CENSUS

A national census of population is "the total process of collecting, compiling, evaluating, analysing and publishing or otherwise disseminating demographic, economic, and social data pertaining, at a specified time, to all persons in a country" (United Nations 1988, p. 3). The United Nations encourages its members to take regular censuses and provides technical assistance. It also publishes an annual demographic yearbook and maintains a web site with selected census data and other population information for most countries.

A nation may conduct censuses on other subjects such as housing, business firms, agriculture, and local governments. Because of the close link between families and housing units, censuses of population and housing are usually combined.

The actual enumeration for a national population census is usually spread over a period of weeks or months, but an attempt is made to record circumstances as of a designated census day (April 1 for the 2000 U.S. census). In some censuses, persons are reported in their legal or usual place of residence (a de jure enumeration). According to the Bible, a census decree issued by the government in Rome ordered that persons be counted and taxed in their home towns. In response, Joseph and his pregnant wife, Mary, traveled from Nazareth to Bethlehem, where Mary gave birth to Jesus. Other censuses record people where they are on the census day (a de facto enumeration). Whatever residency rule is followed, all conceivable special circumstances must be anticipated, questionnaires well designed, and census staff thoroughly trained.

The U.S. census uses the concept of "usual residence," where the person lives and sleeps most of the time, but some exceptions are made. A few examples from the 2000 U.S. census illustrate the range of circumstances. Some persons staying in hotels and rooming houses are transients temporarily absent from their usual homes; they completed special census forms that were later compared to the census forms received from their home addresses to confirm that they were properly reported by other household members. An operation called "Service-Based Enumeration" was designed to locate homeless persons and others with no usual residence. College students were presumed to be residents at the place they lived while attending college, even if they were at home on spring break on census day. Efforts were made to avoid having an away-at-college child erroneously double-counted as a member of the home household. Pre-college students at away-from-home schools were assigned to the parental household. Citizens working in another country for the U.S. government, and their families, were enumerated (primarily through the use of administrative records), but other citizens living abroad at the census date were not counted. Short-term tourists, however, were supposed to be reported by other members of their households, or by late enumeration upon their return home. Persons not legally resident in the United States were supposed to be counted, but many feared all contact with authorities and took steps to avoid detection by census procedures.

The total population count is only one result of a modern census. Information is also obtained on many characteristics, such as age, sex, and relationship to others living in the same household, educational level, and occupation. The census then reports on population size and characteristics for provinces, states, counties, cities, villages, and other administrative units. To obtain all this information, the 2000 U.S. census (U.S. Census Bureau 1999) asked every household to complete a "short form" with six questions about each person and one question about each housing unit. A carefully selected sample of one of every six households received a "long form" that included additional questions on fourteen topics for each person and thirteen topics for each housing unit.

Reporting information for the entire country and for the thousands of cities and other subareas results in many volumes of printed reports. Many nations issue additional data on computer-readable files designed for convenient use by national and local governments, organizations, and individuals. Increasing quantities of data are available to anyone with access to the Internet.

In the United States, the idea for a regular census emerged during debates about the problems of creating a representative form of government. The U.S. Constitution (Article I, Section 2) directs that membership in the House of Representatives be based on population: "The actual enumeration shall be made within three years after the first meeting of the Congress of the United States, and within every subsequent term of ten years, in such manner as they shall by law direct."

The questions asked in a census reflect political and social issues of the day and often provoke spirited debates. For example, slavery was a recurrent divisive issue for the Constitutional Convention. One of the two newly created legislative bodies, the House of Representatives, was to have the number of legislators from each state proportional to the population of the state. Including slaves in the population count would increase the legislative power of southern states. A compromise (Article I, Section 2) provided that: "Representatives. . . shall be apportioned among the several states. . . according to their respective numbers, which shall be determined by adding to the whole number of free persons, excluding Indians not taxed, three-fifths of all other persons." The distinction between "free persons" and "all other persons" (slaves) was eliminated after the Civil War (Amendment 14, Section 2).

The 1790 census was modest in scope. Assistants to federal marshals made lists of households, recording for each household the number of persons in five categories: free white males over sixteen and under sixteen, free white females, other free persons, and slaves (Anderson 1988, p. 13).

Censuses from 1790 through 1840 were conducted with little central organization or statistical expertise. A temporary federal census office was established to conduct the 1850 census. The individual rather than the household became the focus of the enumeration, and the content of the census was expanded to include occupation, country or state of birth, and other items. Experienced statisticians were consulted, and the United States participated in the first International Statistical Congress in 1853.

The temporary office for the 1890 census became one of the largest federal agencies, employing 47,000 enumerators and 3,000 clerical workers. To help with the enormous task of tallying data, census officials supported the development by a young inventor, Herman Hollerith, of an electrical tabulating machine. His punched-card system proved effective in census operations and later contributed to the growth of the IBM corporation.

A permanent census office was established in 1902. An increasingly professional staff assumed responsibility for the population censuses and a broad range of other statistical activities. Deficiencies in the federal statistical system became apparent during the 1930s as the nation tried to assess the effects of the Great Depression and to analyze an array of new programs and policies. Several federal agencies successfully advocated expansion of the social and economic content of the 1940 census. New questions in 1940 asked about participation in the labor force, earnings, education, migration (place of residence in 1935), and fertility. A housing census was paired with the population census to provide information about housing values and rents, mortgages, condition of dwellings, water supply, and other property issues. The practice of providing population and housing information for subareas of large cities was greatly expanded.

Governmental statistical agencies are often set in their ways, concerned with continuity rather than innovation and removed from the higher levels of policymaking. For several decades beginning in the late 1930s, the U.S. Census Bureau was extraordinarily creative. Social scientists and statisticians employed by the Census Bureau were active in research and scholarly publication and played leadership roles in professional organizations. Census Bureau personnel pioneered in development of the theory and practice of population sampling. To accommodate large increases in content and geographic scope, the 1940 census questionnaire was divided into two parts. A set of basic questions was asked of everyone, while a set of supplementary questions was asked of a one-in-twenty sample. Subsequent U.S. censuses have continued to use sampling, as with the "long-form" versions of the 1990 and 2000 questionnaires that went to one of every six households. Sampling theory was also applied to the development of periodic sample surveys to provide timely information between censuses and to provide information on additional topics.

For the 1950 census, the Bureau participated actively in the development and utilization of another new technology, the computer. In later censuses, the schedules were microfilmed and optically scanned for direct input of information (without names and addresses) into a computer for error-checking, coding, and tabulating.

Through 1950, census enumeration in the United States was accomplished almost entirely by having an enumerator conduct a personal interview with one or more members of each household and write the information on a special form. By 1990, most census questionnaires were distributed and returned to the Census Bureau by mail, with telephone follow-up replacing many personal visits. Large-scale use of the Internet and other innovative response technologies may be practicable for widespread use in the 2010 census.

National censuses have so far been the best method for obtaining detailed information about the entire population. Only censuses provide counts and characteristics for small administrative and statistical areas, such as villages, voting districts, city blocks, and census tracts. Only censuses provide reliable information on small population groups, such as the income distribution of female plumbers. Only censuses provide the large numbers of cases needed for complex multivariate analyses. But censuses are so large and difficult to process that by the time the data are released they are, for some purposes, out of date, and this problem increases throughout the interval, commonly ten years, until the next census.

To address the timeliness problem, most national statistical agencies conduct a series of large sample surveys that provide select information on key topics at timely intervals. In the United States, for example, the latest unemployment rate, released on the first Friday of each month, is headline news. It is based on the Current Population Survey, a monthly sample survey that provides current information on many social and economic characteristics.

A few countries keep population registers, continuous records of where people live along with some of their characteristics. If reasonably complete and accurate, these may be used to supplement and update some census information. In many nations, administrative records such as social security, national health, tax, and utility company files, may be adapted to provide periodic estimates of population size and a few characteristics, but difficult questions arise about data quality and comparability with census- and survey-based information. Citizens of many nations are wary about letting the government and corporations compile extensive personal data. Legislative restrictions are increasingly being placed on what information may be gathered and stored and how it may be used.

The questions asked in successive censuses typically change more slowly than the procedures. Keeping the same topics and the same wordings of questions help a government measure change from one census to the next. This appeals to researchers and policy analysts, but policymakers and administrators, whose attention is focused mainly on current programs and next year's budget, often plead for new wording to better serve current concepts. Another reason tending to stifle innovation is that the census is an expensive and visible tool. A lengthy review process confronts any agency that seeks to add, delete, or alter a question. In the United States, both the executive branch and the Congress must approve the final census schedule. A question on pet ownership has been regularly proposed but rejected because there is no compelling governmental interest in such a question and private sample surveys can provide the desired information. A third reason the content of the questionnaire changes little is that proponents of new topics and questions must compete for questionnaire space, while questions previously asked tend to already have a network of users who have a vested interest in retaining the topic. This competition spurs extensive lobbying and mobilization of support from federal agencies, congressional committees, and interest groups.

An innovative approach to providing timely data is "continuous measurement" using a "rolling-sample" survey. The American Community Survey has been implemented by the U.S. Census Bureau (1999), with plans to expand to three million households a year in 2003. Because the survey uses a small, permanent, well-trained and supervised staff (as compared to the large, temporary, briefly trained staff utilized for a census), data quality may be higher than in a census. Another advantage of a monthly survey over a decennial census is the ease of introducing new topics and testing the effects of changes in the wording of questions.

