Academic journal article The Western Journal of Black Studies

The Census, Sampling and African Americans

Academic journal article The Western Journal of Black Studies

The Census, Sampling and African Americans

Article excerpt


African Americans have been systematically and disproportionately undercounted by the census. There are several reasons for this undercount but the most important being the inherent biases of the Bureau's current counting method, the straightforward headcount method. To combat this problem, in the 2000 census, the Bureau proposed to use sampling together with the headcount method. The proposal was widely supported by demographers, statisticians, and census scholars but was met with stiff opposition from politicians who either did not fully grasp the technical nature of sampling or ignored the significance of sampling in the census. This paper seeks to demonstrate that the arguments advanced by sampling opponents are technically unsound and legally unpersuasive and, in light of the undercount problem suffered by African Americans, suggest a new direction for Congress to deal with sampling in the future censuses.

The Decennial Census

The decennial census is one of the most venerable components of the American democracy. It was instituted on the very day the government was founded, and this fact alone has set it apart from all other censuses in the world. It was designed to implement one of the most important agreements of the 1787 Constitutional Convention, the Great Compromise, under which each state would have an equal number of senators and a proportional number of representatives to be determined by a count of each state's population. Slowly but steadily, it has gained widespread acceptance and has accomplished well what it was expected to do: to provide timely and accurate population statistics necessary for the American democracy's population-based system of representation. It has been a paragon of population enumeration for the whole world and an envy of many democracies around the world in their attempts at fair political representation on the basis of accurate population counts.

This is, however, a somewhat idealized textbook version of the census and its role in American history. As an integral part of the American political tradition, it has shared the very peculiar nature of the American dual system, a democracy for Whites and a hegemony for Blacks. It counted Whites as whole persons but Blacks as fractional persons. This counting duality was epitomized by the three-fifths compromise, lying at the heart of the census for a century since its very inception in 1787. This compromise resulted in a quintessential undercount of two-fifths or 40% of the Black population; it has run like an ugly thread through various patterns of American history and contributed to the peculiar shape of the American polity and its conflicts or compromises. In the society that preached equality, liberty, fraternity, and justice, this racial discrimination was an anomaly. An ideology that justified the subjection of Blacks was very difficult to reconcile with the virtue of equity and fairness for all. This barefaced contradiction proved too much to live with and finally was removed by the Fourteenth Amendment in 1866. By mandating the census to count a Black as a full unit of representation, this amendment freed the original counting formula from its White supremacy heritage and heightened the nation's commitment to establishing a system of government in which population serves as the measure of political power.

In addition to the official 40% undercount, Blacks have suffered from various forms of differential undercount. The 1870 census undercounted 0.5 million Blacks or 10% of the Black population (Robey, 1989: 33). The 1940 census undercounted 13% of Black men in the eligible-for-draft age (twenty-one to thirty-five). Nationally, 229,000 more Black men registered for the draft than would have been expected from the census counts (Price, 1947: 44-49). Comprising only 12% of the population, Blacks accounted for more than 50% of the 1980 undercount (Bailar, 1988: C3). In 1990, despite various outreach efforts to count minority groups, Blacks being undercounted reached a higher level at 5. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.