National Academies Press: OpenBook
« Previous: THE STUDY
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 49
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 50
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 51
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 52
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 53
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 54
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 55
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 56
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 57
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 58
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 59
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 60
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 61
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 62
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 63
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 64
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 65
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 66
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 67
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 68
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 69
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 70
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 71
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 72
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 73
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 74
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 75
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 76
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 77
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 78
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 79
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 80
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 81
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 82
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 83
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 84
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 85
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 86
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 87
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 88
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 89
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 90
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 91
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 92
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 93
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 94
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 95
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 96
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 97
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 98
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 99
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 100
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 101
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 102
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 103
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 104
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 105
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 106
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 107
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 108
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 109
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 110
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 111
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 112
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 113
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 114
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 115
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 116
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 117
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 118
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 119
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 120
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 121
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 122
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 123
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 124
Suggested Citation:"THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION." National Research Council. 1969. The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director]. Washington, DC: The National Academies Press. doi: 10.17226/18693.
×
Page 125

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER The Demography of Postdoctoral Education We received usable responses to our census of postdoctoral s from 10,740 persons who determined that they were included within our definition. Assuming that we had a 65 percent rate of return,1 in the spring of 1967 there were approximately 16,000 postdoctorals including U.S. citizens either in this country or abroad and foreign nationals in this country. Compared with Berelson's estimate2 (although he was concerned only with postdoctorals at academic institutions), the number of postdoc- torals has doubled between 1960 and 1967. The rate of doubling has not been uniform across all fields. In chemistry the numbers have doubled in five or six years,3 while in physics the doubling required only four or five years.4 We will examine the situation in each dis- cipline later. For the present it is sufficient to note that until recently the num- ber of postdoctorals has been increasing steadily since World War II. There is evidence that the growth has now begun to level off, if not to de- crease. In spite of an increase in the number of applicants, the number of fel- lowships awarded by the National Science Foundation has almost halved in the last three years. The Committee on Physics and Society (COMPAS) of the American Institute of Physics has reported that although the number of post- See Appendix A-l. Bernard Betelson, Postdoctoral Work in American Universities, pp. 119-130. 3NAS-NRC, Chemistry Opportunities and Needs, Publ. 1292, Washington, D.C., 1963. 4NAS-NRC, Physics: Survey and Outlook, Publ. 1295, Washington, D. C., 1966. 49

50 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION FIGURE 2 Profile of U.S. Postdoctorals. HOST INSTITUTION DEGREE AND APPOINTMENT LEX 1OO- POS1 -PhD — — ^ ••••. > "•~ -• • — POST- MD POST 25 • PhD CO AND ^ MD K O POCTDOC1 C OCODCIC JOCPROOIT GOVT. ICDUCTRY OOROIGC MMODIOTO RMODIOTO CONIOR C Ul 6 o o t. ^ Ul -1 1 Ul •7 0 = FIELD OF STUDY CITIZENSHIP AND SEX ZlOOn Ul U oc Ul Q. 75- U S. FOR EIGN 50- ^•^^ s ^ ^ , ^ . ^ ~ Q. _j y) -I W _i en DC Ul Ul Ul Ul 2 Ul .OGICO CIONCO < Ul COCIO CIOCCO Ul 0 OCOO OCOO ^Z ^0 5 i O w ^ w CO • m Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire.

