An education research concept defining methods used to evaluate interventions, programs, and policy impacts. It governs study design, measurement, and interpretation practices used to estimate effects and assess implementation quality. It does not establish causation without appropriate design choices and careful handling of bias and uncertainty. It supports improvement by identifying what works and under what implementation conditions. The concept is generally stable, though methods and reporting standards evolve over time.

Principle

Detect and correct errors, document data provenance and transformation, and ensure analytic assumptions are supported by the empirical record before conclusions are drawn.

Demonstration

A data review identifies duplicated participant IDs, verifies score ranges against codebooks, reconciles missingness patterns with enrollment logs, and documents imputation decisions prior to statistical modeling.

Misapplication

Treating superficial frequency checks as sufficient or conducting review after analysis (post hoc) so that discovered issues invalidate prior results instead of being prevented.

Consequence

A rigorous data review increases confidence in analytic results, reduces rework, prevents misleading conclusions, and streamlines transparent reporting to stakeholders.

Reversal

Skipping formal data review and relying on raw exports can speed initial analysis but increases risk of bias, analytic error, and the need for corrective reanalysis.

Boundary

Concerns the quality and fitness-for-purpose of data for analysis and reporting; it does not itself perform causal estimation or substitute for peer methodological review of analytic strategy.

Semantic Tension

Tension between rapid preliminary checks for operational decisions and deep audits for publication-grade analysis; the former sacrifices depth for speed, the latter demands thorough documentation.

Synthesis

Program Evaluation Data Review is a quality-assurance stage that verifies data integrity and documents decisions so analysis rests on a transparent, defensible empirical foundation.