Survey methodologists are searching for covariates to use in nonresponse adjustment models, ultimately hoping to find variables that are highly correlated with both the outcomes of interest and the propensity to respond. These covariates can come from auxiliary data that provide information on both respondents and nonrespondents. Two such types of auxiliary data are interviewer observations (a form of paradata) and commercially available data on small areas or households. Interviewer observations intended for use in nonresponse adjustment can be specifically designed to match the outcome variables of interest, while commercial data provide a broad set of small area descriptors that may be correlated with multiple outcomes. This analysis examines these two data sources to determine which is more predictive of the outcomes of interest for a particular survey, thereby fulfilling one of the criteria for a good adjustment variable. The outcomes of interest in this analysis are self-reports of household income and receipt of unemployment benefits from a survey of labor market participation. The findings suggest that at this point in time, compared to commercial data, interviewer observations are better at predicting these outcomes, particularly in the subpopulation that the survey targets. Therefore, the observations share more (accurate) information with the true value, making them better for adjustment on this dimension. The results will inform the work of both researchers wishing to improve their nonresponse adjustments and survey managers looking to make better use of their survey budget.