American Community survey results will be cumulated over a year to provide estimates of numbers and characteristics for the nation, states, and places or groups of 65,000 or more people. For smaller places and groups, or for more reliable and detailed data for larger places, data will be cumulated for periods of up to five years. The five-year cumulated sample is designed to be approximately equivalent to the census "long-form" sample and may eliminate the need for a "long form" in the 2010 census. Moving averages could provide annual updates and be used for time series. There are many issues to be worked through in developing and evaluating this type of survey and in determining how well information averaged over a long period compares to standard statistical measures based on time-specific censuses and surveys.

In many nations, questions on race and ethnicity are a sensitive and contentious topic. In the United States, the groups recognized have changed from each census to the next. For example, special tallies were made from the 1950 and 1960 censuses of persons with Spanish surnames, but only for five southwestern states. Beginning in 1970, and elaborated in later censuses, persons were asked if they were of Spanish/Hispanic origin. Many responses to this question have seemed to be inconsistent with responses to the separate question on race. The Bureau has conducted sample surveys to test various question wordings, and has sponsored field research and in-depth interviews to provide insights into how people interpret and respond to questions about ethnicity. For the 2000 census the basic concepts were retained, but the wording of each was adjusted, provision was made for individuals to report more than one race, and the question on Hispanic ethnicity was placed before the question on race.

In recent decades, extensive social science research has been conducted in many parts of the world on issues of racial and ethnic identity. These identities are almost always more flexible and more complex than can be captured with simple questions. A further complication is that many persons have ancestral or personal links to two or more racial and ethnic groups; how they respond depends on how they perceive the legitimacy and purposes of the census or survey.

In the 1990s, the U.S. Office of Management and Budget conducted an extensive review of its policy statement that specifies the standard race and ethnic classifications to be used by all government agencies, including the Census Bureau (Edmonston and Schulze 1995). Serious methodological problems have arisen as a result of inconsistencies among classifications of individuals on repeated interviews and in different records. A dramatic example occurs with infant mortality rates for race and ethnic groups, which are based on the ratio of counts based on birth registration to counts based on death registration of infants less than one year old. Comparing birth and death certificates for the same person, recorded less than one year apart, revealed large numbers for whom the race/ethnic classification reported at birth differed from that reported at death. The high-level government review also heard from many interest groups, such as Arab-Americans, multiracial persons, and the indigenous people of Hawaii, who were anxious for the government to enumerate and classify their group appropriately.

In the United Kingdom, controversy about a proposed question on ethnic identity led to the question's omission from the 1981 census schedule, thus hindering analyses of an increasingly diverse population.

The most politically intense and litigious controversy about recent U.S. censuses is not that of content but of accuracy. States, localities, and many interest groups have a stake in the many billions of dollars of government funds that are distributed annually based in part on census numbers.

President George Washington commented about the first census that "our real numbers will exceed, greatly, the official returns of them; because the religious scruples of some would not allow them to give in their lists; the fears of others that it was intended as the foundation of a tax induced them to conceal or diminish theirs; and through the indolence of the people and the negligence of many of the Officers, numbers are omitted" (quoted in Scott 1968, p. 20).

A perfect census of a school, church, or other local organization is sometimes possible, if membership is clearly defined and the organization has up-to-date and accurate records, or if a quick and easily monitored enumeration is feasible. A national census, however, is a large-scale social process that utilizes many organizations and depends on cooperation from masses of individuals. Planning, execution, and tabulation of the 2000 U.S. Census of Population extended over more than ten years. Hundreds of thousands of people were employed at a cost of several billion dollars. A discrepancy between the results of a national census and "our real numbers" is inevitable.

Statisticians, recognizing that error is ubiquitous, have developed many models for identifying, measuring, and adjusting or compensating for error. Census statisticians and demographers around the world have participated in these developments and in trying to put professional insights to practical use.

Following the 1940 U.S. census, studies comparing birth certificates and selective service records to census results documented a sizable net undercount. To provide more information on census coverage and accuracy, a post-enumeration survey was conducted following the 1950 census using specially selected and closely supervised enumerators. One finding from the survey was that the undercount of infants in the census did not arise, as first thought, from a tendency for new parents to forget to mention a new baby to the census enumerator. The problem, rather, was that many young couples and single parents had irregular and difficult-to-find living arrangements, and entire households were missed by census enumerators. Based on this and related findings, much effort was given in subsequent censuses to precensus and postcensus review of lists of dwelling units.

Special surveys preceding, accompanying, and following censuses are increasingly used to provide evidence on magnitude of error, location of error in specific groups or places, and procedural means for reducing error. Techniques of "demographic analysis" have also been developed and continually refined to provide evidence of census error. The basic technique is to analyze the numbers of persons of each age, sex, and race in successive censuses. For example, the number of twenty-eight year-old white women in the 1990 census should be consistent with the number of eighteen-year-olds in 1980 and eight-year-olds in 1970. Information on births, deaths, immigration, and emigration is taken into account. Based on demographic analysis, the estimated net undercount for the total U.S. population was about 5.6 percent in 1940, 1.4 percent in 1980 (Fay, Passel, and Robinson 1988, Table 3.2) and 1.8 percent in 1990. Estimates of net undercount have also been made for specific age, sex, and race groupings.

If the net undercount were uniform for all population groups and geographic regions, it would not affect equity in the distribution of seats in Congress or public funds. But census errors are not uniform. The estimates for 1990 show net undercounts exceeding 10 percent of adult black males and small net overcounts for some age and race groups.

In 1980, the mayor of Detroit, the City of Detroit, New York City, and others sued the federal government, alleging violation of their constitutional rights to equal representation and fair distribution of federal funds (Mitroff, Mason, and Barabba 1983). More litigation occurred with respect to the 1990 census. In neither case were the initially reported census counts adjusted to reflect estimated errors in enumeration, although in the case of the 1990 census the Census Bureau technical staff and director thought their methodology would improve the accuracy of the counts.

The Census Bureau developed innovative plans for the 2000 census to integrate an evaluation survey with the census, to use sampling to increase quality and reduce costs of enumerating nonresponsive households, and to produce official counts that already incorporate the best available procedures for minimizing error.

These Census Bureau plans engendered major political and budgetary battles between the Republican-controlled Congress and the Democratic administration. Many of the persons hardest to reach using traditional census methods are poor, often minorities. Many members of Congress assumed that a census with near-zero undercount would increase the population reported for states and localities that tend to vote more for Democrats, and hence reduce, relatively, the representation for states and localities that tend to vote more for Republicans. At the national level, this could alter the reapportionment among states of seats in the House of Representatives. Legislatures in states, counties, and cities are also subject to decennial redistricting according to the "one-person, one-vote" rule, and similar shifts in relative political power could occur.

The controversy between legislative and administrative branches over methods for the 2000 census was taken to the Supreme Court, with plaintiffs seeking a ruling that the Bureau's plans to use sample-based estimates would be unconstitutional. The Supreme Court avoided the constitutional argument, but ruled that current law did not permit such use of sampling. The Democrats lacked the power to rewrite that part of the basic census law. The two parties compromised by appropriating additional dollars to cover the extra costs of complete enumeration while retaining funds for many of the methodological innovations that permit preparation of improved estimates. Reapportionment among states of seats in Congress could therefore use the results of traditional complete enumeration. Most scholars, analysts, and other users of census data are likely to consider the adjusted estimates more accurate and hence more useful.

The percentage net undercount understates the total coverage error in the census. If one person is omitted and another person counted twice, the total count is correct but there are two "gross" errors that may affect the counts for specific places and characteristics. Identification of gross errors is receiving increased attention, to improve information on the nature of census error and potential error-reducing methods, and to facilitate development of techniques of data analysis that compensate for known and unknown errors.

One of the difficulties in debates about undercount and other census error arises from confusion between the idea of a knowable true count and the reality that there is no feasible method for determining with perfect accuracy the size and characteristics of any large population. A national census is a set of procedures adopted in a political, economic, and social context to produce population estimates. It is politically convenient if all parties accept these results as the common basis for further action, much as sports contests use decisions by umpires or referees. The more that policy and budgets depend on census results and the more aware that politicians and citizens become of the importance of the census, the more contentious census taking will be.

Every nation confronts political and social problems with its censuses. Regularly scheduled censuses are often postponed or abandoned because of international conflict. The United States and the United Kingdom canceled plans for mid-decade censuses because of national budget constraints. West Germany was unable to take a census for several years because of citizen fears about invasions of privacy. Ethnic conflict has interfered with census taking in India, Lebanon, and other nations.

The processes by which census procedures are determined, the ways in which census figures are used, and the conflicts that occur about these procedures and numbers are not merely "technical" but are embedded in a broader social process. The character of a nation's census and the conflicts that surround it are core topics for the "sociology of official statistics" (Starr 1987).

(see also: Demography; Population)

references

Anderson, Margo J. 1988 The American Census: A Social History. New Haven: Yale University Press.

——and Stephen E. Fienberg 1999 Who Counts? The Politics of Census-Taking in Contemporary America. New York: Russell Sage Foundation.

