Relation of the Federal Government to Research 



cent years results from his determination to hold to 

 high scientific standards. 



Intercensal Work and Isolation 

 From the Administration of Laws 



When tlie Bureau of the Census was established in 

 1902 there was some expectation that the Bureau would 

 become the central statistical agency of the Govern- 

 ment," free from the responsibility of administering 

 any laws other than laws to enforce the collection of its 

 data. 



Failure to Become Central 

 Statistical Agency 



The Bureau has been kept free from administrative 

 responsibilities but, in view of the growth of other 

 statistical agencies, it can hardly be called the central 

 statistical agency — at least, in the sense, that it is the 

 place to which new statistical inquiries are generally 

 allocated. The freedom from administrative respon- 

 sibilities should, in theory, be an asset from the scien- 

 tific standpoint. Indeed, one basis of the respect for 

 the Bureau's reports is its guarantee to respondents 

 that original schedules are confidential and cannot be 

 used by other agencies, for taxation, rate-making, po- 

 licing, or other regulatory purpose. Yet, however 

 sound in theory may be the separation of such a statis- 

 tical agency from administration, there are many 

 disadvantages. 



One disadvantage is a public and, therefore, con- 

 gressional apathy toward such an agency. It is not 

 easy to arouse public opinion over the importance of 

 statistics. The beneficiaries of statistics are not or- 

 ganized in pressure groups — seldom does a statistical 

 bureau receive expressions of "consumer appreciation." 

 The consumer relationships will be described more in 

 detail in the next section of this report. 



The intrinsically weak strategic position of the Bu- 

 reau of the Census, as a purely statistical agency, for 

 obtaining the backing of influential pressure groups 



" "The act of February 14, 1003, establishing the Department of 

 Commerce and Labor transferred the Census Office to that Department 

 from the Department of the Interior [?,2 Stat. L. 82."i). It was evident 

 that Congress contemplated the possibility of developing the Census 

 Office into a central statistical service, for in the section of the law 

 transferring the Office, the Secretary of Commerce and Labor was given 

 power to rearrange the statistical work of the bureaus and offices in 

 his Department and even to consolidate them. The opportnnlile« for 

 concentrating statistical work in the Census Office were greater than 

 this provision Indicated on its face, for another section of the same 

 law authorized the President to transfer to the Department of Com- 

 merce and Labor from any of the other departments except the De- 

 partment of .Agriculture the whole or any part of any office, bureau, 

 or division, engaged in statisticil work." W. Stull Holt, The Bureau 

 of the Census, Service Monograph of the United States Government, 

 No. 53 (Washington, The Brookings Institution, 1020 j. 

 122099—30 14 



201 



is all the weaker because of the existence of strong sta- 

 tistical agencies closely attached to the agencies charged 

 with the administration of laws. When vast sums are 

 approjiriated to finance new legislation, considerable 

 amounts often must be allocated to statistical studies. 

 Sometimes the allocation is specified in the legislation; 

 often the statistical research is apparently not con- 

 templated by Congress, but must be made, out of the 

 general appropriation for a given undertaking, if 

 the undertaking is to succeed. In neither case is the 

 Bureau of the Census likely to be a beneficiary. The 

 statistical work is almost invariably done in a statis- 

 tical unit attached to the regulatory agency. Thus, 

 statistical work in connection with agriculture usually 

 goes to the Bureau of Agricultural Economics, and 

 statistical work in connection with labor to the Bureau 

 of Labor Statistics. 



In the field of manufactures and trade the Bureau 

 of the Census has managed to get some recognition as 

 performing, in a specific area of governmental interest, 

 functions comparable with those of the Bureau of 

 Agricultural Economics, or the Bureau of Labor Sta- 

 tistics. Even here the Bureau runs head-on into com- 

 petition with other agencies, including one within its 

 own Department, the Bureau of Foreign and Domestic 

 Commerce. 



If, from its beginning as a permanent agency, the 

 Bureau of the Census had been recognized as the place 

 to which most statistical work associated with new 

 legislation would be assigned, the Bureau would have 

 occupied an incomparably stronger position for secur- 

 ing financial support. Instead, the Bureau has been 

 passed by when the huge appropriations to support 

 expanding functions of Government have been made. 

 It has been forced to depend, not on the indirect revenue 

 out of such appropriations, which are backed by strong 

 public opinion, but upon appreciation by Members of 

 Congress of the broad governmental and scientific value 

 of the services and reports of the Bureau of the Census. 



The Bureau's Intercensal Work from 1902 On 



In order to visualize the nature of the intercensal 

 work of the Bureau of the Census, it is necessary to 

 trace briefly the historical background of the Bureau's 

 great miscellany of tasks, from its establishment in 

 1902. 



The inquiries of the Bureau of the Census touch 

 almost every field of social and economic life. Though 

 diverse and comprehensive in scope, the inquiries em- 

 brace only a fraction of the statistical information 

 collected by the Government in some fields. The 

 breadth of coverage is largely explained historically 

 by the fact that most of the intercensal work of the 



