the projects now handled by the Census Bureau. It was not surprising that the overwhelming response was "Yes." I was impressed, however, with the reasons given for the belief that more Federal survey work should be shared with private enterprise. Let me provide actual quotes from several representative letters. Hon. JACKSON E. BETTS, PRINCETON, N.J., September 8, 1967. DEAR CONGRESSMAN BETTS: I am keenly interested in the problem you pose. The government can conduct some types of surveys better than private marketing research firms, such as ours. But, not from the point of view of technology. The decennial census would probably represent too great a project for any or all private firms. Also, there are some studies made by the government, requiring an answer by the respondents, that obviously could not be done by private marketing firms. On the other hand, there are a great many surveys that not only could be done as well by private firms, and at a fraction of the cost. If all expenses were honestly taken into account, I believe that private research firms could conduct most of these surveys for about one-third or one-half of the amount spent by the government. My own fear is that the government will continue to build up its survey machinery and in time will take over more and more of the work that now goes to commercial research firms. From the point of view of quality, the government certainly has no advantage except that it can spend excessive amounts whereas commercial firms must normally work within much smaller budget requirements. As the head of a market research firm, I should add that we have never been very much interested in getting survey assignments from the government because of the time and the money and the difficulties of fighting the Washington bureaucracy. It requires a lot of time of a lawyer, of accountants, etc., even for minor contracts. Moreover, it is always difficult to get money out of the government when the job is completed. In dealing with private industry, a simple contract is enough and we can expect to be paid on the complemention of the study and the delivery of the report. In our experience this is not the case in dealing with the government. Often months go by before some minor detail is cleared up. All of this makes for work for the bureaucrats but it is discouraging to those who could do surveys for the government at substantial savings to taxpayers. Sincerely, GEORGE GALLUP, Director, American Institute of Public Opinion. NEW YORK, N.Y., September 5, 1967. Congressman JACKSON E. BETTS, DEAR CONGRESSMAN BETTS: It is my opinion that private marketing research organizations can complete favorably for the studies undertaken by The Census Bureau for the benefit of government. This has application to studies in other areas of government as well. The Department of Agriculture has demonstrated the success of this approach in a number of studies in which they have utilized private organizations. Most sincerely, A. EDWARD MILLER, President, Alfred Politz Research. RICHMOND, VA., August 14, 1967. Hon. JACKSON E. BETTS, DEAR MR. BETTS: We cannot admit to an unprejudiced viewpoint, but are convinced that private marketing research organizations could successfully contract some of the census projects. Our organization has, from time to time, attempted to persuade the Department of Commerce that this could be done-but to no avail. Insofar as I am aware, the Commerce Department is one of the few in the federal government which relies almost exclusively on its internal personnel for information gathering and other kinds of research. Sincerely yours, J. ALBERT FABER, President, Southeastern Institute of Research, Inc. URBANIA, ILL., September 18, 1967. Hon. JACKSON E. BETTS, DEAR SIR: You asked my opinion on the matter of having private market research firms undertake some of the projects now being down by the Census Bureau. I am quite confident that any one of a number of private research firms could do such work and perhaps do it as effectively as a government agency. I would guess that in a number of instances the contract price for research done by private firms would be lower than the cost of having it done by a government agency. This would be particularly true for special projects where a government agency is not already tooled up to conduct such research investigations. Sincerely, C. H. SANDAGE, President, Farm Research Institute. WATERLOO, Iowa, September 19, 1967. Hon. JACKSON E. BETTS, Washington, D.C. DEAR MR. BETTS: As to your second question: yes, certainly, in my opinion, private interviewing organizations could successfully contract many of the Census Bureau surveys. Sincerely yours, GLADYS L. WALKER, Director, Black Hawk Research Bureau. DETROIT, MICH., August 1, 1967. Hon. JACKSON E. BETTS, Washington, D.C. DEAR MR. BETTS: We firmly believe that the Federal Government is in fact in competition with us along with many tax supported universities and also, in fact, we are being deprived of potential business which would yield the government additional tax revenue. Respectfully yours, RICHARD W. OUDERSLUYS, President, Market-Opinion Research Co. Mr. Speaker, one approach to an objective review of the polls or surveys which might be contracted to private organizations was presented by Mr. Henry Brenner, of the Home Testing Institute: Hon. JACKSON E. BETTS, Washington, D.O. MANHASSET, LONG ISLAND, N.Y., August 16, 1967. DEAR MR. BETTS: I would like to suggest a possible course of action to you. 1. To get from the Bureau of the Census and other Government bodies (Department of Agriculture, for one) that conduct "marketing research" surveys a list of the studies that have been completed during a recent twelve month period along with a description of: (a) The purpose of the study; 95-144-68-5 2. Then I would suggest that this material be forwarded to 15 or 20 heads of marketing research firms or other individuals involved in non-governmental marketing research activities. 