The late-pretest estimator preference holds as long as test score impacts do not grow very quickly early in the school year¶

Claim · All claims
Evidence · 1 study · theoretical r2 · q2
Currency · newest study 2008 · checked against its sources 2026-10-09 (automated, 1 of 1 entry) · not reviewed by a person

Subclaims¶

q2 i? The preference for late-pretest-inclusive estimators holds as long as test score impacts do not grow very quickly early in the school year. → Peter Z. Schochet 2008

Evidence¶

Peter Z. Schochet 2008¶

Peter Z. Schochet. (2008). The Late Pretest Problem in Randomized Control Trials of Education Interventions. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education. https://ies.ed.gov/ncee

q2 · i? · theoretical · r2

Boundary condition stated by the paper for its estimator-preference result: the recommendation is qualified by the growth rate of "test score impacts" early in the school year. No effect size or threshold value is printed.

"This result holds as long as test score impacts do not grow very quickly early in the school year."

Discussion¶

Related Claims¶

• The variance-bias trade-off for late pretests is addressed both theoretically and empirically for several commonly used impact estimators — a broader claim this one bears on
• In RCTs of interventions to improve student test scores, estimators including late pretests will typically be preferred to estimators excluding them or using other baseline data — a broader claim this one bears on
• Including late pretest data in RCT analysis could bias post-test impact estimates when pretests are collected after random assignment — related
• Deciding whether to collect and use late pretest data in RCTs involves a variance-bias trade-off — related
• Empirical examples using real-world education RCT data demonstrate the design-based grouped-data theory — related