Census 2000 Statistical Sampling

by Jared Feldman

The U.S. Constitution mandates that a national census be conducted every ten years. This enumeration serves as the cornerstone of government operations, affecting the allocation of seats in the U.S. House of Representatives; the drawing of Congressional districts; federal allocations for essential services, such as education, law enforcement, health care, and transportation; and implementation of civil rights statutes. Simply put, being counted in the census ensures representation in U.S. government. An inaccurate census count puts cities and states with large numbers of uncounted individuals at a profound disadvantage, both in terms of Congressional representation and allocation of government funds.

In the largest undercount ever recorded, it is estimated that the 1990 Census failed to count 8.4 million people, predominantly poor, homeless and minority populations. To prevent a repeat of this serious error, the Census Bureau developed a plan approved by the National Academy of Sciences that would improve direct enumeration efforts and then supplement those efforts with statistical sampling to ensure the most accurate 2000 Census possible. This plan has been endorsed by the American Statistical Association, the American Demographics Association, and virtually every other professional organization concerned with such issues. However, congressional leaders have persisted in opposing the use of statistical sampling, claiming concern about political manipulation of the figures produced, and worrying about the political impact these numbers might have on re-districting activities.

The Jewish Council for Public Affairs (JCPA) believes strongly that procedures for conducting an accurate census should be determined not on the basis of politics but on objective, scientific grounds. The U.S. government has a moral obligation, and the constitutional power, to ensure that the census it conducts is the most accurate possible and does not further marginalize disadvantaged populations who already live under difficult conditions. The strong possibility of a disproportionate undercount of poor, homeless, and minority populations, when reliable scientific means, such as statistical sampling, are available to ensure a more accurate count, violates the very notion of representative democracy upon which this nation is founded.

The organized Jewish community historically has championed the cause of civil rights and the principles of equal access and equal opportunity for all Americans. We have also recognized the primary role of the federal government in meeting the needs of the most vulnerable segments of our society, and we have expressed concern about the undue burden placed on states, municipalities, and their taxpayers, when that responsibility is not met. Based on these essential values of social justice and equity, and recognizing that the process must be protected by adequate safeguards against distortion by political manipulation, the JCPA urges support for the use of statistical sampling, as a reliable and legitimate means of supplementing direct enumeration in the preparation of Census 2000, to ensure that all segments of the American population are afforded an opportunity for equal representation.

 


About the Author


Jared Feldman