Using Administrative Data to Evaluate Nonresponse Bias

From 2020 to 2024, official poverty rates with the alternative weights were slightly - yet statistically significantly - higher than the survey estimates, by 0.5, 0.5, 0.3, 0.3, and 0.4 percentage points respectively. These differences contrast with the pre-pandemic years 2017 through 2019, which showed no statistically significant differences in poverty rates between the alternatively weighted estimates and the survey estimates.³

To summarize, since 2020, survey nonresponse has continued to bias income statistics upward by 2% to 3% and official poverty rates downward by a fraction of a percentage point.

In ongoing research, the Census Bureau continues to develop and apply new methods that better address nonresponse bias and other sources of nonsampling error. For example, the National Experimental Wellbeing Statistics combine data from a wide variety of sources to produce improved estimates of the income distribution. One component of that work involves applying the alternative weighting scheme discussed above. Whether current trends in response rates continue, this research promises to address well-known challenges with our surveys and advance the Census Bureau's commitment to serve as the country's leading provider of quality data about its people and economy.

Definitions and information on confidentiality protection, sampling error, and nonsampling error are available at www2.census.gov/programs-surveys/cps/techdocs/cpsmar24.pdf.

###