Urban Conference Says Undercount of Non-Whites Deprives Minority Rights

(Last month a group of 58 social scientists and government officials, sponsored by the Harvard-M.I.T. Joint Center for Urban Studies, met to consider ways of improving the enumeration of non-white minority groups, such as Negroes, Puerto Ricans, and Mexican-Americans in the 1970 Census.

The two-day Conference on Social Statistics and the City noted that the Census was explicitly designed to provide the basis for representation in the House of Representatives, but had come to provide the basis on which Federal funds are allocated in many areas.

The most significant paper presented at the Conference was by Jacob Siegel, chief of the Census Bureau's National Population and Estimates branch. He showed that one-tenth of the Negro population was missed in the 1960 Census and that among young Negro males the rate was as high as one in six.

Daniel Patrick Moynihan, director of the Joint Center, spoke for the Conference in a letter to Carl D. Perkins, chairman of the House Education and Labor Committee: "Such substantial undercounting of minority groups raises a new order of question. The census count is the basis on which legislation is apportioned and by which great sums of public funds are allocated. A member of a group significantly underrepresented and concentrated in special neighborhoods might well claim a violation of his Constitutional right to equal representation and protection of the law."

Earlier this week Siegel said that the Census Bureau was now planning to use "irregular, unusual devices" in the 1970 Census to overcome the short count of non-whites. He said the new tactics would include enlistment of shopkeepers, canvassing during early morning hours, and the saturation of some areas with census takers. The Bureau also intends to change its philosophy in ghetto areas by counting a maximum of residents instead of seeking to avoid duplication.

The following resolutions were passed by the Conference)

Improving Enumeration of Negroes, Puerto Ricans, and Mexicans

While American population statistics are among the very finest in the world, papers presented to the Conference have established beyond reasonable doubt that the Decennial Census, the Current Population Survey, and to a lesser degree, the Vital Statistics of the United States, seriously and significantly under-enumerate or under-estimate the size of the Negro, Puerto-Rican and Mexican-American populations. As much as 10 per cent of the Negro population may not have been counted in the 1960 Census, and there is considerable probability that the Puerto Rican and Mexican-American were similarly under-counted.

In 1960 as many as one Negro male in six within the age group of 20 to 39 years may have been omitted altogether.

In a modern society statistical information is not only a primary guide to public and private actions, in itself it profoundly influences patterns of thought and basic assumptions as to the way things are and the way they are likely to be. Were national statistics merely inadequate, but uniformly so, the nation would be at a disadvantage, but no special injury could be claimed by any region or group. As it happens, however, where American population statistics are inadequate, they will normally be found to be so in terms of the under-enumeration and under-estimation of minority groups, defined in terms of race, or national origin, and concentrated in specific neighborhoods, usually in densely populated central city areas. They are also, characteristically, defined by poverty. But a larger issue than simply of efficiency and convenience must enter the consideration of this subject. A constitutional issue enters.

Article I, Section 3, of the Constitution provides for the enumeration once each ten years for all persons residing within the United States. This enumeration is explicitly and primarily designed to provide the basis for representation in the House of Representatives. In the years since the adoption of the Constitution, the Census enumeration and other statistical programs, such as those conducted by the Bureau of Labor Statistics, have come to be the basis for a host of public activities, and most particularly provide the basis on which public funds are allocated in a whole range of government programs at the national, state, and local level. In some cases funds are allocated on a straight "head count" basis. In other cases, as for example the Economic Development Act, Title I of the Elementary and Secondary Education Act, and in the Model Cities legislation, public funds are allocated on the basis of population together with a range of special economic and other information collected by government agencies.

The record of these agencies, notably the Bureau of the Census, in gathering and compiling this information with the highest technical competence, the uttermost standards of impartiality and integrity, and at the most modest cost is a matter of national pride. Typically it has been the Census Bureau itself that has been the most diligent in discovering and analyzing the problems of gathering statistics relating to minority groups.

Nonetheless, the problem of underenumeration of minority groups is likely to persist unless it becomes a matter of more general concern. We believe that what were initially at least technical problems have by their very magnitude been transformed into social problems with powerful legal and ethical implications. Specifically, we hold that where a group defined by racial or ethnic terms, and concentrated in specific political jurisdictions, is significantly under-counted in relation to other groups, then individual members of that group are hereby deprived of the constitutional right to equal representation in the House of Representatives, and by inference in other legislative bodies. Further, we hold that individual members of such a group are thereby deprived of their right to equal protection of the laws as provided by Section I of the 14th Amendment to the Constitution in that they are deprived of their entitlement to partake in Federal and other programs design-for areas and populations with their characteristics.

Injury, while general, is real; redress is in order. This would seem a matter of special concern to the nation in view of recent Supreme Court rulings establishing the one-man, one-vote principle in apportioning legislatures, and in view of the extensive Congressional activity in the establishment of programs designed to improve the economic and social status of just those groups that appear to be substantially under-represented in our current population statistics.

The Bureau of the Census and other government statistical agencies have set a superb standard of public accountability in calling attention to this problem themselves. We feel it is incumbent on the Congress to provide the Bureau of the Census, the National Center of Health Statistics, the Bureal of Labor Statistics and such other agencies as are concerned, with the funds necessary to obtain a full enumeration of all groups in the population, and also to gather the usual information on special and economic characteristics that is necessary to implement the laws of the nation.

