STANDARD DEVIATIONS: New Report Discusses Implications of the Census Bureau’s Blended Base

By Dr. William P. O’Hare, President, O’Hare Data and Demographic Services LLC
Standard Deviations blog posts represent the views of the author/organization, but not necessarily those of The Census Project.

Since the 1970s, the Census Bureau has produced yearly population estimates for states and counties for the decade following each Decennial Census. The estimation method used by the Census Bureau starts with a population base and then adds or subtracts estimated yearly incremental change to that base.

In the past, the Decennial Census counts have provided the estimates base, but the 2020 Census detailed data needed for the base was not available in time to use with the 2021 and 2022 population estimates, so the Census Bureau staff developed a new methodology called the PEP (Population Estimates Program) blended base.

It is important to assess the implications of the PEP blended base for children, because children (ages 0 to 17) had a 2.1 percent undercount in the 2020 Census compared to a 0.25 percent overcount for adults. A new paper provides guidance for child advocates, researchers, and data analysts on the potential impact of the Census Bureau’s new PEP blended base methodology for the child population (ages 0 to 17) by looking at how the data from the blended base compares to the data from the 2020 Census for the population ages 0 to 17 (“Comparing the Accuracy of the 2020 Census Counts to Population Estimates Program Blended Base for Age Groups of Children.”)

Key results are shown below:

  • At the national level, the number of children for April 1, 2020 from the PEP blended base was 74,385,212 compared to 73,106,000 in the 2020 Census. 
  • The PEP blended base estimates were larger than the 2020 Census counts in nearly every state (48 out of 50 states and DC). 
  • The 2020 Census count is larger than the PEP blended base in 1,234 counties. On the other hand, there are 1,903 counties where the PEP blended base provided a larger number of children than the 2020 Census count.
  • In terms of the national share of children in a state, there are 35 states where the Census count is larger than the PEP blended base estimate in terms of national shares. 
  • When national shares were examined, the 2020 Census count was larger than the PEP blended base in half (50 percent) of all counties.

The findings suggest that assessing the impact of the PEP blended base on the child population will be complicated. Patterns are different depending on whether one examines absolute numbers of children or the national share of children. Also, there are important differences by level of geography (national, state or county).

  • Dr. O’Hare, a member of The Census Project Advisory Committee, is a social demographer who has spent forty years using data to increase public understanding of disadvantaged groups. For the past 25 years, he has been involved in the KIDS COUNT project at the Annie E. Casey Foundation. Bill has a Bachelor of Science, Master of Arts, and a Ph.D. from Michigan State University.