The AHS includes more than one-hundred questions covering a broad array of topics. In some instances, households refuse to respond to the entire survey. This is called unit nonresponse. HUD and the Census Bureau use a statistical technique called non-response adjustment to ensure that estimates produced from AHS data reflect both responding and nonresponding households. When households chose to respond to the AHS, they generally provide answers to all survey questions. However, for some households, the Census Bureau is unable to obtain a response to every survey question. This is called item nonresponse. To improve the usability of survey data, survey managers often “fill in” the missing data. This process is called imputation. Imputation improves the completeness and consistency of survey data and prepares the survey data for weighting and estimation processes. HUD and the Census Bureau use imputation to fill in missing responses in the AHS. However, it is important to note that missing responses are imputed in the AHS, but only for certain variables. In fact, most AHS variables with missing responses are not imputed. The purpose of this document is to explain the imputation strategies for item non-response in the American Housing Survey (AHS) for 2015 and later years.