Edmonston, Barry, and Charles Schultze 1995 Modernizing the U.S. Census. Washington, D.C.: National Academy Press.

Mitroff, Ian I., Richard O. Mason, and Vincent P. Barabba 1983 The 1980 Census: Policymaking and Turbulence. Lexington, Mass.: Lexington Books.

Scott, Ann H. 1968 Census, USA: Fact Finding for the American People, 1790–1970. New York: Seabury Press.

Starr, Paul 1987 "The Sociology of Official Statistics." In W. Alonso and P. Starr, eds., The Politics of Numbers. New York: Russell Sage Foundation.

Steffey, Duane, and Norman Bradburn (eds.) 1994 Counting People in the Information Age. Washington, D.C.: National Academy Press.

United Nations 1998 Principles and Recommendations for Population and Housing Censuses. ST/ESA/STAT/SER.M/67/Rev.1. New York: United Nations.

United States Census Bureau. Census Bureau Home Page. On the Internet at http.//www.census.gov. See links to Census 2000 and American Community Survey.

KARL TAEUBER

CENSUS

CENSUS. The word census is a Latin term, and efforts during the early modern period to conduct population surveys were historically descended from the Roman census process, which was based on sworn declarations of the age, number of family members, and property of individual households. Early modern political writers were impressed by the Roman state's ability to enumerate and assess its subject population. Because the equation of a commonwealth's population with its strength had by the later sixteenth century become a commonplace, enthusiastic recommendations of the census were made by a host of thinkers, including Jean Bodin (1576), Giovanni Botero (1588), Justus Lipsius (1589), and other political thinkers. Yet, if the historical memory of the Roman census had survived the collapse of the Roman Empire, the administrative ability to actually conduct one did not. As a result, full territorial enumerations were only sporadically carried out in most of Europe until the end of the eighteenth century.

The most significant exception to this generalization was Italy itself, where true censuses (as opposed to household listings and tax surveys) were already being carried out by the end of the Middle Ages. The Italian city-states were especially (and unsurprisingly) advanced in this regard, and censuses had already been carried out in Florence (1380), Treviso (1384), Padua (1411), Verona (1473), Reggio (1473), Palermo (1479), Brescia (1493), Parma (1508), Venice (1509), and Rome (1526). Enumerations were also conducted in Italian territorial principalities, including the duchies of Ferrara (1431) and Mantua (1451), and in Sicily (1501). The mature administration of these censuses reflects the much greater sophistication of public administration in Italy than elsewhere in Europe, and early modern Italian censuses were much more than simple head counts. The Sicilian censuses, of which there were fourteen between 1501 and 1747, listed every individual by name and relationship to the head of the household, and separated out those males of arms-bearing age. By the sixteenth century, Italian censuses often recorded detailed information on the age structure of the population, and the exact age of every inhabitant was recorded at Pozzuoli (1489), Sorrento (1561), and Carpi (1591). By comparison, the English and American censuses did not list every individual by name until 1841 and 1850, respectively. It is also worth noting that the registration of births began in Siena in 1381, in Florence in 1450, and in Bologna in 1459, whereas parish registers do not survive before the mid- to late-sixteenth century in Protestant Europe, and before the seventeenth century in most of Catholic northern Europe.

Italian precocity did not mean, however, that the rest of Europe had ceased to carry out population surveys altogether, and the early modern period generated a mass of such material. Thus, population surveys begin to appear on monastic estates in France and Germany as early as the ninth century, and by the later Middle Ages full population counts were taken in several German cities, for example in Nuremberg (1449), Nördlingen (1459), and Strasbourg (1473), although these were not followed up on a regular basis.

Especially at the level of the local territory or community, a huge variety of other surveys were conducted with ever greater frequency during the early modern period. Muster rolls listing all men eligible for military service were drawn up on an irregular basis in various European communities. There was also a variety of specialized censuses, such as the Norwich Census of the Poor (1570) and the Castilian educational census of 1764, which was designed to determine the number of students who were attending various education institutions. In the Holy Roman Empire a number of territorial authorities (for example, the Bishop of Speyer in 1530, and the monastery of Ottobeuren in 1548, 1556, 1564, and 1586) compiled Leibeigenbücher, 'serf registers', which recorded the free (or servile) status of every man, woman, and child in the territory. During the sixteenth century both Catholic and Protestant episcopal authorities began conducting parish-by-parish counts of the number of communicants (all persons over twelve to fourteen years of age), and these surveys became ever more detailed and systematic during the seventeenth and eighteenth centuries.

Still more detailed was the liber status animarum, or listing of each parish resident, which the papacy, in 1614, ordered every parish priest to maintain. These listings were less commonly compiled than the more familiar baptismal, marriage, and burial registers, but many seventeenth- and eighteenth-century parish listings have survived from Catholic Europe, and in a few areas (Malta after 1687, for example) complete listings have survived for every single parish. Similar records were maintained in many Protestant areas. In England, listings have survived from scattered locations from the later sixteenth century, although it is a rare parish where more than one such survey has survived. More systematic efforts were undertaken on the Continent, especially in Sweden, where parish registration had begun in 1686. In 1749, the Lutheran parish clergy in Sweden and Finland (then a Swedish possession) were further required to maintain a continuously updated list of parish residents and submit quinquennial tabulations to the Tabellverket (Tabulation Office) of population numbers broken down by sex, age, marital status, occupation, and social status, in addition to annual statistics of births, marriages, deaths, and (in the nineteenth century) migration.

By far the most common type of early modern enumeration, however, was a survey of hearths or heads of households, made almost always for fiscal reasons. As with the census itself, the earliest such territorial hearth tax surveys were carried out in Italy, as at Pavia (1250), Pistoia (1255), Perugia (1278), Padua (c. 1281), Reggio Emilia (1315), Florence (1351), Sicily (1374), and Venice (1379). The Florentine catasto (tax survey) of 1427 went so far as to record not only the name, age, marital status, and profession of the household head, but also the number of other individuals in the family, the type of residence (owned or rented), the number and value of livestock, the value of private and public investments, and the capitalized value of real property.

Beyond Italy, England stands out as a kingdom of very early tax surveys. Because of its unusually centralized monarchy, national tax surveys began in England as early as 1086 (William I's famous Domesday Book), and were repeated with varying degrees of completeness in 1279–1280 (the Hundred Rolls), 1377 (Edward III's Poll Tax), 1524–1525 (Henry VIII's Lay Subsidy), and 1662–1674 (Charles II's Hearth Taxes). Elsewhere in Europe full territorial tax surveys were conducted in France (1328), the German lands of the Holy Roman Empire (1495), Portugal (1527), Bohemia (1653–1655), Moravia (1655–1657), Ireland (1659), and Austria (1749–1750). Before the eighteenth century these large-scale surveys were only infrequently attempted; thus in Portugal there was only a single national survey between 1527 and 1736, and this one (in 1636) was seriously inaccurate. Outside of Italy, perhaps the most regular set of national household surveys were conducted in Castile between 1528 and 1536, 1541, 1552, 1561, 1571, 1587, 1591, and 1596, and were supplemented by the so-called relaciones topográficas of 1575–1578, a set of questions about local customs, economic conditions, and institutional characteristics administered in each locality in the kingdom. Even then, the frequency of survey fell off in the following century.

Local hearth and household tax surveys were much more common than their national counterparts and grew in frequency over the course of the early modern period. Nevertheless, there were significant regional differences in detail. Thus, in northern France, local taille (direct property tax) rolls recorded little more than the payment made by each household (and even they are rare before 1650). By contrast, German tax surveys often itemized and valuated each item of a household's property, and the level of its debts, as early as the beginning of the seventeenth century.

The close connection between census taking and taxation was, of course, recognized both by administrators and those they surveyed, and fears of excessive taxation would move the British Parliament to reject a census bill as late as 1753. Nevertheless, by the end of the eighteenth century the census was recognized as an essential tool of government, and regular population surveys were initiated (or at least attempted) in Norway (1769), France (1774, 1790), Denmark (1787), Belgium (1797), England (1801), Bavaria (1818), Saxony (1834), and Austria (1850).

See also Property ; State and Bureaucracy ; Taxation .

BIBLIOGRAPHY

Beloch, Julius. Bevölkerungsgeschichte Italiens. 3 vols. Berlin, 1937–1961.

Blaschke, Karlheinz. Bevölkerungsgeschichte von Sachsen bis zur industriellen Revolution. Weimar, 1967.

Herlihy, David, and Christiane Klapisch-Zuber. Tuscans and Their Families: A Study of the Florentine Catasto of 1427. New Haven, 1985.

Hollingsworth, Thomas Henry. Historical Demography. Ithaca, N.Y., 1969.

Mols, Roger. Introduction à la démographie historique des villes d'Europe du XIVe au XVIIIe siècle. 3 vols. Gembloux, 1954–1956.

Govind P. Sreenivasan

Census

The modern census

BIBLIOGRAPHY

A census of the population—that is, a counting of the people within the boundaries of a country— has become indispensable to any modern government. How many people are there? What are their basic socioeconomic characteristics? Where do they live, and how are they affected by the processes of social and biological change? These questions arise daily in the governments of all industrially developed countries, not to mention the governments of those that are still developing.

