Design effects from aggregation are tolerable with moderate numbers of groups and few covariates¶

Claim · All claims
Evidence · 1 study · theoretical r2 · q2
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Design effects from aggregating individual data to groups are tolerable when there are moderate numbers of groups and few covariates, suggesting little information is lost in these cases. → Peter Z. Schochet 2020

Evidence¶

Peter Z. Schochet 2020¶

Peter Z. Schochet. (2020). Analyzing Grouped Administrative Data for RCTs Using Design-Based Methods. Journal of Educational and Behavioral Statistics, vol. 45, issue 1. https://www.mathematica.org/publications/analyzing-grouped-administrative-data-for-rcts-using-design-based-methods

q2 · i? · theoretical · r2

The article reports that "design effects from aggregation are tolerable with moderate numbers of groups and few covariates," indicating "little information is lost in these cases." No numeric design-effect values are printed in the available text.

"Furthermore, design effects from aggregation are tolerable with moderate numbers of groups and few covariates, suggesting little information is lost in these cases."

Discussion¶

Related Claims¶

• Ecological inference methods for subgroup analyses yield large design effects — reports the opposite
• Test-taking disengagement distorts both individual and aggregated test scores — related
• School-level aggregated scores on the computer-based PISA-Based Test for Schools were highly stable despite considerable differential engagement among schools — related
• E-M scoring leaves residual score distortion, suggesting some disengaged test responses are nonrapid rather than rapid guesses — related
• Responses used in effort-moderated scoring can sometimes reflect less-than-full, or partial, engagement, making E-M scores less trustworthy — related