Developing and evaluating a geographic information dashboard to improve spatial task performance

Abstract

Information dashboards are decision-support tools that pull data from multiple sources and display those data on a single screen. Information dashboards are becoming common in fields such as medicine, computer science, and business, given their perceived ability to facilitate faster and more accurate assessments by users. However, there is very little peer-reviewed research on information dashboards that support this assumption. This research focuses on applying the concept of an information dashboard visualization within the spatial sciences and evaluating the effectiveness of a geographic information dashboard, or GID, on improving user performance related to spatial thinking tasks. A review of literature from multiple disciplines highlights what is, and what is not, understood about dashboard visualizations. Borrowing from ideas such as Cognitive Fit Theory and past work in evaluating the effectiveness of map animations, an appropriate method for evaluating the GID is proposed. A Web-based GID and an alternative “tabbed” visualization were developed using the R Shiny package to support an analysis of grassland vegetation development for a site located in northeastern Kansas. A controlled experiment was conducted using a survey completed by volunteer student participants who responded to a series of benchmark tasks related to the interpretation of 6 related maps and graphs. Data for three dependent variables (task completion time, task response accuracy, and an integrative measure of performance accounting for both time and accuracy) were collected directly from the survey or post-survey grading of responses. Three independent variables and their impact on spatial task performance were analyzed, including the type of visualization, assessed spatial thinking ability, and cognitive task type. Results showed that participants using the GID completed the benchmark tasks faster and more accurately, but that a users’ spatial thinking ability had the most significant influence on performance regardless of visualization. Evidence was found to support the idea that the GID improved spatial thinking performance, especially for users with more experience in spatial reasoning, and that the GID format may improve user performance beyond what is expected based on an independent assessment of spatial thinking ability.