Final Evaluation of 2020 Census Quality from Statisticians’ Task Force

Today, the American Statistical Association’s 2020 Census Quality Indicators Task Force issued a final evaluation of the 2020 Census state-level apportionment numbers. It examined 10 process statistics relating to the collection and post-data collection operations of the 2020 Census, including Master Address File (MAF) development, self-response, nonresponse follow-up (NRFU), data compilation and processing, and group quarters enumeration.

The task force concluded that:

  1. “Indicators released to date by the bureau do not permit a thorough assessment of the 2020 Census data quality.”
  2. “Despite concerns that census numbers could be jeopardized by political interference, the task force found no evidence of anything other than an independent and professional enumeration process by the Census Bureau. The bureau appropriately delayed release of data products to ensure careful review and processing of the data according to bureau quality standards.”
  3. “Across the limited set of state-level process statistics evaluated by the task force, it found no major anomalies that would indicate census numbers are not fit for use for purposes of apportionment.”
  4. “The task force’s ability to more thoroughly evaluate the quality, accuracy, and coverage of the 2020 Census has been hampered by limits on available information and research.”
  5. “The set of process statistics evaluated by the task force is relevant for evaluating the quality of census numbers for apportionment, but not redistricting or distribution of federal funds. These and other assessments require scrutiny at more detailed levels of geography and subgroups of the population.”

While commending the Census Bureau for working with the National Academies “and other experts to conduct a more thorough assessment of 2020 Census data quality,” the ASA task force recommended that the 2030 Census must “incorporate explicit attention to evaluating and reporting on data quality.” That will require investment of resources in advance “to enable the use of process statistics for evaluating data quality before apportionment counts are released for future censuses.” Census quality assessment measures usually come “long after the release of census data products, but this is no longer sufficient since process statistics and other data can be available much sooner.”

The ASA task force was formed in September 2020 in response to concerns about the quality of the 2020 Census data, after many data collection challenges in 2020.