This assessment presents the item nonresponse and imputation rates for household-level and person-level items from the 2020 Census. It also compares the 2020 item nonresponse and imputation rates to the 2010 rates.

Item nonresponse and imputation rates are response data quality metrics. Item nonresponse occurs when a respondent provides some information but does not respond to all census questions. Item nonresponse rates are calculated before preediting or characteristic imputation and do not take into consideration the validity of a response. Imputation is a statistical technique used to fill in missing information and considers both respondent cooperation and response validity.

The items included in this assessment are: population count, undercount, tenure, telephone number, relationship, sex, age/date of birth, Hispanic origin, race, and overcount. A “population count only” metric is also calculated, which refers to the respondent provides the household population count but does not provide responses for the household-level items nor the person-level items for any person on the roster. This quality metric measures which housing units provided minimal data.

This assessment will not address the quality of the imputation process. It will only measure how much imputation was done. Characteristic imputation, applied after the household population is established, is in scope for this assessment, but count imputation is not in scope.