A Case Study in Nonresponse: Plaintiff vs. California State Board of Equalization
展示如何利用协变量论证即使无应答率较高，无应答偏差也可能很小，该论证在一宗税务案件中帮助胜诉方。
In any sample survey, nonresponse bias is a potential issue. Even with a moderately high nonresponse rate, however, covariates can sometimes be used to show that the nonresponse bias is likely to be small. This note presents such an argument, which was used by the winning side in a tax case.