As part of our annual data review process, the American Community Survey (ACS) program calculates and publishes quality measures for the ACS data including response rates, coverage rates, item allocation rates, sample size and number of interviews. These quality measures are available on our Sample Size and Data Quality webpage for years 2000 to the present. In addition to informing data users about the overall quality of the data, these measures serve as input to our assessment of whether the data meet the Census Bureau’s Statistical Quality Standard F1: Releasing Information Products as described in the Statistical Quality Standards.

As part of addressing Requirement F1-7, the ACS assesses the item response rates for 17 key estimates to determine whether the item response rate for each is equal to 70 percent or higher. Refer to this user note for more information on how the characteristics were selected.

To calculate this metric, the ACS calculates the median of the ACS 5-year county-level item response rate for each characteristic and compares that median to the 70 percent threshold. We define, for our purposes, the item response rate as (100% minus the item allocation rate). Similar allocation rates for the ACS 1-year data product at the national and state-level can be found on our Sample Size and Data Quality webpage.

For data year 2022, the characteristic “percentage of households with income between $15,000 and $24,999” had an item response rate that fell below the 70 percent threshold. For this characteristic, we use the “Some or all income allocated” allocation rate when deriving the item response rate. Total income is not a direct question on the questionnaire but rather the sum of eight income components. Thus, for the standards, the derived allocation rate for total income is considered allocated (and hence a nonresponse) if any of the eight components of income are allocated.

Based on this, the 2022 ACS products fail the Statistical Quality Standards for this criterion. We are, however, continuing to publish the data under the waiver process described on the Statistical Quality Standards webpage.

We believe the data have value and that this decrease in the item response rate below the threshold represents only an incremental change from previous years. In addition, it should be noted that each of the eight components of total income easily surpass the item response rate set by the standards. Thus, the failure for total income to meet the 70 percent threshold is due in part to its construction where if even one of the eight components is allocated for an individual, regardless of the magnitude of its contribution to total income in dollars, then total income is considered allocated. This has the tendency to inflate this particular metric and contributes to it having a higher allocation rate, and thus a lower item response rate, compared to all other key characteristics.