One of the oldest and most widespread use of counting and mathematics is the census. The census is the official count of the people in a geographic area. The Constitution of the United States calls for a census to be taken every 10 years. Originally, the purpose of the census was to provide information for the periodical  reapportionment of the House of Representatives. Today, census data is used for many reasons, including establishing where and how federal and state monies should be distributed.

   The U.S. Constitution requires its citizens to be “enumerated” or physically counted. But the precise meaning of the word “enumerate” has been debated in recent years. The beginning of the twenty-first century still sees census officials and researchers recommending different methods for estimating the number of people living in an area rather than counting. Debate continues around estimation methods because they can give different results than traditional counting methods.

Problems with Undercounting

Under the original Constitution, Native Americans were not counted in a census. African Americans were counted, with 92 percent listed as slaves,  and each slave counting as three-fifths of a man. Other groups, such as Asians, were not included at all at this point. Over the years, different groups gained recognition and were included in the census. But even in the twentieth century, some groups continued to be undercounted, such as those living in dense urban areas. With multiple family units often living at one address, the chance of finding an address for all individual families is less than for single-family residences.

Although counting methods today recognize all ethnicities and socioeconomic classes, concerns about undercounting are still an issue. After the 1990 census, at least three cases went before the United States Supreme Court regarding problems or disagreements with either the manner in which the census was completed or its results. In Wisconsin v. City of New York (1996), the charge was that the census had undercounted a population of New York by not including some members of certain minority groups. This alleged undercount benefited the State of Wisconsin by increasing the number of representatives from Wisconsin. The U.S. Secretary of Commerce refused to make any statistical change to the data: this decision was upheld by the Supreme Court.

      In two other cases brought to court in 1996, states were required to reapportion the congressional districts in their state. In doing so, districts highly irregular in shape were created in order to given voting strength to a minority group. The Supreme Court ruled these redistricting plans violated the equal protection clause of the United States Constitution.

The Debate over Estimation

Around 1770, a science known as statistics was defined. Statistics is a branch of mathematics dealing with the collection, analysis, interpretation and presentation of masses of numerical data. Statistics were so new in the late 1700s that the authors of the United States Constitution did not have any confidence in this “new” science, and did not use it for census purposes. By the end of the 1800s, however, statistical methods and knowledge increased, and the concept of estimating a population by sampling a small part of it was no longer a strange thought.

Today, mathematicians believe that using sampling and estimation methods will reduce undercounts. Some ideas presented at a 1999 conference included basing estimates on water demand, birth records, death records, IRS tax returns, master address files, and census blocks, all of which have been utilized with varying degrees of success. The National Academy of Sciences recommended using an estimation method combined with traditional enumeration to fulfill federal law requirements and increase the accuracy of the census.

      But a political debate centers on how the census count will affect the congressional districts. By using traditional enumeration methods, the likelihood of minorities being undercounted is greater. Therefore, individuals who were elected by majority constituents desire the census to remain the same. In contrast, individuals who would benefit from a larger participation of minorities in elections prefer the census to be conducted statistically which would therefore increase estimates of minorities.

The Next Census

The 2000 census, with all of its faults, was the most detailed in U.S. history. It included race self-identification questions as well as ethnicity identification. Although this is more information than ever requested before, some minority groups allege that the census remains incomplete. Some of the problems encountered with the 2000 census included non-English-speaking citizens who could not read the form, the inability to include more than one race for couples who are biracial, and the lack of a permanent address at which to be contacted. Additionally, some believe the Census Bureau does not have the right to ask detailed personal questions, such as income or race.

       Every tenth year during census time, many of these same questions resurface. Why does the Census Bureau need to know such information?

Why does the U.S. Code not allow mathematicians to take advantage of statistics to simplify and make the count more accurate? These questions will surely be addressed again before the 2010 census occurs.

______________________________________________________________________________________________

Reference

Anderson, Margo J. and Stephen E. Fienberg. Who Counts? The Politics of CensusTaking in Contemporary America. New York: Russell Sage Foundation, 1999.

Cohen, Michael L., Andrew A. White, Keith F. Rust. Measuring a Changing Nation:Modern Methods for the 2000 Census. Panel on Alternative Census Methodologies,

National Research Council. Washington, D.C.: National Academy Press, 1999.

Keever, Beverly Ann Deepe, Carolyn Martindale, Mary Ann Weston. U.S. Coverage of Racial Minorities: A Sourcebook, 1934–1996. Westport, CT: Greenwood Press,

1997.

National Academy of Sciences. Modernizing the U.S. Census. Commission of Behavioral and Social Sciences and Education. Washington, D.C.: National Academy

Press, 1994.