This study investigated the influence of visual representation of qualitative data at progressively greater levels of abstraction on decision-making processes in order to address a gap in research that currently focuses predominantly on the final choice phase of decision-making and representation of quantitative information. Specifically, this research investigated how four forms of data representation, varying progressively in their use of visualization and data abstraction, compare in the effort required to arrive at a decision, the ease with which themes are identified, satisfaction with the level of detail obtained, the confidence in decisions made, and the intuitiveness of representations. The experimental design closely simulated a real world decision-making scenario with a decision-making task developed in consultation with industry experts and a large qualitative dataset obtained from a survey on workplace environmental design conducted by a large global company. Qualitative data can be open to interpretation and final decisions or conclusions can be difficult to evaluate for accuracy. Therefore, decisions were compared across the four conditions to see how varying the level of visual abstraction of data representation encouraged participants to focus on certain themes versus others and to arrive at their decisions. The results indicate that in a matter of an hour, most participants identified key themes in the data regardless of the level of abstraction of the visual representation. However, there were significant differences in operational effort required, the intuitiveness of representations, and a marginaly significant difference in the ease with which themes were identified. Participants reported satisfaction with the level of detail reached and ratings of confidence in decisions-made were low or neutral and at odds with their objective performance. Insights into the effect of the visual representations and the subjective experience of the decision-maker are discussed.
展开▼
