The Role of Cognitive Load and Individual Differences When Interpreting Human-Resource Data Visualizations

Abstract

Data visualizations (e.g., bar graph, dashboard) can be used as decision-support and storytelling tools that aid users’ interpretation of sometimes complex information, including within the human resource management (HRM) context. As HRM evolves towards implementing more data-informed decisions, it is important to understand how users interpret data visualizations. The aims of this thesis are to (a) identify whether cognitive load affects the amount of time users spend arriving forming and interpretation and the accuracy of their interpretations, and (b) to evaluate whether cognitive load moderates the association between individual-difference variables and interpretation time and accuracy. The individual differences that are of particular interest are locus of control and the personality dimensions of extraversion, neuroticism, openness to experience, and conscientiousness. A sample of 58 undergraduate business students were randomly assigned to three different cognitive load levels (control, moderate, high), and each participant -- irrespective of their group -- responded to the same four data-visualization vignettes. Hypotheses were tested using a moderated multiple linear regression model. None of the proposed hypotheses were supported in the initial analysis, although after further analyses, cognitive load was a strong moderator of the association between neuroticism and interpretation accuracy for participants who experienced a moderate level of cognitive load, such that the association was negative when cognitive load was moderate. Theoretical and practical implications are included for developers of these data visualizations to keep in mind.