February 13, 2001
For Immediate Release
Contacts: John Chambers
PORTLAND STUDY DEMONSTRATES THE HARMFUL EFFECTS OF CENSUS UNDERCOUNTS ON PUBLIC SCHOOL PLANNING
Findings Underscore Need for Corrected Census Data in 2001
Washington D.C. -- A study released today by the Presidential Members of the U.S. Census Monitoring Board found that school enrollment forecasts for the Portland Public School District, which had a relatively low undercount in 1990, was off by as many as 1,250 students for the decade ending in 2010 as a result of 1990 census undercount.
Public schools rely on enrollment rates for planning over a 15 to 20 year period. These rates help to determine the need for additional schools, where a school should be constructed, the appropriate size of a schools physical plant and the level of investment needed for school buses, and other purposes.
"The use of modern statistical methods in Census 2000 will lead us to the first accurate census in U.S. history and give school districts nationwide the reliable data they need to plan effectively," said Gilbert F. Casellas, Presidential Co-Chair of the Monitoring Board.
The study, authored by Dr. Barry Edmonston, Director of the Population Research Center at Portland State University, highlights the importance of accurate census data in ensuring that students K-12 are afforded the necessary resources to obtain the best possible education.
"If Portland, a comparatively well counted city, is adversely affected by a census undercount, then school districts across the nation, and especially those with larger undercounts, should be extremely concerned with these findings," said Edmonston.
Edmonston examined the effects of a correction for the undercount in the 1990 Census on the major factors affecting enrollment changes in the Portland School District -- births, migration and public school enrollment. The study can be viewed at http://www.cmbp.gov.
A corrected 1990 census would have added 1,400 school aged children to the count. Edmonston determined that a correction could have led to more accurate enrollment estimates over the last 10 years. "I found a much greater uncertainty in school planning forecasts than I ever imagined," said Edmonston.
"In Portland, 1,250 children is enough to fill two new grade schools," Edmonston added. "These findings make it crystal clear that a corrected census can improve school enrollment analysis and lead to better planning decisions."
The Census Bureau is scheduled to release block level data by age, race and sex for the 2000 census next month.