Suchergebnisse

Measurement errors are often a large source of bias in survey data. Lack of knowledge of the determinants of such errors makes it difficult for data producers to reduce the extent of errors and for data users to assess the validity of analyses using the data. We study the determinants of reporting error using high quality administrative data on government transfers linked to three major U.S. surveys. Our results support several theories of misreporting: Errors are related to event recall, forward and backward telescoping, salience of receipt, the stigma of reporting participation in welfare programs and respondent's degree of cooperation with the survey overall. We provide evidence on how survey design choices affect reporting errors. Our findings help survey users to gauge the reliability of their data and to devise estimation strategies that can correct for systematic errors, such as instrumental variable approaches. Understanding survey errors allows survey producers to reduce them by improving survey design. Our results indicate that survey producers should take into account that higher response rates as well as collecting more detailed information may have negative effects on survey accuracy.

We document the extent, nature, and consequences of survey errors for receipt of cash welfare and SNAP in three major U.S. household surveys linked to administrative program records. Our results confirm high rates of misreporting of program receipt, particularly failure to report receipt. The surveys inaccurately capture patterns of participation in multiple programs, even though there is little evidence of program confusion. Error rates are higher among imputed observations, which also account for a large share of false positive errors. Many household characteristics have significant effects on errors in reporting receipt, both false positives and false negatives. We find large differences in survey errors by race, ethnicity, income and other household characteristics. We provide evidence on the consequences of these errors for models of program receipt. Estimated effects of income and race are noticeably biased. We then examine error due to item non-response and imputation, as well as whether imputation improves estimates. Item non-respondents have higher receipt rates than the population, even conditional on many covariates. The assumptions for consistent estimates in multivariate models fail both when excluding item non-respondents and when using the imputed values. In binary choice models of program receipt, estimates from the linked data favor excluding item non-respondents rather than using their imputed values. The biases in each case are well predicted by the error patterns we document, so such analyses can help researchers make more informed decisions on the use of imputed values.