2021 Articles

Supporting the Initial Work of Evidence-Based Improvement Cycles Through a Data-Intensive Partnership

Bowers, Alex J.; Krumm, Andrew

Purpose: Currently in the education data use literature there is a lack of research and examples that consider the early steps of filtering, organizing, and visualizing data to inform decision making. The purpose of this study is to describe how school leaders and researchers visualized and jointly made sense of data from a common learning management system (LMS) used by students across multiple schools and grades in a charter management organization operating in the United States. To make sense of LMS data, researchers and practitioners formed a partnership to organize complex data sets, create data visualizations, and engage in joint sensemaking around data visualizations to begin to launch continuous improvement cycles.

Design: We analyzed LMS data for n=476 students in Algebra I using hierarchical cluster analysis heatmaps. We also engaged in a qualitative case study that examined the ways in which school leaders made sense of the data visualization to inform improvement efforts.

Findings: The outcome of this study is a framework for informing evidence-based improvement cycles using large, complex datasets. Central to moving through the various steps in the proposed framework are collaborations between researchers and practitioners who each bring expertise that is necessary for organizing, filtering, and visualizing data from digital learning environments and administrative data systems.

Originality: We propose an integrated cycle of data use in schools that builds on collaborations between researchers and school leaders to inform evidence-based improvement cycles.