51 THE COMPOSITION OF THE POSTDOCTORAL POPULATION doctorals in physics has increased slightly between 1965-66 and 1966-67, the figure was expected to decrease in 1968-69 as the hiring of new postdoctoral s was deferred because of the uncertainty in federal support.5 (It did decrease, by about 3 percent.) The COM PAS survey of 130 department chairmen re- vealed that the number of physics postdoctoral s per faculty member was expected to fall from 0.34 (where it has stabilized for three years) to 0.29. (The implications of a reduction of the number of postdoctoral appointments will be pursued in Chapter 6.) The Composition of the Postdoctoral Population As is shown in Figure 2, 81 percent of the postdoctoral s are at academic insti- tutions in the United States, 8 percent are at U.S. nonprofit organizations, 7 percent are at federal research establishments, 4 percent are in other countries, and only 0.4 percent are in industrial installations. Although the universities predominate as host institutions, it is important to keep in mind that signifi- cant numbers of post doctorals have chosen other places to do research. It will become clear that the nature of the experience and the aspirations of the post- doctorals are relatively independent of the host institution. A more significant difference among the segments of the postdoctoral popu- lation is the type of degree that the postdoctoral has earned. According to the responses to our census,6 62 percent hold a research doctorate only (PhD or equivalent), 31 percent hold a professional doctorate only (MD, DDS, DVM, etc.), 3 percent hold both the PhD and the MD, and 4 percent reported no doctorate.7 Because of the different nature of the predoctoral experience, the postdoctoral activity is different for the PhD and the MD. PhD's, having had more research experience, play the role of apprentices, whereas most MD's, receiving perhaps their first research training, tend to have the status of students of research. Another critical difference among the postdoctorals is the level of their professional seniority. An established researcher will generally neither seek 5Survey of the Committee on Physics and Society-Report No. 1, American Institute of Physics, February 27, 1968. Unless otherwise indicated all data will be presented in terms of what we collected from the various questionnaires. If we have not received uniform return rates from the various segments of the population, the actual distribution will differ from what is reported. Un- fortunately, there is no way to correct such errors. A number of scholars receive appointments and fellowships of the postdoctoral charac- ter without having earned a doctoral degree. Some of these are from foreign countries where the doctorate has a different significance from that in the United States.

52 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION nor expect the same kind of appointment that a fresh PhD will accept, nor will their activities necessarily be the same. From this perspective, several cate- gories are usually established. The "regular" postdoctoral with the PhD is one within five years of his PhD. The senior postdoctoral with the PhD is more than five years beyond his PhD. A similar distinction can be made among those with the MD except that we have used seven years as the dividing point. This allows the man to serve one year of internship and several years of resi- dency before taking a postdoctoral appointment. In this study the post-PhD categories are defined somewhat differently from those in most fellowship programs in order to group the postdoctorals in more homogeneous sets. With a complication to be described below there are three basic subcategories: immediate postdoctoral, intermediate postdoc- toral, and senior postdoctoral. The immediate postdoctoral is within two years of his doctorate, the intermediate postdoctoral is between two years and five years from his doctorate, and the senior postdoctoral is more than five years from his doctorate. A fourth category is important and overlaps those already given. This group comprises the long-term postdoctorals, defined as those who, however far from their doctorate, have spent more than two years on a postdoctoral appointment and who are not on leave from another position. It is clear that the long-term postdoctoral as we have defined him is not necessarily to be identified with the postdoctoral on indefinite appointment. Some of the long- term postdoctorals are simply completing work that has taken more than two years. The professional research appointee, since he did not perceive of him- self as on a "temporary" appointment, may not have responded to our ques- TABLE 5 Number of Postdoctorals by Level of Appointment and Percent Foreign Percent Postdoctorals Foreign Level of Appointment Number Percent at Level Immediate post-PhD 3,997 37.2 44 Intermediate post-PhD 905 8.4 64 Long-term post-PhD 979 9.1 54 Senior post-PhD 815 7.6 44 Recent post-MD 2,391 22.3 26 Senior post-MD 937 8.7 62 Both PhD and MD 334 3.1 84 No reported doctorate 382 3.6 64 Total 10,740 100.0 46 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire.

