Seemingly Unrelated Regressions (SURE) for Joint Covariate Balance
RCTClinical Trial
Testing experimental balance by examining one baseline variable at a time creates a multiple hypothesis testing problem, risking false positives and failing to evaluate whether baseline covariates are jointly uncorrelated with treatment assignment.
Picture this
Instead of checking a sports team's fair distribution by looking at height, age, and experience in separate independent tests, you evaluate all these features together in a single system to ensure no combination of traits creates a hidden advantage for one side.
What the evidence says
SURE joint hypothesis testing yielded p-values of 0.163 across the initial randomized sample, 0.282 for the state test sample, and 0.405 for subsequent peer characteristics, statistically confirming successful experimental balance across all baseline features simultaneously.
- Who was studied
- N = 31,439 randomized students in grades 4–8 and high school across 6 urban school districts.
- How
- Zellner’s Seemingly Unrelated Regressions (SURE) model stacking multiple baseline student characteristic regressions to test the joint null hypothesis of zero correlation with assigned teacher effectiveness.
What to do
Specify Zellner's SURE system estimation to execute joint hypothesis testing across all baseline covariates when validating randomization balance in multi-site RCTs.
From the source
"The joint hypothesis test was estimated using the 'seemingly unrelated regressions' or SURE model proposed by Zellner (1962)."
Have We Identified Effective Teachers? Validating Measures of Effective Teaching Using Random Assignment