Ten-RCT empirical comparison of CACE variance formulas¶

Element · All elements
Evidence · no claims cited

Description¶

An empirical analysis component of the report in which the authors "used data from ten large-scale RCTs" to compare significance findings for the CACE estimator. The comparison was conducted with the estimator "measured in nominal and standard deviation units", contrasting the correct variance formulas with those typically used in practice, and it grounds the report's conclusion that variance correction terms matter little.

Design Implications¶

Context¶

Requirements¶

• Requires outcome and compliance data from large-scale clustered education RCTs

Constraints¶

• The comparison concerns significance findings for the CACE estimator, not effect magnitude estimation

Target Learners¶

• education researchers conducting RCT analyses

Target Learning Goals¶

• understanding when CACE variance corrections affect significance findings

Claims¶

Related Elements¶

• Set of 10 large-scale education RCTs used to compare CACE variance formula significance findings
• Empirical example using data from a multi-armed education RCT

Examples¶

-

Key Sources¶

• Peter Z. Schochet, Hanley Chiang. (2009). Technical Methods Report: Estimation and Identification of the Complier Average Causal Effect Parameter in Education RCTs. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education. https://www.mathematica.org/publications/technical-methods-report-estimation-and-identification-of-the-complier-average-causal-effect-parameter-in-education-rcts