Tracking Students’ Academic Progress in Data-Rich but Analytically Poor Environments

• In Event: Longitudinal Data Systems and Tracking Academic and Nonacademic Outcomes

Abstract

This study validates a data-driven algorithmic procedure for identifying students at-risk of academic failure through multi-cohort, school, and district longitudinal comparative studies. Results indicate it is possible and feasible using readily available school SIS data to identify students who appear at substantial academic risk, what some risks are, and who may benefit from specific academic support interventions. Today's school environment is "data-rich." Even relatively simple analyses (dashboard profiles predictive of individual success, case studies of particular interventions, summaries of service utilization, climate assessments) can benefit educators and students. This is the essence of data-driven decision-making. Our schools have capability and staff talent to produce these sorts of analyses, but seldom time or resources to develop such capabilities on their own.

Authors

• Warren E. Lacefield, Academic Software, Inc.

• Brooks Applegate, Western Michigan University