March 29, 2000
For Immediate Release
Contact: Christina Horzepa, 212-339-0520
The US Decennial Census: Political Questions, Scientific Answers
The US census--now underway--has been the target of sharp partisan attacks over methodology. The current controversy is misguided, says Census Bureau Director Kenneth Prewitt. In the latest issue of Population and Development Review Prewitt traces the history of the census and explains why neither the traditions nor the competencies of the Census Bureau are consistent with advancing a partisan agenda.
It is no surprise that there is a partisan edge to the focus on census numbers, Prewitt says, noting that the allocation of more than $2 trillion in federal funds--plus many millions in state and local spending--will be affected by Census 2000 counts. However, accusations that the nonpartisan, professionally managed Census Bureau might choose a data-collection methodology so as to favor one political party over another are deeply disturbing, Prewitt says.
The decennial census was established in 1787 to serve the political purpose of allocating seats in the House of Representatives. After World War II the purposes to which census data are put expanded to include allocation of federal funds and election redistricting. "Political leaders now argue that the census can be designed to predetermine the apportionment outcome in a way that favors one or the other of the political parties. This invites partisan disagreement not just about how the census count should be interpreted or applied, but how the census should be designed," Prewitt notes.
No census has been able to account for every resident. A combination of citizen resistance and flaws in the enumeration procedure results in an underestimation of the population. Insofar as the benefits of the census are allocated on a share basis--as is true if the benefit is a fixed number of congressional seats or a fixed amount of federal funds--an undercount distributed equally across geographic units and population groups does not result in inequitable outcomes. Inequity or injustice emerges only if some areas or groups are counted at lower rates than others: that is, if the undercount is differential, Prewitt explains.
After World War II, demographers uncovered the first systematic evidence of a differential undercount when they compared statistics from the selective service draft registration to census figures and noticed that African-American males of draft age had been missed in the census at much higher rates than white males.
There are two ways to measure the differential undercount. One method compares census counts with population estimations based on demographic analysis of births and deaths and immigration and emigration rates. The second method, known as dual system estimation, derives from sample survey data collected independently of the census, but for the same time frame, that are then matched with census records. This statistical method has a major advantage over demographic analysis in that it permits the calculation of undercount rates for many more population groups and in greater geographic detail than national or state estimates. Dual system estimation also allows demographers to adjust the initial census count to correct for any detected under- or overcounting.
Recognizing that some population groups--the linguistically isolated, recent immigrants, those without legal status, the highly mobile, the poor and uneducated, and those indifferent to civic responsibility--are more likely to be missed, the Census Bureau accordingly proposed to use a post-census sample and dual system estimation as part of its methodology. Republican Party leaders in Congress objected and the Speaker of the House successfully sued the Department of Commerce to preclude the use of sampling for apportionment.
According to Prewitt, the accusation that the Census Bureau has acted or would act on a narrow partisan agenda fails to account for several factors: By tradition the Census Bureau is professional and apolitical. It lacks statistical expertise in reapportionment or redistricting and has no insight on trends in voting behavior.
Furthermore, the census is designed and executed under a spotlight. For Census 2000, for example, four congressional committees have oversight responsibility; continuous examinations are undertaken by the General Accounting Office and the Inspector General; a specially appointed Census Monitoring Board is given the explicit charge to report to the Congress and the President as to whether there is any evidence of political bias; several dozen journalists track census issues; and a half-dozen major standing advisory committees, including one appointed by the National Academy of Sciences, pay close attention to the census design and operations. "None of these oversight processes, advisory groups, or public watchdog efforts has identified any instance of partisan intention in the Census 2000 design--not for lack of looking, but because none is to be found," Prewitt concludes.
Kenneth Prewitt is Director, U.S. Census Bureau.
Population and Development Review, Vol. 26 No. 1, March 2000, also includes Notes and Commentary, Data and Perspectives, Archives, Book Reviews, and Documents.
For subscription information call 212/339-0514 or fax 212/755-6052. For further information, contact Christina Horzepa, 212/339-0520, or Sandra Waldman, 212/339-0525.
The Population Council is an international, nonprofit, nongovernmental institution that seeks to improve the wellbeing and reproductive health of current and future generations around the world and to help achieve a humane, equitable, and sustainable balance between people and resources. The Council conducts biomedical, social science, and public health research and helps build research capacities in developing countries. Established in 1952, the Council is governed by an international board of trustees. Its New York headquarters supports a global network of regional and country offices.
# # #
RSS