Effective and economical: Cost analysis of the ASSISTments online math program

By Ashley Grant, Johns Hopkins University

With the growing interest in AI-supported and technology-focused interventions, there is a need for more information on the relative cost-effectiveness of interventions to help policymakers choose the ones that maximize student learning but remain within school budgets, which are often tight.

A West Ed study from Feng and colleagues analyzed the cost and impact of the ASSISTments program, an online, 1-to-1 homework support program for math which can be used in and out of the classroom to provide students with instant feedback and teachers with real-time progress reporting on their students. The original impact study (a randomized control trial of 5,912 7th grade students from 63 schools in 41 districts in North Carolina) showed a strong impact of ES=+0.10, which was sustained one year after the program ended.  

The current study builds on this evidence by estimating the cost required to produce these gains. Specifically, they estimated that it cost $46.23 per student for this 0.1 standard deviation increase in math achievement. This estimate (and cost-effectiveness ratio) is quite low compared to many other similar interventions and is more impressive given the rigorous research design that produced the estimated impacts (with randomization; large, diverse sample; and long-term impact estimate).

This study provides a guide for a quick application of cost-analysis methods, including a breakdown of each component’s contribution to the cost (Table 1). Additionally, the authors discuss the potential for alternate implementation approaches and how to integrate them into the cost-analysis.