Use process and usage data alongside impact data in edtech evaluations¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article calls for 'evaluations that make greater use of process, implementation, and usage data as well as measures of impact'. Digital platforms automatically collect copious data on learning processes and usage; process measures can supply behavioral evidence that a presumed learning process, such as surfacing and addressing misconceptions, actually occurs. The strategy reframes the evaluation question: what can be well-implemented and well-used matters as much as what works.

Design Implications¶

Context¶

Requirements¶

• Usage data should be extended to center equity — whether usage is reaching all students and target groups

Constraints¶

• Traditional evaluation relies on distal outcomes; end-of-year tests may be administered many months after the intervention

Target Learners¶

• K-12 students using digital learning platforms
• district evaluators

Target Learning Goals¶

• Validating logic models, guiding implementation, and monitoring equitable usage

Related Strategies¶

• Broaden measured outcomes beyond high-stakes assessments to include durable skills
• Invest in technological infrastructure for evidence generation, including synthetic data, shared platforms, and privacy-enhancing technologies
• Evaluate edtech investments by Value of Investment alongside Total Cost of Ownership

Examples¶

-

Key Sources¶

• Roschelle, J., Ruiz, P., Means, B., Griffiths, R., Lee, K. (2026 April). Towards an Evidence Framework for Emerging Educational Technologies. Digital Promise. https://doi.org/10.51388/20.500.12265/289