July 2, 2001

For Immediate Release

Contacts: John Chambers
301-713-6672

MONITORING BOARD FILLS VOID OF CENSUS BUREAU SILENCE

Census Undercount Estimates For The Nation's Largest Counties Released

Washington D.C. - Today, the Presidential Members of the U.S. Census Monitoring Board released estimates of the number of people missed in the largest U.S. counties.

Links to the charts:
Main Chart | Chart By Percentage of People Missed | Chart By Number of People Missed

Recognizing that the raw population count missed 6.4 million people -- mostly Black, Hispanic, Asian-Pacific-Islander, urban and rural poor and children -- Dr. Eugene P. Ericksen, a decennial census expert and professor of statistics at Temple University, calculated undercount estimates for all counties with populations over 500,000 and for the remainders of the states in which they are located. (Visit www.cmbp.gov for state undercount data.)

"Taxpayers paid more than six-and-a half billion dollars for the census and they deserve to see all the census results." said Gilbert F. Casellas, Presidential Co-Chair of the Monitoring Board. "My concern is that more than 6.4 million people were missed. If the Census Bureau doesn't see the importance of this, it's our duty to provide Congress and the American people with our own expert analysis."

The 6.4 million undercount estimate was calculated by experts at the Census Bureau using the results of the Accuracy and Coverage Evaluation (A.C.E.), a scientific quality control check to correct for those excluded in the head count. This more accurate estimate of the population (especially for larger geographic areas) could be of significant use to state and local governments for planning decisions and funds distribution.

Recently, however, the Census Bureau has refused to abide by congressional statutes and Freedom of Information laws requesting access to census data. Their strong stance against transparency, peer review, public accountability, scientific scrutiny and congressional oversight, has led to litigation by state and local governments and an outcry by many in the scientific and non-profit communities.

"People at the state and county levels have been demanding release of these figures, but their requests have fallen on deaf ears," added Casellas.

Today's release of undercount estimates is consistent with the Monitoring Board's statutory mandate to monitor Census 2000 operations and report to Congress. With less than two months before the Board issues its final report to Congress, the Board hopes these estimates will assist the scientific community and the public at large, to better understand the accuracy of Census 2000 and to prepare for Census 2010.

The bipartisan Census Monitoring Board was established in 1997 to monitor Census 2000 operations. Its findings are reported to Congress every six months. For further information, visit www.cmbp.gov.

Statement on Methodology from Dr. Eugene P. Ericksen:

In contrast to the parallel estimates for states, which equal the estimates calculated by the Census Bureau, the county estimates are reasonable approximations to the Census Bureau calculations. They were calculated to provide some insight into within-state variations in undercount rates.

Our method was to estimate each county's population subdivided into categories defined by race, Hispanic origin, renter/owner status, and metropolitan location. These categories match the major post strata defined by the Census Bureau when calculating undercount estimates. We applied the undercount adjustment factors calculated by the Census Bureau to the county population distributions. To illustrate with a simple example, if half the population was non-Hispanic Black or Hispanic, for whom the undercount rate was 2 percent, and the other half was non-Hispanic White, for whom the undercount rate was 1 percent, the overall estimated county undercount rate would be 1.5 percent.

These estimates should be close, but not exactly equal, to the numbers that the Census Bureau would calculate. Our calculations make no claim to being "official Census Bureau numbers." We also point out that the Census Bureau is currently reevaluating its estimates of the undercount, and they may revise the estimates at a future date.

Our county estimates, like the state estimates, and like the national estimates calculated by the Census Bureau, apply to the non-group quarters population. Like the Census Bureau, we make no statements regarding the possible undercounts of the group quarters populations.

Links to the charts:
Main Chart | Chart By Percentage of People Missed | Chart By Number of People Missed

###