1. What is item nonresponse?
Missing data for a particular question or item is called item nonresponse. It occurs when a respondent fails to provide an answer to a required item. The ACS also considers invalid answers as item nonresponse.
2. How does the ACS correct for item nonresponse?
The Census Bureau uses imputation methods that either use rules to determine acceptable answers or use answers from similar housing units or people who provided the item information. The first of these two methods is known as "assignment," while the second is referred to as "allocation."
Assignment involves logical imputation where a response to one question implies the value for a missing response to another question. For example, first name can often be used to assign a value to sex.
Allocation, on the other hand, involves using statistical procedures, such as within-household or nearest neighbor matrices populated by donors, to impute for missing values.
3. Why is it important to measure item nonresponse?
Item nonresponse measures allow data users to judge the completeness of the data in which the survey estimates are based. Final estimates can be adversely impacted when item nonresponse is high and bias can be introduced if the characteristics of the nonrespondents differ from those reported by respondents. Item nonresponse and unit nonresponse both contribute to potential bias in the estimates.
4. How does the ACS measure item nonresponse?
Item nonresponse is measured through the calculation of allocation rates which are published with the survey estimates. The Census Bureau calculates measures of item nonresponse for two distinct universes. The American Factfinder (AFF) includes allocation tables specific to the tabulation universes. This Quality Measures Web page includes allocation rates for the universe that was eligible for editing and imputation. In some instances these will be the same, but in many instances they will differ. For example, we edit and impute data collected for educational attainment for the total population 3 years and over, so that is the universe referenced to calculate the allocation rates shown on the Quality Measures Web page. However, the tables for educational attainment in the AFF are restricted to the population age 25+ and therefore the imputation tables on AFF are restricted to this universe. The specific universe associated with each of these Quality Measures are shown in the tables, displayed below the title of each item.
The 2000-2005 item allocation rates are for the housing unit population only. The item allocation rates for 2006 and thereafter are for the total population which includes the housing unit and group quarters populations.
5. How are item allocation rates calculated?
Allocation rate for item A (state x, year y) =