Censuses have come to include many topics other than population. Censuses of manufactures, agriculture, mineral industries, housing, and business establishments are taken by many countries, often independently of the census of population.

The following discussion is concerned primarily with censuses of population, but many of the comments—especially the comments on methods, tabulation, and quality of results—apply with equal force to other kinds of censuses.

Some early censuses. Counting the people, or some portion of them, is a practice that is probably as old as government itself. No one knows which ruler first enumerated the men for military purposes, or drew up a list of households with a view to taxing them. Figures obtained from censuses have long served as items of political propaganda, particularly in order to justify territorial expansion.

Population counts were reported in ancient Japan and were taken by the ancient Egyptians, Greeks, Hebrews, Persians, and Romans. Many of these early censuses appear to have covered only part of the population, often the men of military age, and the results were generally treated as state secrets. In Europe censuses on a city-wide basis or, as in Switzerland, on a canton-wide basis, were reported in the fifteenth and sixteenth centuries. A 1449 census of Nuremberg presumably was taken to determine the needed food supplies when a siege was threatened. The city of Madras in India reportedly took a census in 1687.

Various censuses have been claimed as the first held in modern times with purposes and methods resembling those of today. Among these are a census in New France (the early possessions of France in North America) taken at intervals between 1665 and 1754 and a census in Sweden in 1749. Census taking began early in the North American and South American colonies. The British Board of Trade ordered 27 censuses in the North American colonies between 1635 and 1776, and censuses were taken by these former colonies between independence in 1776 and the establishment of the United States.

The oldest continuous periodic census is that of the United States, which has been conducted every ten years since 1790. The census of the United Kingdom dates back to 1801, and a census has been taken there every ten years except in 1941, during World War II.

If one were to define a modern census as one in which information is collected separately about each individual instead of each household, then the beginning of modern censuses would have to be dated about the middle of the nineteenth century. Censuses along these lines were taken in Brussels in 1842, in all Belgium in 1846, in Boston in 1845, and in the entire United States in 1850. This is the procedure that is now generally in use.

International activities. Toward the end of the nineteenth century, the International Statistical Institute recommended international publication of the results of all censuses. Around the turn of the century the institute repeated its recommendation, pointing out that the results of some 68 different censuses were available, covering about 43 per cent of the world’s population. The institute had already adopted, in 1897, a set of rules for conducting censuses and presenting their results, while the whole topic had been discussed by the International Demographic Congress as early as 1878.

One of the early proposals considered by the United Nations was that it develop plans for a 1950 world census of population. Although it found that conditions were not ripe for such an effort, the United Nations took steps to foster census taking and made recommendations to improve the comparability of results. The Inter American Statistical Institute meantime had undertaken a program for the 1950 Census of the Americas, and 18 of the 21 countries in the Americas took censuses between 1945 and 1954. Throughout the world at least 150 areas took censuses in this period, collecting individual data on more than 2 billion people. For the decade centered on 1960, the number of censuses was about 180, including 2.2 billion people.

The use of censuses. The reasons for taking a census vary with the needs of the countries involved. The current concern with social and economic development is one of the prime reasons. Much information is needed in order to institute programs that will improve health, literacy, education, income, levels of living, supplies of food and other consumer goods, agricultural production, and industrial output. Census data are also collected in order to determine the representativeness of legislative bodies, the number of persons eligible to vote, and the areas or groups that have a claim to benefits deriving from the state.

The census provides a basis for much demographic, economic, and social research. It makes possible the identification and description of such groups as the labor force, economically dependent persons, recent migrants to cities, rural and urban populations, racial or religious minorities, refugees, scientific and technical workers, and others. Comparisons of successive censuses show changes in the numbers, characteristics, and location of the population. The census returns are also used as a frame from which samples are selected for subsequent inquiries.

The modern census

The United Nations (1958, p. 4) gives the following definition of a modern population census: “A census of population may be defined as the total process of collecting, compiling and publishing demographic, economic and social data pertaining, at a specified time or times, to all persons in a country or delimited territory.”

It also listed six essential features of a census, as follows. (1) A census must have national sponsorship. Only a national government can provide the necessary resources and enact suitable legislation, although provincial and local governments may share a part of the responsibility and sometimes a part of the cost. (2) A census must cover a precisely defined territory, boundary changes that affect comparisons between successive censuses should be clearly and explicitly stated. (3) All persons in the scope of the census must be included without duplication or omission. (4) The people must be counted as of a fixed time. Persons born after the census date are to be excluded, and persons who die after the census date are to be included. Some information, such as that relating to labor force participation or migration, may relate not to the census date but to another period, which must be clearly defined. (5) Census data must be obtained separately for each individual. This does not preclude making some entries for the entire household—and in exceptional circumstances summarized information for a group of persons may be acceptable—but the objective of a modern census, insofar as possible, is to collect data separately for each individual. (6) The data from a census must be published. Although at one time census reports were treated as state secrets, it is now recognized that a census is not complete until the data are compiled and published.

What is included. The content of a census, no less than its purpose, is determined by the country’s needs at the time. Questions of high importance in one country may be of relatively little importance in another, and questions of great significance at one time may be of little significance later. Every national census has changed over the years. Conditions in the country, alternative sources of information, and the ability of the census organization to provide the proposed information chiefly determine the census content.

The United Nations has approved a set of recommendations for national population censuses that includes the following list of question topics: location at time of census and/or place of usual residence, relation to head of household or family, sex, age, marital status, place of birth, citizenship, whether economically active or not, occupation, industry, status (as employer, employee, etc.), language, ethnic or nationality characteristics, literacy, level of education, school attendance, and the number of children born to each woman (Principles and Recommendations … 1958). For countries that could not include all items in the list, the following were suggested as a minimum: sex, age, marital status, and some indication of economic activity. Each item listed, of course, requires specific definition. Some items, such as marital status, type of economic activity, or level of education, apply to only part of the population.

Collection of data. Discussions of census methods generally distinguish between a count based on the actual location of the population on the census date and one which relates each person to the place of which he is a resident. Frequently residence is interpreted as the place where the person usually lives, although “usual residence” has no statutory meaning. On either basis national totals are usually about equal; community totals, however, may differ substantially.

To facilitate international comparison, the United Nations (Principles and Recommendations … 1958) has suggested the adoption of an “international conventional total.” This would include all persons present in the country on the census date except foreign military, naval, and diplomatic personnel: it would also include the country’s own military, naval and diplomatic personnel and their families located abroad, and merchant seamen normally resident in the country but at sea on the census date. The same document recommends that counts or estimates of the following groups should be given where feasible: indigenous inhabitants and nomadic tribes, civilian national residents temporarily abroad on the census date, and civilian aliens temporarily in the country on the census date. This is because whether or not these groups are included in a national census is usually determined by the laws and needs of each country; thus detailed comparisons between national totals can be made only if separate enumerations of these groups are made available.

Customarily two basic methods of collecting census data are recognized—direct enumeration and self-enumeration. Under the direct enumeration system, an enumerator collects information directly from the individual concerned, the head of the household, or some other member of the household who may be authorized to report for him. Under the self-enumeration system, the questionnaire is given to the individual, or the head of the household, who is expected to enter the required information and return the questionnaire to the census office. In self-enumeration the enumerator often delivers and collects the questionnaire. He may assist persons in making the entries, and he is responsible for assuring the accuracy and completeness of the completed questionnaires.

In a few countries the people are required to present themselves at a designated place to be enumerated. In a few others the population is immobilized on the census date, and no one is permitted on the streets unless he has been enumerated. In the great majority of cases the census enumerator seeks out the potential respondents and delivers questionnaires to them or asks them directly for the required information.

Census legislation normally provides that respondents must give full and correct information. This information must be given confidential treatment by all census employees. For persons who do not wish to have other members of their households or local officials see the information about themselves, census offices have provided special forms on which the person can insert the information and send it directly to the local, regional, or national office.

Population registers and lists of households, where available, are often used to help make the census complete. In some censuses the first step in the field work is to establish a list of housing units. The numbers assigned to these units are then used to control the completeness of the enumeration.

The questionnaire must be designed to permit ready entry of the required information and also be adapted to the tabulation procedure. In self-enumeration there is customarily a separate form for each household, with space for entering information for each household member. When information is collected directly by enumerators they may carry a separate form for each household or a form on which information is recorded for a number of households. Sometimes a separate document is used for each person.

Some countries have a long-established tradition of taking a census once every ten years, a few countries require one every five years, but most countries take a census when it is needed and do not have a fixed date. In a few unusual situations censuses have been taken more often than once in five years. Whether by law or by custom, the ten-year term appears to be most common.

In an increasing number of countries a pilot census is taken before the full census to test the inquiries, the procedures, the development of the field organization, and sometimes the tabulation program.

Use of sampling. Sampling has been extensively used in connection with censuses, although a census in principle requires that information be collected for each person. The term “sample census,” which is sometimes used, is a misnomer; a collection of information relating to only a specified part of a population should be termed a sample survey. In some censuses a part of the questions are asked of only a sample of all individuals; in this way more information can be collected without a comparable increase in respondent burden or tabulation work load. Many countries base some tabulations on a sample of the population instead of the total to provide preliminary totals for early release or to reduce the cost of some final tabulations. Sampling has been used to control the quality of the work at several stages of the processing of the data. In some countries a sample has been selected for a pretest and the resulting data have been used to validate the processing operations.

