3 ms·
There's too little info to go on in the article. If it is true, as they might possibly seem to imply, that they triggered more audits but didn't find cheating a
by vidarh 3y ago
There's too little info to go on in the article. If it is true, as they might possibly seem to imply, that they triggered more audits but didn't find cheating at an elevated rate over other subsets, then the first step to do would be to look at what assumptions were made about how EITC correlates with cheating or ucaught errors that'd justify an audit, and whether there's data to support those assumptions.