The Conference likewise emphasizes that there is an obligation on the part of every resident to be enumerated.

The Conference commends the Bureau of the Census for the innovative use of the Post Office for the purpose of insuring a more complete enumeration. The Conference believes that the Post Office will contribute to the improvement of the coverage of the census not only in general, but particularly, in the central cities.

The Conference wishes to encourage the Bureau of the Census to explore more flexible personnel procedures from the standpoint first, of enlarging the number of people who can be located to serve usefully as regular enumerators and, second, to employ people in various auxiliary roles to help insure completeness of count.

The Bureau of the Census, working in concert with other groups, public and private, should work to develop enumerator skills, particularly for conditions in the central cities, and to instill a professional spirit among enumerators.

The Conference suggests consideration of the appointment of enumerators or supervisors by Presidential commission as a step to impress them with the seriousness of the task.

The Conference wishes to encourage the Bureau of the Census to take an experimental approach toward the use of various incentives for respondents.

The Conference is impressed with the extent to which improvements in close-out procedures (Close-out procedures determine the number of calls an enumerator must make at a particular household in an attempt to gather information. If no information can be gathered after the specified number of calls, characteristics of the household are allocated by a computer) and procedures for following back to apparently vacant housing units may contribute to the reduction of under-enumeration. We put high priority, therefore, on changes in procedures and in allocation of resources that promise an improvement in this area.

The Conference would like to urge financial support for further studies of under-enumeration in the 1970. Census such as the study of 1960 under-enumeration by Jacob Siegel, extended to specific estimates of under-enumeration by age, sex, race, ethnic group, and residence.

Improving Vital Statistics for Negroes, Puerto Ricans, and Mexicans

We recommend that the Bureau of the Census and the National Center for Health Statistics study methods for improving the completeness and quality of Census data to enable the computation of vital rates for minority groups, particularly those groups which have been included with the white population in the past.

We recommend that the National Center for Health Statistics through the mechanisms of the Public Health Conference on Records and Statistics and the American Association of Vital Registrars and Public Health Statisticians ask state and local health departments to make more tabulations and classifications of vital records for minority groups.

We recommend that the National Center for Health Statistics make known material state and local health agencies collect, tabulate, and publish concerning vital rates for minority groups.

We recommend that the Bureau of the Census and the National Center for Health Statistics conduct a birth registration and enumeration matching test in conjunction with the Census of 1970. The aims of this test should be: test completeness of birth registration; ascertain census under-enumeration of the young; determine quality of both birth registration and census data; study differentials in infant mortality by characteristics in family or household.

We encourage the National Center for Health Statistics to work through the Public Health Conference on Records and Statistics and the American Association of Vital Registrars and Public Health Statisticians to assure that a question concerning legitimacy will appear on the birth certificate and certificate of fetal death. This can be accomplished by having this item in the confidential section as recommended for the standard certificates for 1968.

We recommend that the National Center for Health Statistics encourage record linkage studies utilizing birth, fetal death, death, marriage and divorce records. We recommend that the Bureau of the Census encourage record linkage studies based on the 1970 Census of Population and Housing.

We encourage states to retain the question concerning race on the birth and death certificates and we encourage states to include questions on race on marriage and divorce certificates. This can be accomplished by having this item in the confidential section as recommended for the standard certificates of 1968.

We urge the National Center for Health Statistics to continue and expand studies of the completeness of death registration and studies of the quality of data obtained on death certificates.

We urge the National Center for Health Statistics to encourage local agencies to classify both birth and death certificates by city block or census tract, perhaps using the address register being developed by the Bureau of the Census for the Census of 1970. This will make possible further studies of the vital rates of minority groups.

We support efforts to obtain a quinquennial census and we recommend that such a census obtain adequate information to permit study of the vital rates of minority groups.

We urge states to affiliate as rapidly as possible with the Marriage and Divorce Registration Areas established by the National Center for Health Statistics.

We encourage the National Center for Health Statistics and the Bureau of the Ceneus to explore, support and conduct surveys of fertility expectations and performnace. In particular, adequate information should be obtained for minority groups with distinctive fertility patterns.

Needed Enlargement in Available Social Statistics for Negroes, Puerto Ricans, and Mexicans

The Conference recommends that the Census Bureau continue to clearly identify Negroes, Puerto Ricans, Cubans, Mexicans, white persons of Spanish surname, and American Indians in the Census. For the specific groups for which more data are desired Census publications should replace the categories of white and non-white with more specific ethnic or racial designations; for example, Negro, American Indian, and Cuban.

Federal statistical surveys should be designed from the beginning to provide data for specific minority groups. This may require special sampling and tabulation programs in addition to specifications of content appropriate to the social situation of minorities.

The Bureau of the Census and other statistical agencies should continue to experiment with methods of identifying ethnic groups; for example, by ethnic origin, language spoken in the home and birthplace of grandparents.

The presentation of survey and census results should be expanded in two ways--in summary form for general users and in easily available special tabulations for professional users.

The Conference recommends that the Census Bureau conduct frequent surveys to provide for individual cities and the minority groups within cities data of the type included in the program of the Current Population Survey.

A full census should be carried out more frequently than at present. A 5 year census should replace the present 10 year census.

Apprising Organizations About the Census and other Sources of Social Statistics

The Joint Center for Urban Studies of M.I.T. and Harvard should apprise organizations interested in social statistics, particularly those concerned with minority groups, of the proceedings of this Conference