Tabulation of results. To be useful, individual census returns must be converted to statistical summaries. Tabulation methods vary from simple hand counts to processing on high-speed electronic computers. Processing may be done in the field, in provincial offices, or at the central office.

Questionnaires as received from the enumerators must be reviewed to locate incomplete or inconsistent entries. Procedures have been developed for correcting such errors by utilizing other information on the questionnaires or by using probability distributions based on data from other sources. In many instances the entries on the schedule must be converted to a numerical or other suitable code to facilitate counting and grouping into appropriate categories.

The most common procedure for grouping the entries is to punch cards and tabulate them by mechanical means. More recently, electronic computers have been adapted to this work, and devices have been developed to transfer the original questionnaire entries directly to the magnetic tape used in these computers. The capacity and speed of the computers have led to important advances in the amount of material tabulated and in the timeliness of the publications.

Completeness and accuracy. In view of the many public and private uses for census data, there is increasing concern with the quality of the results in terms both of completeness of coverage and accuracy of returns.

Completeness of coverage has received major attention. Unless special precautions are taken, some individuals may not be counted and some areas may not be enumerated. Undercounts of the population are more frequent than overcounts, although a few zealous census officials have been known to overcount the population of an area. Omission of individuals may arise from incomplete enumeration of a household or an area. Underenumeration has been particularly marked in the case of infants and young adults who are highly mobile or who lack firm occupational attachments. In cultures where men are considered more valuable than women, the number of women may be understated. In a few instances a fear that the census may be used for military conscription has led to some omission of men or to misreporting them as women.

Reporting residence may present problems. Some persons, such as migratory workers, have no usual place of residence. Others, such as college students, members of the armed forces, and persons engaged in long-distance transportation, may live away from their families for short or extended periods. Unless precise instructions are given for reporting these persons, they may be counted twice or not at all.

Age is often misreported either because people see some advantage in giving the wrong age or because they do not know their correct age. In many countries births are not recorded, and age is not precisely known. Some reasons for misreporting are that a person believes that despite pledges of confidentialness, the data will be used to his disadvantage; a parent believes that overstating a young child’s age may gain him entrance to school or that overstating an older child’s age may free him from compulsory school attendance; an older person may overstate his age to get social security benefits or to acquire the higher status sometimes accorded to the very old. Often people report a number ending in zero or five—even when asked for year of birth.

Nationality, citizenship, and mother tongue are particularly subject to misstatement when the correct answer may be considered detrimental in the context of the political situation of the group controlling the area.

Improvement of quality. Public confidence in the census greatly affects accuracy of returns. Public confidence can often be increased through an informational campaign, which explains that the data about any individual cannot be used to his detriment. Police officers, who are used as enumerators in some countries because of their availability and authority, are specifically excluded in others because they may give the impression that the information is not collected solely for statistical purposes.

To improve the quality of results, research has been devoted to reliability of response, role of enumerators, question wording, and attitudes of respondents in different situations. Increased attention is also being given to training the temporary field forces in census methods and concepts and in training the office staff in principles of editing and coding.

The quality of census statistics can often be determined by analyzing them after the complete census tabulations have been published. Formerly this work was performed mainly by scholars or by census officials willing to do scholarly work on their own time; but modern census offices increasingly are undertaking such analysis as part of their regular work. Checks are made for internal consistency of the census, for consistency with previous censuses, and also with other statistics, including estimates, that are entirely independent of the census. Both underenumeration and overenumeration of an age group can be determined by comparing the census results with the comparable age cohort for the previous census, adjusted for deaths and migration [seeCohort analysis]. Inaccuracies in age reporting that result from the preference of respondents for certain ages can be identified by internal analysis of the data. Sometimes the accuracy of census information for individuals is tested by comparison with information given for these same individuals in administrative records. Sample surveys may be taken to check on completeness of coverage and accuracy of census returns.

These evaluations aid in increasing the accuracy of future censuses, but they also provide users of the statistics with information about the reliability of the data. Although there has been some hesitation about revealing errors in the census, there is a growing recognition that full and frank disclosure leads not only to improvement in future censuses but also to increased public confidence.

Conrad Taeuber

[See alsoFertility; Government statistics; Mortality; Population; Vital statistics.]

BIBLIOGRAPHY

Brunsman, Howard G. 1963 Significance of Electronic Computers for Users of Census Data. Pages 269–277 in Milbank Memorial Fund, Emerging Techniques in Population Research. New York: The Fund.

Canada, Bureau of Statistics 1955 Ninth Census of Canada, 1951: Administrative Report. Volume 11. Ottawa: Cloutier.

Coale, Ansley J. 1955 The Population of the United States in 1950 Classified by Age, Sex, and Color: A Revision of Census Figures. Journal of the American Statistical Association 50:16–54.

Eckler, A. Ross; and Hurwitz, William N. 1958 Response Variances and Biases in Censuses and Surveys. International Statistical Institute, Bulletin 36, part 2: 12–35.

Jaffe, A. J. 1947 A Review of the Censuses and Demographic Statistics of China. Population Studies 1:308–337.

Mayr, Georg von (1895–1917) 1926 Statistik und Gesellschaftslehre. 3 vols. 2d ed., enl. Tübingen (Germany): Mohr.

Principles and Recommendations for National Population Censuses. 1958 United Nations Statistical Office, Statistical Papers Series M, No. 27.

Steinberg, Joseph; and Waksberg, Joseph 1956 Sampling in the 1950 Census of Population and Housing. Bureau of the Census Working Paper, No. 4. Washington: U.S. Department of Commerce.

Taeuber, Conrad; and Hansen, Morris H. 1964 A Preliminary Evaluation of the 1960 Census of Population. Demography 1, no. 1:1–14.

Taeuber, Conrad; and Taeuber, Irene B. 1958 The Changing Population of the United States. U.S. Bureau of the Census, Census Monograph Series, 1950. New York: Wiley.

United Nations, Statistical Office 1958 Handbookof Population Census Methods. Volume 1: General Aspects of a Population Census. New York: United Nations.

U.S. Bureau of Labor 1900 The History and Growth of the United States Census. Prepared for the Senate Committee on the Census. Washington: Government Printing Office.

U.S. Bureau of the CensusBureau of the Census Catalog. → Published since 1947.

U.S. Bureau of the Census 1909 A Century of Population Growth From the First Census of the United States to the Twelfth: 1790–1900. Washington: Government Printing Office.

U.S. Bureau of the Census 1955 The 1950 Censuses: How They Were Taken. Procedural Studies of the 1950 Census, No. 2. Washington: Government Printing Office.

U.S. Bureau of the Census 1960 a Inquiries Included in Each Population Census, 1790 to 1960. Washington: Government Printing Office.

U.S. Bureau of the Census 1960b The Post-enumeration Survey, 1950: An Evaluation Study of the 1950 Censuses of Population and Housing. Bureau of the Census Technical Paper, No. 4. Washington: Government Printing Office.

U.S. Bureau of the Census 1964 Evaluation and Research Program of the U.S. Censuses of Population and Housing, 1960: Accuracy of Data on Population Characteristics as Measured by CPS–Census Match. Series ER 60, No. 5. Washington: Government Printing Office.

U.S. Bureau of the Census 1966 1960 Censuses of Population and Housing: Procedural History. Washington: Government Printing Office.

U.S. Library of Congress, Census Library Project 1950 Catalog of United States Census Publications, 1790–1945. Washington: Government Printing Office.

Willcox, Walter F. 1930 Census. Volume 2, pages 295–300 in Encyclopaedia of the Social Sciences. New York: Macmillan.

Zelnik, Melvin 1961 Age Heaping in the United States Census: 1880–1950. Milbank Memorial Fund Quarterly 39, no. 3:540–573.

Zelnik, Melvin 1964 Errors in the 1960 Census Enumeration of Native Whites. Journal of the American Statistical Association 59:437–459.

CENSUS

A population census, which usually is just called a census, is a count of the population of a country on a fixed date. National governments conduct censuses to determine population sizes, growth rates, and characteristics (such as sex, age, marital status, and ethnic background) for the country as a whole and for particular regions or localities. Generally governments collect this information by sending a questionnaire in the mail or dispatching an interviewer to every household or residential address in the country. The questionnaire asks the head of the household or a responsible adult living in the household (the respondent) to list all the people who live at the address on a particular date and answer a series of questions about each of them. Over a period of months or years the government census office aggregates and tabulates the answers and reports the results to the public.

Censuses are very expensive and elaborate administrative operations and thus are conducted relatively infrequently, generally at five- or ten-year intervals. Between censuses governments estimate the size and characteristics of the population by extrapolating past trends or drawing on other data sources. Periodic sample surveys are one such source. In the United States, the Current Population Survey of around 50,000 households is conducted monthly. The Census Bureau is planning a new rolling sample survey called the American Community Survey to provide the same level of local area detail available in a decennial census by cumulating and averaging sample estimates over a five-year period.

