheses based on the newly-discovered correlational dependencies, and providing feedback on the research procedure. The involvement of research participants in the data interpretation process enables a critical appraisal and enrichment of quantitative results. Insights extending (or challenging) the quantitative results are gathered, which leads to confirmed, revised or extended correlational models and revised or extended research procedures. These value-adding research outcomes are important for management research because they contribute to theory development, they create ideas on how to solve work-related problems, identify industry-specific dependent and independent variables, or foresee future management trends and developments.