Using Contact History Information to Adjust for Nonresponse in the Current Population Survey

John S. Dixon


The Current Population Survey (CPS) adjusts the sampling weights for nonresponse to match population controls based on cells which combine similar primary sampling units (PSU) based on size and urbanicity. The adjustment method assumes that the nonresponse is random within the adjustment cells. This adjustment increases weights for responding units in PSUs with higher nonresponse. The present study uses information from the Contact History Instrument (CHI) to adjust the weights based on the patterns of responses interviewers experience in contacting and attempting to interview households. This paper explores whether this additional adjustment has the potential to reduce nonresponse bias.