History

Censuses have been conducted since ancient times. Early censuses were conducted sporadically and generally were used to measure the tax or military capacity of an area. Examples include Roman and Chinese censuses, the Domesday Book, occasional city surveys such as the Florentine Catasto, and records of medieval manors. Unlike modern censuses, they tended to count only adult men, men liable for military service, or tithables (people liable to pay taxes) along with landholdings. The results were used for administrative purposes and were not extensively tabulated or regarded as public records. Nevertheless, historical demographers have derived estimates of total populations from them.

Census taking in the modern sense requires the conception of a uniform, countable unit of analysis. Hence, census taking had to await the development of the state and the emergence of the concept of the commensurate household. The latter occurred in the medieval European west: In the ancient world rich households with large slave labor forces could not be considered "commensurate" with the hovels of the poor. In modern censuses the household or family serves as the unit of analysis or the locus for counting the members within it.

The modern periodic census of all persons is an invention of the early modern period in Europe. One of its purposes was to monitor the progress of overseas colonies. Thus, repeated counts were taken of the colonial American population in the seventeenth and eighteenth centuries, starting in the 1620s in Virginia. In Canada, French efforts to count the population began in 1665–1666 in what is now Quebec, and censuses were conducted at irregular intervals after Canada became a British colony in 1763. Sweden began to conduct censuses in the mid-eighteenth century by tallying the records in its vital registration. England and Wales instituted a regular census on a ten-year cycle in 1801. By the early nineteenth century census taking had begun to be a regular function of government in Western Europe and North America, and in the twentieth century it spread throughout the world.

Functions and Techniques

Censuses serve a variety of purposes in different countries. At a minimum a census provides a measure of the size of the population of a country, which can be compared with the population in the past and the populations of other countries and used to make estimates of the likely population in the future. Governments use census information in almost all aspects of public policy, such as determining how many children an educational system must serve, determining where to put new roads, and providing the denominators of other measures (e.g., per capita income, crime rates, and birth rates and death rates). Private businesses use census data for market analysis in deciding where to locate new businesses or where to advertise particular products. Government agencies and private researchers use the census to provide the "sampling frame" for other types of survey research.

In the United States the census is taken during the tenth year of each decade. The resulting population count provides the data for reapportioning seats among the states in the House of Representatives and the Electoral College and for redrawing district boundaries for seats in the House, in state legislatures, and in local legislative districts. In Canada and many European countries a full census is taken during the first year of every decade. Canada also takes an abridged census during the sixth year of the decade. Canadian population data are used to apportion seats among the provinces in the House of Commons and to draw electoral districts.

Most countries create a permanent national statistical agency to take the census, such as the United States Bureau of the Census or Statistics Canada. This agency usually undertakes a public review process to determine the questions that will be asked. Most censuses ask for basic demographic information such as the age, sex, educational background, occupation, and marital status of an individual. Race, ethnic or national origin, and religious affiliation are important questions in some countries. Other questions often include a person's place of birth, relationship to the household head, individual or family income, type of house, citizenship, movement in the last five years, and language spoken at home. Questions that are routine in one nation may be controversial in another. In the United States questions on religious affiliation are not asked in the census because they are seen as an infringement of the First Amendment right to freedom of religion. Other nations, such as India, do collect this kind of information. Questions on the number of children born to a woman were quite controversial in China in the early twenty-first century because of their connection with the government's one-child policy. A question on income was considered controversial in the United States in 1940, when it was first asked; it is no longer considered problematic.

Questions also change in response to public debate about the state of society. For example, Americans wanted to know which households had radios in 1930 and introduced questions on housing quality in 1940. Canada asks census questions on unpaid work done in the home.

Census taking can be divided into several phases. In the first phase the census agency divides the country into geographic divisions, makes maps and address lists, and prepares instructions for the local census takers. To conduct the count, large numbers of temporary workers may be hired or other government employees, such as schoolteachers, may be called on. The census agency prepares, prints, and mails the questionnaires to households or has them delivered by enumerators.

In the second phase a responsible adult or household head in every household, family, or equivalent entity is asked to fill out the form or respond to the enumerator and supply the required information about each member of the household. (In the current U.S. practice a brief set of questions on a "short form" is asked of all people, usually including name, age, sex, race and ethnicity, and relationship to the household head. A sample of households is asked to complete a more complicated "long" form, which can have many detailed questions on work status, income, housing, educational background, citizenship, and recent moves.)

In the third phase the census agency enters the data into a computer and tabulates the responses for the nation, states or provinces, and cities, towns, and other local jurisdictions. The agency also cross-tabulates the answers, for example, reporting not only the number of people in a local area but the number of people in five-year age cohorts, for each sex, and for local areas. The agency publishes only the tabulated results of the count and keeps the individual responses confidential. In the United States the individual census forms are stored in the National Archives and eventually opened to the public. People then may use them to research the history of their families or construct genealogies.

The choice of census technique for a particular country is the result of its social and political traditions and technological capacities. The U.S. Census is highly automated and is conducted primarily by mail. Canada sends enumerators to deliver the census form to each household, to be completed and returned by the household head. Other nations use more labor-intensive techniques for collecting and tabulating the data, sometimes requiring people to stay home to await the census taker on census day.

The U.S. Census

The U.S. Census was mandated in the 1787 Constitution. This census was the first count in the world designed to provide population figures for apportioning the seats in the national legislature. Direct taxes levied on the states were also to be apportioned on the basis of population. At that time almost 20 percent of the American population consisted of enslaved African Americans. The framers debated whether slaves were "persons" or "property" and thus whether states should receive representation for their slave populations. The framers developed what came to be called the Three-Fifths Compromise, which discounted the size of the slave population as the equivalent of 60 percent of the free population when determining the apportionment of the House. (The abolition of slavery also abolished the compromise, but the tradition of counting the different racial groups in the population continued.)

In the first census, taken in 1790, assistant U.S. marshals were instructed to travel the country and ask six questions at each household: the name of the family head and for each household the number of free white males age 16 and over; the number of free white males under 16, the number of white females, the number of other free people (the free colored), and the number of slaves. The marshals recorded and totaled the figures for the local jurisdiction and sent them to the U.S. marshal for the state, who totaled the figures for the state and sent them to the President. The census counted 3.9 million people.

In later years the census became more elaborate, with more questions asked and more data published. In 1850 Congress mandated a census schedule (form) with a line of questions for each person, including name. A temporary Census Office, as it was then called, was set up in Washington to compile the responses and publish the results. By 1880, when the American population topped 50 million, the census was still compiled by hand, using a primitive tally system. In 1890 the Census Office introduced machine tabulation of the responses, and each person's answers were converted to codes punched into Hollerith cards, a precursor to the IBM punch card. The cards then were run through counting machines. This was the beginning of modern data processing and led to further innovations in tabulating large amounts of data. In the 1940s the Census Bureau commissioned the construction of the first nondefense computer, UNIVAC, to tabulate the 1950 census. In the late 1950s the Census Bureau developed an electronic scanning system called FOSDIC (Film Optical Scanning Device for Input to Computers) to transfer the answers on the census form to a computer.

In 1940 the United States began to collect some census information from a sample of the population and thereafter slowly shifted the detailed questions on the census to the long form sent to only about 15 to 25 percent of households. In 1970 the census became primarily a mail enumeration as the Census Bureau developed automated address files for the country. In the year 2000 over 90 percent of the roughly 110 million residential addresses in the United States received the census form by mail. If the Census Bureau does not receive a response, it sends an enumerator to determine whether the address is correct and to get the information from the household at the address.

Availability of Census Data

Until the 1980s statistical agencies published census results in large volumes of numeric tables, sometimes hundreds of those volumes. Since that time census results have become available electronically on disc, magnetic tape, and CD-ROM or the Internet. Retrospective print compilations of census data are available in libraries, and some have begun to be converted to an electronic format and posted on the World Wide Web. For example, basic population tabulations from American censuses from 1790 to 1960 are available from data compiled by the Interuniversity Consortium for Political and Social Research (ICPSR).

Availability of the original census forms varies by country. With the exception of the forms from the 1890 census, the original schedules for the U.S. censuses survive. If a country's forms are available, historical public use samples of population censuses may exist in electronic form. These samples have been compiled online, including the Integrated Public Use Microdata Series (IPUMS) for American data and planned international compilations that can be viewed on the Internet.

Issues

Censuses can become embroiled in political or social controversy simply by reporting information relevant to ongoing issues in a society. Complaints about the census generally involve concerns about the accuracy of the count, the propriety of particular questions, and the uses to which the data are put.

Censuses require public understanding, support, and cooperation to be successful. Concerns about government interference in private life can prevent people from cooperating with what is an essentially voluntary counting process. People may be wary of giving information to a government agency or may regard particular census questions as invasions of privacy.

When public trust is lacking, people may fail to cooperate. Individuals in illegal housing units, those who are resident in the country illegally, and those who do not wish to reveal their economic or social situation to a government agency are reluctant to respond to a census. In a more serious challenge, some people claim that censuses should not be conducted at all on the grounds that the results will not be held in confidence. In the Netherlands the legacy of the Nazi era, during which census records were used to identify Jews for deportation, was one of the major justifications for ending census taking in 1971.