53 THE COMPOSITION OF THE POSTDOCTORAL POPULATION tionnaire (in fact, he should not have responded). Thus the reader is cautioned that the long-term category is at best an ill-defined group. The distribution of postdoctoral s by seniority and degree type is given in Table 5. It should be noted that the immediate postdoctoral represents 60 percent of all post-PhD's (3,997 out of 6,686). This is the group that most people refer to when discussing postdoctorals. They have taken postdoctoral appointments as their first employment after completing their degree require- ments. The same may be true of some of the long-term postdoctorals, but they constitute less than 14 percent of the post-PhD's. The intermediate postdoc- torals have been employed elsewhere and they are either on leave of absences or are in transition to new employment. To understand the composition of the postdoctoral population it is neces- sary to explore another dimension. In each discipline there exists the spectrum of levels just described and, to a lesser extent, a mixture of both post-PhD's and post-MD's.8 Similarities across fields are not absent, but similarities within a discipline and across host institutions are often striking. Table 6 shows the distribution of the postdoctoral s in the various fields. It is clear that the social sciences and humanities do not participate in post- doctoral education to the extent that the natural sciences do. Whether these fields ought to be more involved or not is discussed in Chapter 6. It should be noted, however, that these data were collected before the National Endowment for the Humanities made its first awards. An important categorization of the entire population can be made in terms of the citizenship of the postdoctoral. Tables 5 and 6 give the fraction of all individuals at each level and in each field who are foreign. The details of the foreign component of the population and its relation to federal and educa- tional policy will be discussed in Chapter 8. At this point we should be re- minded that international travel of scientists and scholars generally is a well established pattern. Between the end of the last century and the first third of this century many American scientists went abroad, mostly to Germany, for postdoctoral training. It is not at all unlikely that as many as half of the post- doctorals in Germany at that time were not Germans. What has changed is that the locus of scientific excellence has shifted to the United States and the availability of support in this country is now much larger. We must also re- member that 8 percent of all U.S. postdoctoral s (35 percent of senior post- doctorals) are abroad. An important feature of the foreign postdoctoral population is the concen- tration of citizenship in only a few countries. Over half of all foreign postdoc- 8The term post-MD is used here and elsewhere as a generic term that includes all post- professional doctorates. The MD degree is by far the most predominant of these (approxi- mately 95 percent).

54 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION TABLE 6 Number of Postdoctoral s by Field and Percent Foreign Percent Postdoctoral Foreign Postdoctoral Field Number Percent in Field Astronomy 106 1.0 56 Mathematics 240 2.2 40 Physics 1,267 11.8 SO Chemistry 1,660 15.5 63 Earth sciences 189 1.8 54 Engineering 274 2.6 64 EMP* Total 3,738 34.9 Biochemistry Other basic life sciences Other biosciences Agricultural sciences Internal medicine Other medical sciences Allied medical sciences Life Sciences Total Psychology Social sciences Social Sciences Total Arts and humanities Other fields Total 1,322 1,030 907 55 1,059 1,166 425 5,964 246 196 442 12.3 9.6 8.4 0.5 9.9 10.8 4.0 55.5 2.3 1.8 4.1 228 2.1 368 3.4 10,740 100.0 51 40 44 62 36 35 37 11 36 23 36 46 ^Engineering, mathematics, and physical sciences. Source: N RC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. torals are from only five countries (United Kingdom, India, Japan, West Ger- many, and Canada) and 75 percent are from 13 countries. Thus, the remaining 68 countries represented account for only 1,211 postdoctorals, or slightly less than 18 postdoctorals per country. Appendix B-3 presents data for foreign postdoctorals by their country of origin. The Postdoctoral in U.S. Academic Institutions In 1967 there were approximately 13,000 postdoctorals of all varieties at U.S. institutions of higher education. Of these, 8,654 responded to the census ques-

