Relation of the Federal Government to Research 



171 



Of the 1,556, 508 are over 65 years of age and pre- 

 sumably are less active than the 1,048 who are 65 or 

 younger. In Table II ^ are listed the total number of 

 starred men and the number 65 or under in each of 

 our universities and colleges. To save space, 21 insti- 

 tutions, having only one starred man, and he 66 years 

 of age or older, are not listed. 



It is interesting to note that if we summarize Tables 

 I and II we find that : 



We can safely assume that the ablest research men 

 in the humanities and social sciences are distributed in 

 about the same proportion as the men in the natural 

 sciences. Evidently most men are trained where the 

 ablest research workers are located. 



The Proportion of Ph. D. Graduates 

 Becoming Productive Research Men 



Two studies have been made of the productive work 

 of the Ph. D. graduates. Of 1,188 Ph. D. graduates 

 in mathematics, 1862-1933, not over one-third con- 

 tinued in active research.^ In 1927 Dr. M. W. Jerne- 

 gan sent a questionnaire to 500 men holding the Ph. D. 

 in history. From the replies he concluded that less 

 than 25 percent were consistent producers. 



Other studies seem to indicate that about three- 

 fourths of our Ph. D. graduates are employed where 

 research is a very minor interest. It is certainly true 

 that many on whom the Pli. D. is conferred have 

 very little interest in research. It may be remarked 

 here that probably increasing numbers will be em- 

 ployed in teaching in our stronger high schools and 

 especially in junior colleges connected with high 

 schools. 



Probably a fair estimate would be that from one- 

 fourth to one-third, or from 700 to 900 out of 2,700 

 receiving the Ph. D. in 1937, will continue active in 

 research in the sense that they continue to publish 

 after receiving the doctorate; whether a larger number 

 would carry on research if opportunity offered is un- 

 certain. It may be that these figures do not give 

 adequate recognition to men engaged in research in 



industry and business who do not publisli. If, as seems 

 probable, from 40,000 to 60,000 men and women ara 

 actively engaged in research in the country from 1,200 

 to 2,000 would be required for replacement annually. 

 If only 700 to 900 of the 2.700 Ph. D. graduates con- 

 tinue in research work in spite of the iai'ge numbers 

 graduated the supply of men with taste and talent 

 for research is less than the demand. Far too little 

 is known of the life work of men with tlie doctorate. 

 We need some searching studies into the occupations 

 of men with the doctorate and into the demand I'oi-, 

 and the supply of, research men in each field. 



Number of Research Workers Employed 

 in the United States 



The sixth edition of American Men of Science in- 

 cludes 28,000 men and women in the natural sciences 

 who are rated as research workers. On the assump- 

 tion that there is a slightly smaller number, 22,000, in 

 the humanities and social sciences, we have about 

 50,000 research workers in all. 



In the agricultural experiment stations in 1935-36 

 4,000 research men were employed on work costing 

 $16,000,000. An expenditure of $4,000 per man, in- 

 cluding salary, supplies, eitc., was the cost on the 

 average. In the University of Chicago survey it was 

 estimated that $7,000 was spent per man engaged in 

 research. If we average these figures and accept $5,500 

 as the cost of research per man employed we would 

 have : 



= 9, 100 in universities. 



= 11, 800 in the Government. 



= 27, 200 in industry and Ijusine.ss. 



• Several other tables showing the distribution of starred men in the 

 Governmpnt, in the foundations, and in Industry are in the Appendix, 

 pp. 191-193. 



■ Dean K. S. B. Richardson of Brown University. Am. Math. Monthly, 

 vol. 42, p. 199. 



50, OOP, OOP 



5,500 

 65, 000, OPO 



5,500 

 150, OOP, PPO 



5,500 



48, lOP research workers. 

 Another approach to this matter is as follows : Dr. 

 Wm. A. Hamor of the Mellon Institute states in a letter 

 that "the number of industrial research workers at pres- 

 ent is said to be approximately 25,000, embi-acing scien- 

 tists and engineei-s engaged in basic, production, plant, 

 and merchandizing investigations. Some authorities, 

 however, believe that 20,000 is a much closer calcula- 

 tion." As the above does not include research of com- 

 mercial establishments which conduct inquiries in 

 banking, marketing, and other social-science fields we 

 could certainly add 5,000 in such employment. In the 

 Government 10,000 seems the lowest figure we can use 

 considering the money spent. In all universities and 

 colleges 100,000 are on the faculties. Of these, 16,649 

 are on the faculties of the fifteen institutions spending 

 over $1,000,000 each on research, where research 

 is a major activity with a large percent of the staff. 

 Another 13,110 are on the staffs of 39 institutions spend- 