Some political challenges to the census claim that the census does not count the population well enough. All censuses contain errors of various kinds. People and addresses are missed, and people may misunderstand or fail to answer some questions. Census officials have developed elaborate procedures to catch and correct errors as the data are collected and to impute missing answers from the answers to other questions. Nevertheless, some errors inevitably remain.

Various methods are used to measure the accuracy of censuses. Census results may be compared with population information from other sources, such as the records of births, deaths, and marriages in vital statistics. Commonly, a second, sample count (a postenumeration survey, or PES) is collected shortly after the complete census. Its results are matched against those of the census, allowing estimates to be made of the number of those missed and those who have been counted twice or are in the wrong geographic location. Some nations, such as Canada and Australia, adjust their census results for omissions and other errors.

In the United States, city dwellers, the poor, and minorities tend to be undercounted in the census relative to the rest of the population. Officials representing such undercounted jurisdictions claim that these jurisdictions have suffered loss of political representation and government funding as a result of incorrect data. Litigation seeking to compel adjustment of census results has been unsuccessful in the United States. The question of adjustment has also emerged as a political controversy in Congress: Republicans generally have opposed adjusting for the undercount, and Democrats have supported it. In 2001 the U.S. Census Bureau certified the unadjusted results of the 2000 census as the official results on the grounds that it could not guarantee that the adjusted census results were more accurate than the unadjusted count.

See also: Databases, Demographic; Data Collection, Ethical Issues in; Demographic Surveys, History and Methodology of; Demography, History of; Population Registers.

bibliography

Anderson, Margo. 1988. The American Census: A Social History. New Haven, CT: Yale University Press.

Anderson, Margo, and Stephen E. Fienberg. 2001. Who Counts? The Politics of Census Taking in Contemporary America. New York: Russell Sage Foundation.

Cassedy, James. 1969. Demography in Early America: The Beginnings of the Statistical Mind. Cambridge, MA: Harvard University Press.

Choldin, Harvey. 1994. Looking for the Last Percent: The Controversy over Census Undercounts. New Brunswick, NJ: Rutgers University Press.

Curtis, Bruce. 2001. The Politics of Population: State Formation, Statistics, and the Census of Canada, 1840–1875. Toronto: University of Toronto Press.

Eckler, A. Ross. 1972. The Bureau of the Census. New York: Praeger.

Glass, David. 1973. Numbering the People. Farnborough: Saxon House.

Herlihy, David. 1985. Medieval Households. Cambridge, MA: Harvard University Press.

Seltzer, William. 1998. "Population Statistics, the Holocaust, and the Nuremberg Trials." Population and Development Review 24: 511–552.

Wells, Robert. 1975. The Population of the British Colonies in America before 1776: A Survey of Census Data. Princeton, NJ: Princeton University Press.

Worton, David A. 1997. The Dominion Bureau of Statistics: A History of Canada's Central Statistics Office and Its Antecedents: 1841–1972. Kingston, Ontario: McGill-Queens University Press.

internet resources.

Integrated Public Use Microdata Series (IPUMS). 2002. <http://www.ipums.org>.

Interuniversity Consortium for Political and Social Research (ICPSR). 2002. <http://www.icpar.umich.edu>.

Statistics Canada. 2002. <http://www.statcan.ca>.

U.S. Department of Commerce, U.S. Census Bureau.2002. <http://www.census.gov>.

U. S. Historical Census Data Browser. 2002. <http://fisher.lib.virginia.edu/census/>.

Margo Anderson

CENSUS

An official count of the population of a particular area, such as a district, state, or nation.

The U.S. Constitution requires that a census of the entire population, citizens and noncitizens alike, be made every ten years (Article I, Section 2, Clause 3). The fourteenth amendment to the Constitution directs that the census will be used to determine the number of members of the U.S. House of Representatives from each state. The census is conducted by the U.S. census bureau, an agency established in 1899 within the U.S. commerce department. The data gathered by the U.S. Census Bureau are used by the states to draw boundaries for congressional and state legislative districts, and by local governments to establish districts for other representative bodies such as county legislatures, city councils, and boards of supervisors.

Census data are also used to allocate federal and state funding and services. By the mid-1990s, more than $50 billion in federal aid for education, housing, and health programs to states and cities was distributed annually based on census numbers. In addition, census information is used in academic research and is sought by product manufacturers and marketers who want to know the demographics of potential consumers.

The first U.S. census took place in 1790 when some six hundred u.s. marshals went door-to-door counting approximately 3.9 million people. The 1790 census consisted of fewer than ten questions, which for each household included the name of the head of the family, the number of free white males over and under 16 years of age, the number of free white females, the number of all other free persons, and the number of slaves.

The 1890 census counted 63 million U.S. citizens and reflected a dramatic increase in immigration, urbanization, and industrialization. That census showed that for the first time fewer than half of all U.S. workers were employed on farms. The 1890 census included questions regarding military service during the Civil War, number of years in the United States, naturalization status, reading and writing ability, and mental and physical disabilities.

By 1980, the Census Bureau conceded that the decennial censuses were undercounting portions of the population, usually low-income and minority groups in the inner cities. In follow-up surveys after the 1980 census the bureau determined that it had missed some 3.2 million persons, or 1.4 percent of the population. For example, a 1986 post-census survey of East Los Angeles estimated that the 1980 census missed about ten percent of the Latino community, seven percent of the Asian community, and nine percent of the black community. Census officials determined that overall, nearly six percent of the black and Hispanic populations were uncounted

and less than one percent of the white population.

By May 1987, the Census Bureau had determined that the 1990 census could be adjusted for undercounting by using a technique called a post-enumeration survey (PES). The PES would allow the census to be checked for accuracy by sending census takers back to a given number of households that would be representative of the entire U.S. population and comparing the information gathered with the initial head count. If discrepancies arose, the bureau could make corrections and project them to neighborhoods with similar demographic characteristics.

But in October 1987, officials from the Commerce Department, which oversees the Census Bureau, decided against making any statistical adjustment to the 1990 census. As a result, in 1988, New York, Los Angeles, and several other cities, as well as a number of states and organizations, brought suit in federal district court. They claimed that the secretary's decision not to adjust the 1990 census violated their right to equal protection under the fifth amendment to the Constitution and asked the court to enjoin the census. They also argued that the Commerce Department's actions were politically motivated by a Republican administration that realized that the undercounted population is historically Democratic. The defendants moved to dismiss the complaint, contending that the secretary's decision was not subject to judicial review. In City of New York v. United States Department of Commerce, 713 F. Supp. 48 (E.D.N.Y. 1989), the district court denied the motion to dismiss, holding that the plaintiffs had standing (the legal right) to challenge the census on constitutional grounds and that the court could review the secretary's decision.

Following the district court's decision the parties entered into a stipulation in July 1989 by which plaintiffs would withdraw their motion to enjoin the census and the Commerce Department would reconsider its 1987 decision not to adjust the 1990 census. The agreement required the Commerce Department to conduct a PES of not fewer than 150,000 households as part of the 1990 census in order to produce corrected counts usable for congressional and legislative reapportionment and redistricting. The agreement also required the department to develop guidelines under which the secretary would assess any proposed adjustment. In March 1990 the Commerce Department issued final guidelines. The plaintiffs challenged them in court on the grounds that they were impermissibly vague and were biased against any adjustment to the 1990 census. In City of New York, 739 F. Supp. 761 (E.D.N.Y. 1990), the district court held that the guidelines satisfied the defendants' obligations under the 1989 stipulation. The Census Bureau then began the 1990 census.

The 1990 census employed more than 425,000 workers who gathered information on an estimated 250 million people in 106 million households. For the first time, the Census Bureau combined technology with traditional door knocking, using coast-to-coast computerized maps of all 7.5 million census tracts in the United States. The bureau predicted that these maps would reduce the number of errors caused by census workers' reliance on outdated state and local maps. The census cost some $2.6 billion—65 percent more than the 1980 census—making it the most expensive count ever conducted.

In March 1990, the bureau mailed or hand delivered more than 106 million questionnaires, one to every household in the United States. Most households received a short form consisting of 14 questions covering personal characteristics and housing. One in six U.S. households received a long form with 45 additional questions on topics such as utilities, tax, mortgage, and rent payments; place of birth; ethnic origin; and work habits. From March to June 1990 census workers continued the data collection. The bureau set aside March 20, 1990, as "homeless night." On that night, census takers, many hired from among the homeless population or those who worked with them, visited shelters and low-cost motels from 6:00 p.m. to midnight; counted homeless people on the streets from 2:00 a.m. to 4:00 a.m.; and from 4:00 a.m. to 6:30 a.m. stood outside abandoned buildings, counting those who emerged.

The homeless count caused a great deal of controversy. The 1990 census reported 228,600 homeless persons in the United States, compared with earlier estimates of 500,000 to 3 million. Advocates for homeless persons argued that the Census Bureau had surveyed only a third of the country's cities and counties and had visited only a limited number of locations. The bureau acknowledged that its workers had avoided actually going into hideaways such as abandoned buildings and dumpsters because of safety concerns and admitted that many winter shelters had closed by the time the census was taken in late March. The bureau maintained that its homeless survey was not intended to produce a definitive count of the homeless population.