55 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS tionnaire. Of the 212 universities that had granted a PhD by 1966, only 147 or 70 percent had postdoctorals. In addition, 27 other colleges or newly formed graduate institutions had postdoctorals. Appendix B-2 contains a listing of the institutions with postdoctorals. The distribution of postdoctorals among these 174 institutions is, however, highly skewed, as is shown in Figure 3. From the curve it can be seen that 50 percent of the postdoctorals are in only 9 percent of the schools that have any postdoctorals and 80 percent of the postdoctorals are in only 25 percent of the schools. Harvard alone can claim 7 percent of the postdoctorals. In spite of the different total number of institutions in the base, the distribution of PhD production is strikingly sim- ilar. The relationship to federal funding9 is also shown in Figure 3. Another way of looking at the concentration is to examine the number of institutions in each field that have postdoctorals compared with the number of institutions that have granted the PhD. Table 7 gives the number of schools having half of the postdoctorals in a given field as well as the fraction of avail- able schools these numbers represent. Although postdoctorals are most widely dispersed among the potential universities in chemistry and internal medicine, the concentration of postdoctorals among a few of the universities is almost independent of field, as can be seen in the last column. The small attention generally paid to postdoctoral activity might be explained by the fact that only at a handful of schools is the number of postdoctorals large enough to be noticeable outside of the departments. In terms of departments, the distribution of postdoctoral activity is given in Table 8. It is not surprising that postdoctorals tend to go to the more pres- tigious schools.10 What might be unexpected is that postdoctorals are present in liberal arts colleges that do not award the PhD. The percentages given for colleges at which less than half the faculty have the PhD may be inflated since the return rate may have been higher from departments with postdoctorals. The current pattern does not differ significantly from what Berelson found in 1960. He found that the institutions in the Association of American Uni- versities (AAU) did about two thirds of the postdoctoral work in American universities.11 At that time the AAU had about 40 members, which would imply that approximately one fifth of all schools had 67 percent of the post- doctorals in 1960. 9A total of 298 schools received funds in excess of $12,000 in 1966 to support research from the AEC, NASA, or the Department of Defense. Since NSF and HEW contribute funds for nonresearch purposes, it is difficult to determine whether the funds from them represent research support. The fit in Figure 3 would not be nearly so close if all of the schools receiving federal support were included. I °The grouping of institutions by reputation is explained in Appendix B-2, which also includes summary data for postdoctorals at U.S. academic institutions. II Berelson,/oc. cit.

66 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION FIGURES Distribution of 1967 Postdoctorals among U.S. Institutions and Comparison to 1960-66 PhD Production and 1966 Federal Academic Science Obligations. lOO-i- • Postdoctorals PhD.s Produced Federal Obligations 20 40 60 80 100 PERCENTAGE OF POSTDOCTORALS, PhD.s. DOLLARS OBLIGATED Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire and Doctorate Records File NSF, data compiled for the Committee on Academic Science and Engineering (CASE) An adequate picture of postdoctoral activity in the universities can be ob- tained only if we examine the various kinds of postdoctoral s there. Table 9 gives the distribution among levels in the various fields. The significance of the activity, both for the university and for the individual postdoctoral, depends on the level of appointment. Usually the young man who proceeds to a post- doctoral appointment immediately after his doctorate is motivationally and professionally different from a seasoned researcher. Moreover, he is at a much more critical point in his career than the older man. Since 84 percent of these

i -s i "^ i ' Hi •i ||| r^ ID in «-. co f; oo o & Q- -1- co "& in co co CM o 1 "o H- $ i if! i oo in «- co CM w o 8 loo in o co o o «- r^ c. .? « \ g I § | a. 5 1 i | 1 i 6 co co o in * o r^ tn £ j| 1 c M £ s •^ 5 -| t> ? ^ al S § "•3 0 'fi £ fi k O o0 CO O o0 O O o M g 1 11 1 in •- ^f co K ^' co C 1 0 C * 1 1 | 1 a 1 g 2! sssssss DC i ro O M 1 1 Postdocto slumber of 1 i! Personcel, C n CO ^ CO «- CM ^ CM •s o CM in ^ in oo «- | 0 | 1 w c C 1 1 X ^ "o s c •" it 2 |1 o .! e.lf o" Postdoctoral Oield 1 1 fr 1 1 8 1 c z LU i j 1 1 i i I iliiiik I m H- S Q. cj CD m — 57