3. Those who wish to would then be given an opportunity to discuss with you and other interested Congressmen and officials of Government bodies involved the capability of marketing research firms to undertake studies similar to those conducted by the Government bodies. Cordially, HENRY BRENNER, Home Testing Institute/TvQ, Inc. Mr. Speaker, the officers of the major trade associations for the marketing and opinion research industry should be asked to provide suggestions on the formation of industry-academic committees to advise Congress, the Bureau of the Budget or specific agencies. There is a need for a better understanding of where lines are drawn separating federally conducted research efforts from those contracted with private firms. Such a determination might expand, decrease or not materially affect the amount of Federal research work. This would not be the objective of such a determination. It would be to set to a standard everyone understood and to follow that course in future decisions. Fourth. A broad, inclusive program of Federal statistical research must include nonprofit research organizations and universities. These institutions, important statistical users amass much information, also develop new research technology. It is true that some new techniques and methodology are now shared and this should continue. A closer relationship between these loci of vast information should not become one of domination by the Federal Government or reluctant cooperation from the nongovernment sector. This is a relationship to be considered as part of an overall national statistical plan. Mr. Speaker, there are many ramifications of any commentary on improving the information gathering apparatus in the United States. I feel my remarks as they pertain to the Census Bureau and the 1970 census questionnaire lend support for the passage of H.R. 10952. I will let the experts consider the other aspects of my suggestions to strengthen the vital information sources in this country. (The following statements of Hon. A. Ross Eckler. Director. Bureau of the Census; John H. Aiken, executive director, Federal Statistics Users' Conference; and Hon. Arnold Olsen, a Representative in Congress from the State of Montana; were received by the subcommittee for inclusion in the record.) STATEMENT OF A. Ross ECKLER, DIRECTOR, BUREAU OF THE CENSUS I appreciate the opportunity to testify regarding H.R. 10952, a bill to amend Title 13, United States Code, to limit the categories of questions required to be answered under penalty of the law in the decennial censuses of population, unemployment, and housing. I recommend that the bill be disapproved. Its enactment into law would constitute a clear reversal and retrogression from a policy position the Congress has expressed repeatedly over the 177 years of census history in this country. It would devalue the significance and importance of the national census at a time when its results are more critically needed than ever before. It would seriously interfere with the conduct of the census, and it would jeopardize the information to be obtained. The proposed act seems to reflect the misapprehension that the requirement to answer all census questions is one which has been newly, and improperly, added to census-taking. A brief look at the history of the census shows that this is not the case. The requirement that all census questions must be answered goes back to the first census of the United States, in 1790. It has been incorporated into every census act which has been passed since that time, and in the present codification of Title 13, United States Code, it applies not only to the censuses of population, unemployment, and housing, but also to the censuses of manufactures, mineral industries, business, transportation, and agriculture, and to many major surveys as well. On March 1, 1790, the Congress adopted an act providing for the enumeration of the inhabitants of the United States. Section 6 of that act reads as follows: "And be it further enacted, That each and every person more than sixteen years of age, whether heads of families or not, belonging to any family within any division of a district made or established within the United States, shall be and hereby is, obliged to render to such assistant of the division, a true account, if required, to the best of his or her knowledge, of all and every person belonging to such family respectively, according to the several descriptions aforesaid, on pain of forfeiting twenty dollars, to be sued for and recovered by such assistant, the one-half for his own use, and the other half for the use of the United States." In 1850 and again in 1880 the original penalties were strengthened, and penalties have been included in every subsequent repassage and codification of census law. The record of the intent of the Congress that mandatory reporting is required for the Nation's basic statistical program is repeated and clear. The other nations of the world recognize the same necessity. They require mandatory reporting by their citizens in their national censuses. Along with authority goes responsibility for using it rationally. In connection with the most recent decennial census, only two persons were prosecuted for wilful failure to comply with the law; both were fined and both verdicts were upheld on appeal. To my knowledge no person has ever been imprisoned for failure to cooperate in the census. Later in my statement I will discuss the reasons for needing the authority despite the moderate and infrequent exercise of it in actual prosecution. A more specific implication underlying the proposed legislation is that the content of the decennial census has become more burdensome, involves a greater demand on the respondent, and has tended to include questions which are of such limited importance that a low response would not be serious. This implication is not valid. From the very beginning, the census was not confined simply to a count of the number of persons. With your permission, I would like to make available for the record a list showing the items which have been included in each of the decennial censuses. The Congress decided not to limit the 1790 census to the bare essentials which were needed for apportionment. After considerable debate in the Congress the Federal marshals were instructed: “*** to cause the numbers of the inhabitants within their respective districts to be taken; omitting in such enumeration Indians not taxed, and distinguishing free persons, including those bound to service for a term of years, from all others; distinguishing also the sexes and colours of free persons, and the free males sixteen years and upward from those under that age *** The additional information was apparently desired for the purpose of ascertaining the military and industrial strength of the country. The realization of the value of the census as a means of obtaining essential and accurate