Rigorous Data Validation for Accurate Dashboards: Experience From a Higher Education Institution

Abstract

Data have become an indispensable aspect of our daily lives. The demand for data visualization tools such as dashboards is driven by the desire to make data—and more importantly, the power of using data to inform decision making—accessible to all. During the current pandemic, the availability of realtime data via various dashboards at the global, national, and local levels empowered many to accurately assess the situation and take appropriate actions, a testament to the value of data visualization. For the same reasons, data dashboards are increasingly popular in higher education to promote data consumption and data-driven decision making. California State University, Fullerton (CSUF) is no exception. CSUF has developed a suite of dashboards in the past few years to promote an operational culture that is rooted in evidence. Discussions about data visualization tools often gravitate towards dashboard design, accessibility, and usability, while neglecting a fundamental (and arguably more critical) issue—the importance of having appropriate and accurate underlying data. The accuracy and adaptability of a dashboard are determined by the accuracy and adaptability of the data behind it, and to ensure such requires a meticulous, streamlined development process. This article is intended to do a “deep dive” into this process.