Relation of the Federal Government to Research, 



219 



the Bureau than in younger statistical organizations, 

 which will confront the same difficulties eventually, 

 unless there is constant expansion. One-third of the 

 clerical employees in the Bureau are over 55 years of 

 age, thus greatly reducing efficiency in performance 

 of tasks which put a premium on agility. 



A source of some of the Bureau's difficulties in the 

 past has been the size of its temporary force. Even in 

 intercensal periods the temporary force is large and 

 fluctuating, and at the peak of the decennial census 

 period the temporary force exceeds 125,000. The tem- 

 porar}- field people are frankly political employees in 

 large part. The Washington temporary staff is largely 

 on civil service. The most serious consequence of this 

 situation is perhaps less the real inefficiency of these 

 political employees than it is the danger to which the 

 whole morale of the Bureau is exposed. The primary 

 effect has been felt in influencing the choice of Director, 

 who often in the past has been named as a reward for 

 party service. The present administration of the Bu- 

 reau has made substantial progress in repairing some 

 of the harm done to the Bureau in the past. But the 



fact that the Bureau, by tradition, has available a largo 

 number of political jobs of a temporary character, does 

 not make the future secure against the appointment of 

 subsequent directors who lack the respect for scientific 

 work which characterizes the present leadership. 



Mention was made of the educational work within 

 the Bureau, including courses taught by Bureau ofli- 

 cials for which university credit is given. The Bureau 

 also is hospitable to graduate students who desire to 

 receive some practical statistical training. 



As we have seen, some serious handicaps to the im- 

 provement of personnel are beyond the control of tlie 

 Bureau. It is extremely important, however, for the 

 Bureau to have a will for improvement. This will 

 has not always been present in the administration. It 

 is present now. Slowly, but unmistakably, this will is 

 finding expression in personnel policies which are 

 changing the atmosphere of the organization. How 

 lasting these changes will be depends to a considerable 

 extent on the extrication of the Bureau, politically and 

 financially, from the insecure position whose historical 

 background was traced in part I of this Section. 



IV. ANALYTICAL RESEARCH 



It doubtless would be generally agreed that the fii'st 

 great function of such an agency as the Bureau of the 

 Census is to collect, tabulate, and publish basic data, 

 which serve as raw material for further processing by 

 research workers everywhere, in the Government and 

 outside of the Government. The Bureau's first respon- 

 sibility is to mine the statistical ore and produce the 

 statistical pig iron. 



How much farther does the Bureau's responsibility 

 extend ? 



To what extent does it, or should it, conduct analj^i- 

 cal research and fashion the statistical pig iron into 

 consumers' goods? 



These are. perhaps, debatable questions. 



Three Types of Analytical Work 



It may clarify the problem if one distinguishes three 

 types of analytical work conducted by Government sta- 

 tistical organizations : 



1. Investigations of the accuracy of the basic data, with the 

 aim of finding ways of improving its collection and tabulation and 

 helping to determine what new questions to ask on schedules. 



2. Research on the social and economic significance of rela- 

 tionships derived from basic tabulations, with the general aim 

 of making additions to knowledge. 



3. Quick analysis of data witli the aim of providing factual 

 guidance to an administrator facing a particular administrative 

 problem. 



It would be quite generally agreed that the first of 



these three types of analysis is clearly a responsibility 



of such an agency as the Bureau of the Census. The 

 Bureau is in a better position than any other agency to 

 conduct investigations of the accuracy and validity of 

 its own data. If for no otlier reason, this is true be- 

 cause the original schedules are usually confidential, 

 open only to sworn employees of the Bureau, and be- 

 cause the Bureau has custody of vast quantities of un- 

 published tabulations, in the form of machine sheets 

 and consolidation sheets. There are four times as many 

 unpublished tables relating to the census of population, 

 for example, as there are published tables. A still more 

 cogent reason, however, for the Bureau to regard this 

 type of research as a major responsibility is the fact 

 that the Bureau itself would be the primary immediate 

 consumer of such investigations, whose aim is to im- 

 prove ways and means of collecting the Bureau's basic 

 data and of presenting tlie data accurately to the 

 public. 



There also might be quite general agreement that 

 the third type of analysis is, in general, not the func- 

 tion of the Bureau of the Census as Government sta- 

 tistical work is now organized. This form of quick 

 analysis to inform an administrator with respect to 

 a particular problem is one of the most important so- 

 cial science functions within the Government. It is 

 the point at which facts enter into policies. Quite ob- 

 viously, this function is best performed by analysts 

 directly attached to a given administrator. It is the 

 type of work which might be expected in such an office 