In October 1990, the Census Bureau issued estimated U.S. population figures of approximately 254 million, based on a tracking of birth, death, and immigration records. In December, the bureau released a final U.S. population tally of some 249 million, based on the actual mailed census questionnaires and house-to-house interviews. The discrepancy between the two sets of numbers indicated that the 1990 census missed some five million U.S. residents.

By December 31, 1990, the bureau reported to the president population figures for each state as well as the number of seats in the U.S. House of Representatives that each state would receive. Between January and March 1991, states with early deadlines for redrawing legislative districts received totals of all persons of voting age, broken down by race. By April 1, 1991, most other states received the voting age and race data. Between April 1991 and 1993, the Census Bureau released statistics compiled from the long forms, including information on income, marital status, disabilities, types of housing, and education.

In April 1991, the bureau announced the results of its PES. Estimates drawn from the PES revealed that the census had resulted in a national undercount of 2.1 percent, or approximately 5.3 million persons out of a total population of approximately 255 million, the largest undercounting in the history of the census. For example, in one south central Los Angeles neighborhood, officials determined that census takers had underreported the number of occupants in 38 percent of 5,800 households. As expected, the undercount was greater for members of racial and ethnic minorities. Hispanics were undercounted by 5.2 percent, Native Americans by 5.0 percent, African Americans by 4.8 percent, and Asian Pacific Islanders by 3.1 percent. The PES-calculated undercount for non-African Americans was 1.7 percent and for non-Hispanic whites, 1.2 percent. Among major cities with high undercounts were Los Angeles (5.1 percent), Houston (5 percent), Washington, D.C. (5 percent), Dallas (4.8 percent), Miami (4.6 percent), Detroit (3.5 percent), and New York (3 percent).

Among the reasons given for the low counts were that certain segments of the population did not believe the Census Bureau's promise that information is confidential and will not be shared with other government agencies such as the Immigration and Naturalization Service (INS), the local housing authority, or the police; did not have addresses and thus were missed because the 1990 census was conducted primarily by mail; lived in urban high-crime areas where census takers were afraid to go door-to-door; were illegal immigrants; feared the government in general; or lacked proficiency in English.

According to the bureau, if the adjusted count were adopted, Arizona and California would each gain a seat in the House of Representatives and Wisconsin and Pennsylvania would each lose one seat. These discrepancies led state officials to renew their plea for an adjustment of the census using the PES.

In July 1991, Secretary of Commerce Robert A. Mosbacher announced his decision not to adjust the 1990 census to account for the estimated five million people undercounted by the census. Mosbacher said that although he was troubled by the undercount of minorities, his decision supported the integrity of the census and that the resulting disadvantage to minorities should not be remedied in the official census. He also expressed concern that adjustment might not improve distribution of representatives among the states and that uncertainty as to the methods of adjustment and assumptions behind them might cause even more dispute about the accuracy of the census.

The plaintiffs in Wisconsin v. City of New York 517 U.S. 1, 116 S. Ct. 1091, 134 L. Ed. 2d 167 (1996), attacked the secretary's decision, contending that it was tainted by partisan political influence and violated the Constitution, the administrative procedure act of 1946, and the 1989 stipulation agreed to by both parties in the case. After a 13-day bench (non-jury) trial, the district court concluded that it could not overturn the secretary's decision (City of New York, 822 F. Supp. 906 [E.D.N.Y. 1993]). On appeal, the court of appeals concluded that, given the admittedly greater accuracy of the adjusted count, the secretary's decision was not entitled to be upheld without a showing by the secretary that the refusal to adjust the census was essential to the achievement of a legitimate government objective (City of New York, 34 F.3d 1114 [2d Cir. 1994]). On appeal, the Supreme Court reversed the decision of the Second Circuit,

holding that the secretary's decision not to adjust the census was within the government's discretion (___ U.S. ___, 134 L. Ed. 2d 167, 116 S. Ct. 1091 [1996]).

By October 1991, at least five state legislatures had filed requests under the freedom of information act (FOIA) (5 U.S.C.A. § 552 et seq.) to see the adjusted census figures in order to decide which set of numbers should be used to redraw state political boundaries. Secretary Mosbacher refused to make the adjusted numbers public, claiming they were flawed and their release could disrupt the redistricting process. In Assembly of California v. United States Department of Commerce, 797 F. Supp. 1554 (E.D. Cal. 1992), California state officials brought an action under the FOIA to enjoin the Commerce Department from withholding computer tapes containing statistically adjusted census data for California. The department claimed that the information was protected from disclosure under an exemption to the FOIA. But the district court said the exemption did not apply to the census data and ordered the Commerce Department to release the tapes. The court of appeals affirmed the district court's order to release the tapes (Assembly of California, 968 F. 2d 916 [9th Cir. 1992]).

In a similar case, the U.S. Court of Appeals for the Eleventh Circuit reached the opposite result. In Florida House of Representatives v. United States Department of Commerce, 961 F. 2d 941 (11th Cir. 1992), the Florida House of Representatives brought a FOIA action to compel the Commerce Department to release all the adjusted census data for Florida. The district court granted summary judgment for Florida and the Commerce Department appealed (Florida House of Representatives, No. TCA 91-40387-WS [N.D. Fla. 1992]). The Eleventh Circuit reversed, finding that the census data were exempted from disclosure under the FOIA. The U.S. Supreme Court declined to review the case (Florida House of Representatives, 506 U.S. 969, 113 S. Ct. 446, 121 L. Ed. 2d 363 [1992]).

In light of the controversy over the 1990 census, government officials and demographers debated how best to conduct the census in 2000 and later. Many demographers argued that the U.S. population had become too mobile and too uncooperative to allow reliance on mail-in-surveys and door-to-door interviews. An increase in the number of non-English speakers, undocumented immigrants, and homeless persons makes census taking more difficult and residents will become more diverse and less tolerant of government intrusion in the future. The American Statistical Association urged the government to use scientific sampling surveys to estimate the population that has been the most difficult to count.

In preparation for the 2000 census the bureau conducted a test census in the spring of 1995 at three sites—Paterson, New Jersey; Oakland, California; and six parishes in northwestern Louisiana. The sites were selected because of their ethnic diversity and their large number of multidwelling housing units. In Paterson the bureau experimented with a multimedia kiosk, which allowed residents to answer census questions by touching a screen. In Oakland all identified households were sent a census form and blank forms were also made available at libraries, post offices, and the state department of motor vehicles. The bureau also experimented with using statistical samples from random surveys to estimate total population.

From these test projects the Census Bureau announced that it would use statistical sampling to take into account historically undercounted populations. These populations included minorities, renters, children, poor persons, and illegal aliens. Although the American Statistical Association supported this approach as a valid methodology, the announcement set off a political firestorm. Congressional Republicans, worried that sampling would lead to congressional apportionment that favored the democratic party, filed a lawsuit challenging the constitutionality of the proposed practice.

The Supreme Court, in Commerce Dept. v. U.S. House of Representatives, 525 U.S. 316, 119 S. Ct. 765, 142 L. Ed. 2d 797 (1999), ruled, in a 5–4 decision, against the use of statistical sampling, holding that the 1976 amendments to the Census Act (1954) prohibit the use of statistical sampling for purposes of population head counts. Justice sandra day o'connor, writing for the majority, stated that there had been over two hundred years of history "during which federal census statutes have uniformly prohibited using statistical sampling for congressional apportionment."

The 2000 census revealed that the U.S. population had grown to approximately 281 million. There was little public controversy over the results, a sharp contrast to the 1990 census. However, one state did file suit in an attempt to throw out census figures derived from a method the state considered impermissible sampling. Utah, noting that its population grew by 30 percent in ten years, was disappointed it did not gain another seat in the U.S. House of Representatives. In reviewing the census data it noted that the Census Bureau had relied on a statistical method called imputation to estimate the number of members of a household that census takers could not contact after repeated efforts. Utah discovered that if it could have these imputed numbers removed from the population count it would gain a House seat that had been awarded to North Carolina.

A three-judge panel heard Utah's lawsuit but dismissed it at the urging of the Commerce Department. The panel ruled that the imputation method was not impermissible under the 1999 Supreme Court decision and that it did not violate the Constitution's Census Clause. The Supreme Court, in Utah v. Evans, 536 U.S. 452, 122 S. Ct. 2191, 153 L. Ed. 2d 453 (2002), upheld the lower court ruling. The Court, in a 5–4 decision, dismissed Utah's contention that actual enumeration under the Census Clause was intended as a description of the only methodology for counting U.S. citizens. As for the imputation method, the Court saw it as different from sampling: "sampling seeks to extrapolate the features of a large population from a small one, but the Bureau's imputation process sought simply to fill in missing data as part of an effort to count individuals one by one."

The Census Bureau has established at its website (<http://www.census.gov>) a portal for accessing all 2000 census data. The site provides researchers with tables of data while also providing the public with breakdowns of data in easily searchable formats.

further readings

Anderson, Margo J. 1990. The American Census: A Social History. New Haven, Conn.: Yale Univ. Press.

Katz, Bruce, Robert Lang, and Franklin Raines. 2003. Redefining Urban and Suburban America: Evidence from Census 2000. Washington, D.C.: Brookings Institution.

U.S. Census Bureau. Available online at <www.census.gov> (accessed May 28, 2003).

cross-references

Apportionment.