Nonrespondents and Nonresponse Bias: Evidence from a Survey of Former Welfare Recipients in Iowa

Examines whether differences between respondents and nonrespondents in a survey of Iowa families that left TANF resulted in findings that overstated the economic well-being of these families. Finds that nonrespondents were less likely to have health insurance and more likely to have housing-related problems than respondents. Concludes that the nonresponse bias was small despite differences between respondents and nonrespondents, and that findings from the study would not have been much different if the response rate (76 ) had been considerably higher.