• The San Juan Daily Star

PR population overcounted by 5% in 2020 Census

U.S. Census Bureau Director Robert L. Santos

By The Star Staff

The U.S. Census Bureau on Tuesday announced that it overcounted Puerto Rico’s population by 5% as it released new estimates of net coverage error and components of coverage for Puerto Rico from the 2020 Post Enumeration Survey (PES).

The Census Bureau also overcounted the number of people in productive years and the number of owned housing units, according to a statement.

Those results provide insight into the quality of the 2020 Census counts for the population and housing units in Puerto Rico.

“The release of coverage estimates for Puerto Rico represents a final piece in completing our comprehensive assessments from the 2020 Post-Enumeration Survey,” Census Bureau Director Robert L. Santos said. “This is one of many self-assessments that allow us to think critically about how we transform our operations and plan for the 2030 Census.”

The numbers are important because they are used to determine things like the amount of federal aid to the island.

The PES was used to develop independent estimates of the number of people and housing units in the United States and Puerto Rico on April 1, 2020. Those estimates were compared to 2020 Census counts to explore the strengths and limitations of the data and to inform plans to improve the 2030 Census.

Puerto Rico is the largest and most populous U.S. territory and was treated as a separate governmental entity. As such, the sample was designed up front to allow for subpopulation estimates of coverage.

According to the PES, the 2020 Census overcounted the Puerto Rico population by 5.7%, or 174,000 people. That was not statistically different from the 4.5% overcount in the 2010 Census.

The estimated rate of erroneous enumerations in the 2020 Census was 9.8%, or 319,000. This includes an estimated 9.1%, or 294,000, duplicate records. The PES estimated an overcount of people in owned housing units by 8.6%. Renters did not have a statistically significant undercount or overcount.

The PES also estimated overcounts of the population for ages 30 to 49 years old by 3.6% and ages 50 years and older by 9.7%. The estimated net coverage error rates for the remaining age groups, including children, were not statistically significant. Adult males (5.7%) and females (7.3%) were statistically significantly overcounted in the 2020 Census. The PES also estimated statistical overcounts for 2020 Census counts of females ages 30 to 49 years (4.9%) and females 50 years and over (11%), as well as males 50 years and over (8.3%).

The PES estimated overcounts in the San Juan-Bayamón, PR Combined Statistical Area and in the rest of Puerto Rico. Because of the demographics in Puerto Rico, coverage estimates are not computed by race and Hispanic origin.

The PES did not estimate a statistically significant undercount or overcount of housing units in the 2020 Census in Puerto Rico. The 2010 Census also did not have a statistically significant net coverage error.

The 2020 Census undercounted the number of vacant housing units in Puerto Rico by 5.8%. There was no statistically significant overcount or undercount for occupied housing units.

“Understanding the devastation caused by Hurricane Maria, we adapted our operational design to count Puerto Rico as we prepared for the 2020 Census,” Santos said. “We will continue to collaborate closely with key stakeholders in Puerto Rico to work toward an accurate count for future censuses.”

112 views0 comments