58 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION TABLE 8 Percentage of Academic Institutions Having Postdoctorals by Type of Institution and Department Type of Percentage with Postdoctorals by Department Basic Academic Physical Medical Social Institution Sciences Engineering Biosciences Sciences Sciences Humanities Ten leading 96 72 86 100 61 30 Twenty other major 78 67 79 97 36 18 Established 58 21 71 71 16 8 Developing 26 5 20 69 6 1 Others More than half PhD faculty 4 6 7 25 1 0 Less than half PhD faculty 1 0 0 0 0 1 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. immediate postdoctorals have chosen to do their work at universities, we should discuss their situation next. Immediate PhD Postdoctorals An increasing number of PhD recipients have been selecting postdoctoral appointments as their first appointment after the doctorate. In 1962, 8.5 percent of all PhD's produced in this country went immediately into postdoc- toral positions.12 By 1967, the fraction had increased to 11.6 percent. Since the number of graduating doctorates had grown from 11,507 to 20,295 in the same time interval, this relatively small percentage change indicates almost a tripling in the number of postdoctorals. The behavior of doctoral recipients in the various fields shows even more striking changes with time (Table 10). The percentage in physics and astronomy taking a postdoctoral appointment has moved from 16 percent of the 1962 class to 26 percent of the 1967 class. Biochemistry sent 36 percent of its doc- toral recipients on to postdoctoral work in 1962; by 1967 that fraction had 12These data are derived from the Doctoral Records File, maintained by the Office of Scientific Personnel of the National Research Council from the annually conducted Sur- vey of Earned Doctorates. A questionnaire is filled out by doctoral candidates when they have completed the requirements for their degrees. The respondents are asked to indicate their anticipated employment. Follow-up studies show that their responses are accurate.

s 1| m — • in eo ^ CM cn eo co r^ r^ co o co ^ co «- o in «- CM «- •- CM •- SCM r-- t in in oo s1 1 H 1 C S co" in r^ CD '~ r^ S CM" oo cn S ^ CM" eo". •o I s eu 0O$ t> 0) o ^ eo r^ co in Tf cn CM ^ «- CO in o co cn t cn CM cn ® LL zoo co '«• co CM ^ «*• t eo v CM eo co «- CM CM «- CM r- O en O T> C CO i o | c «- o «- oo o in r^ en oo in o in co cn CM CO in o co CM ,,1 1 «- o o o o CM o in oo co C0 C0 SCO 01 co CM o j* 8. Q. «- in «- CM cn cn •s Q g 0) C £ H 5 S c r-. CM r- co CM co oo r- oo r^ «- O> CD CM oo eo in «- en «- r- —i £ -i in oo v eo i^- oo CM co in co in CM •- «- 6 •- cn in oo oo en _Q c 1 C a Q • O 2 £ •f "o 0. g '3 B ti 'e co in ^ in oo CM t oo co 05 in CM r*- «- oo in co ^ o ^ i- c 1 II CM in in co in co t- «- co co oo in 0 0 0 0 •* CM in in CD o «- CM TJ- «- K o •^ S O 'i CD iA Si (0 m •o «o o O if 0) r- oo «- o CM «- f oo co M- CD «- CM t CO C0 01 ^ ^ «^ <D U . *rf e2 = 05 r^ CM eo CM «- CM «- o in o en T- O O O oo cn o r« eo i S CO O ^ 0 => § CB •s 3 CO s. D| 8 1 2 f 1 '« | 8' o 1 1 co Tr Tf eo <t ^ v en o «- CD O) eo t cn oo co en o •- o> 1 1 ^ ^ «- eo o r- CM in co co co co co co CM r- oo in *«• CM 5? in CM eo co S CO 2 CM 8 g • "i 1 0- a o •s s • ^ S m 15 .2 0 Ogricultural sciecces Biochemistry Other basic medical sciecc ^ Biosciecces Agric. and biol. sci. To 43 Psychology Cocial sciecces Orts acd humanities Oducatioc acd professioc! o TABOE 9 Distributi Postdoctoral Oield V % in Icternal medicice Other clicical medicice Ollied medical sciecces Medical sciences Total Total All Fields aOcgiceericg, mathematic Cource. . NRC, Office of C .ii >- C ? S E 2- .i - <» E ° ui S if <" "*• 1 1 1 II i ,* £ S 5» t B •— Ui '± t > CD t. 0> ^« JS m £ £ <o c 5 < Q. O UI 01 59

