APRIL 26, 2021 — Today the U.S. Census Bureau released information about the quality of the 2020 Census from two methods: (1) analyses that compare the first census results to other ways of measuring the population, and (2) metrics that provide insight into the census operations.
The Census Bureau accelerated plans for releasing this information because of increased interest in how COVID-19 and its impact on operations may have impacted the quality of the 2020 Census. This is the first time that the Census Bureau has released such detailed analyses and operational quality metrics on the same day as the first population counts.
“Despite all the challenges of the pandemic, the completeness and accuracy of these first 2020 Census results are comparable with recent censuses,” said Census Bureau Acting Director Ron Jarmin. “We had numerous quality checks built into collecting the data, and we have conducted one of the most comprehensive reviews in recent census history during data processing. We are confident that today’s 2020 Census results meet our high data quality standards.”
The report A Preliminary Analysis of U.S. and State-Level Results From the 2020 Census compares the 2020 Census population counts to three other population benchmarks — previous decennial census results, estimates from the Census Bureau’s Population Estimates Program, and the 2020 Demographic Analysis estimates.
The report analyzes the differences between the 2020 Census counts and each of the benchmarks. The analyses show that these first results from the 2020 Census are in line with the three sets of benchmarks:
The findings are also discussed in the Comparisons to Benchmarks as a Measure of Quality blog.
The operational metrics released today include information on how people responded to the 2020 Census, as well as how the Census Bureau accounted for addresses that did not respond. These metrics add to the unprecedented number of data quality indicators the Census Bureau has already provided on 2020 Census data collection.
These metrics show by state the percentage of addresses in the 2020 Census that the Census Bureau determined were occupied, vacant or nonexistent. That status was determined by:
The operational metrics also show how the Census Bureau resolved households through the Nonresponse Followup operation, such as through an interview with a member of the household, or a proxy, or through the use of administrative records. Finally, the metrics provide the percent of vacant addresses and the percent of deleted (i.e., nonexistent) addresses resolved with a proxy interview or by using administrative records, respectively.
The dashboard and the downloadable table provide comparable metrics from the 2010 Census. It is important to note there will be differences between the two decades across geographies. Many differences result from changes in how the 2020 Census was conducted compared to the 2010 Census. For example, the 2020 Census invited people to respond online.
While no single number can definitively quantify the quality of the census, looking at these metrics collectively and comparing them with past censuses provides an important perspective to consider. Preliminary analyses of the operational metrics are available in the Examining Operational Quality Metrics blog.
The Census Bureau will continue to evaluate the 2020 Census and release additional quality metrics and assessments in the months and years to come. The Census Bureau will use these assessments to prepare and innovate for the 2030 Census.
For more information, visit our 2020 Census Quality and Data Processing Press Kit with materials from today’s data release.