information about the nation appears again and again in the debates about later censuses. The determination of the questions to be included in the census is a long process in which the needs of government and the general public are carefully weighed. A question is included only if it is of substantial value for fulfilling significant purposes. The needs of the Goverment are given primary consideration. Questions which have proven their value in the past are reexamined to make sure that they are still needed. The list of questions for 1970 is expected to follow the 1960 list very closely. In its hearings on mid-decade census legislation, the Subcommittee heard extensive testimony concerning the uses that are made of census data and the many decisions which are influenced by such data. They are used extensively in developing, planning, administering, and evaluating programs by Federal, State, and municipal and other government agencies. Census data are used not only in apportionment and districting for national and State legislative bodies, but also for distribution of Federal and State funds, and for a large number of administrative actions. Census data are vital in planning, administering, and evaluating public programs relating to employment, education, health, poverty, urban renewal, and many others. Many government agencies have a need for statistics for specific small areas, such as those in the central cities, rural problem areas, river basins, school districts, State legislative districts, and others. Their need for information is not limited to head counts but covers the range of census data. The census must be prepared to provide adequate statistics for all such areas. To omit some or to provide only partial data for some would be to fail in one important task which has been assigned to the Census Bureau. Demands made on the Bureau by the Congress and by administrative agencies in the summer of 1967 provide additional illustrations of the extensive use made of census data. The Government Operations Committee of the Senate, the National Advisory Commission on Civil Disorders, and many other agencies concerned with the urban violence problem have eagerly sought and used such data. Uses by nongovernmental interests and agencies are also considered, but no question is included in the census solely because some private profit or nonprofit group has asked for it. All of the questions included in the census are important. If a question is not important, it should be excluded rather than included but downgraded by being placed on a voluntary basis. A third implication of the proposed legislation is that the Government has no basic right to require replies to census questions other than those necessary to carry out the constitutional mandate for a decennial census. On this point, I call your attention to a decision emphatically affirming the constitutionality of the census act as made by Judge Edward B. Thomas, United States District Court, 2nd Circuit, Southern District of New York, on February 20, 1901, in the case of United States vs. Moriarity (106 F. 886). Judge Thomas said: "It would be curious governmental debility that should incapacitate the Nation from directing its census enumerator to ask an inhabitant concerning his business because for certain purposes he was only to be counted, and perhaps his gender ascertained. The functions vested in the National Government authorize the obtainment of the information demanded by section 7 of the census act, and the exercise of the right befits an exalted and progressive sovereign power, enacting laws adopted to the needs of the vast and varied interests of the people, after acquiring detailed knowledge thereof;" "For the National Government to know something, if not everything, beyond the fact that the population of each State reaches a certain limit, is apparent, when it is considered what is the dependence of this population upon the intelligent action of the general Government." "If one shall select any of the powers reposed in the Federal Governmentand consider thoughtfully the direct essential aid that the information, in whole or in part, sought by the Act, must give, he will not hesitate to ascribe plenary ability to Congress to obtain such information. A Government whose successful maintenance depends upon the education of its citizens may not blindly legislate, but may exercise the right to proclaim its commands, after careful and full knowledge of the business life of its inhabitants, in all its intricacies and activities." Some of the persons who have been critical of the content of the census questionnaire have drawn particular attention to the questions on housing and on household equipment. A Census of Housing is taken in response to a mandate from the Congress. When the legislation for a Census of Housing in 1940 was approved, the language specifically provided for a census of "housing (including utilities and equipment)." This language was retained in later versions of the basic census legislation. As to personal privacy, it has long been recognized that the requirement that residents provide information about themselves carried with it the obligation to protect that information from misuse. Early in the development of the United States census, the Congress took steps to make sure that the information collected would not be used for any purpose other than the compilation of statistical totals. Title 13, U.S. Code, recognizes this obligation specifically and provides penalties for any employee who makes any wrongful disclosure of any information coming to him by virtue of his employment in the Bureau of the Census. The penalties for such disclosure are more severe than those provided in the case of refusal to answer questions, or the giving of false information. Finally, I should like to discuss some of the potential problems the proposed legislation would create in the execution of an effective, meaningful national census, and the serious effects this legislation would have on the validity of the census results. It would be impossible to carry out an acceptable census under the conditions which would be created if this legislation were enacted. There would be numerous problems of incomplete returns, respondent confusion, and follow-up visits to householders if an effort were made to conduct a census partly on a mandatory basis and partly on a voluntary basis. This would result in increased costs, delays |