r^ 8 S 2 o o •o o p. Q. I S to .•& cu J cn C I I •o (O Ul _l DO 2 s 5? 51 ? 11 55 11 s •+• n 0 o ield cn •- n co oq «- co co CM CM' t' oo CM «- CO «- «-CM in in CM CM «- •- CM «- 5 •- CM in «- ? o CM «-" -~cN"" rf"' 8" S OCQlf) «- ^ CM S ScN«-™^ ^ ^ Lfi CO CO CO CO CD »4" LO r^TOCor*•inCMO co «-"«-" CM« «-" co r^« q CD CM CM oq iq cn oo. to q r^ in o> CM co «-" «-" CN~ -oocNoor^n «-cNooinr^- ooMCMnM co" ocn'~oo«~oo^ 's;^^^.?Ist^»^ ®? ^S^co^^S cocM«-cy'~ ^ «- OU)CM•-CMtf)«- CNCOnOlDCOr«« ^* OlCD^lDcD^rs lf)lf)^'CMlf)<TCO CM 1 8 00" oi oi o co a! cri gj d W CM' CM d ri o' i g T-" •-" CM" CM" Q CM oo. ^; cn oq oo CM «-. CM cn t* - v in CM in , oi in r^ «-' pi in co in in oo CM •-' o CM' C0" «- CM rt cN «- ooocnr^ini*-co cNCMr-r-coooco r^ £ cor^cMT-cMcr)cN tr*-oo^'«-ooin in .£ o - * ;c? 60

61 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS increased to 58 percent. On the other hand, the fraction of new doctorates taking postdoctoral appointments in the humanities and in the social sciences has remained stable at 1 and 2 percent, respectively, and in mathematics it has dropped from 9 to 7 percent. The drop in the number of positions funded in the 1969 fiscal year occurs in the face of a rising demand. It is to be expected that the uncertain impact of the draft on graduate enrollments will not affect the PhD production for the next three or four years.13 Consequently, the reduction in positions will result in the failure of potential postdoctorals to realize their training goals. An obvious question is, Why are so many seeking postdoctoral appoint- ments? The answer is not simple. Not only is there no single answer, even for an individual, but the emphasis changes as we move from field to field. Never- theless, it is possible to enumerate several categories of motivations that are present in varying degrees among most of the postdoctorals in the natural sci- ences. The humanities and the social sciences require separate treatment. It is important to realize that only one out of nine PhD recipients seek postdoctoral appointments, and among these there is a great spread of talent, accomplishment, and background. A man who received his degree from a small university and who did his research with a relatively unknown faculty member might have a different motivation from the graduate of a major institution whose mentor was a Nobel Laureate. Moreover, a man whose field is theoretical physics is likely to perceive the requirements for his future career differently from the man in biochemistry. The unifying theme of postdoctoral work is, by definition, research. More relevant here, however, is the commitment of virtually all the postdoctorals to research and scholarship as a career. Another almost universal feature of postdoctoral activity in the academic world is that most of the participants are anticipating an academic career. With one exception, all the postdoctorals we visited on 18 different campuses preferred to be employed subsequently in a university where they could work with graduate students and carry out re- search. The one exception was a man who had taken the postdoctoral appoint- ment to determine whether he wanted a research career. He did not and is now headed for a position in a state college system. The others not only were looking to the university setting, but also were hoping to be employed in the more prestigious institutions (at least as prestigious, that is, as the university at which they were taking their postdoctoral appointment). Several at a top institution turned down faculty appointments at lesser places in order to take the postdoctoral positions. Their attitudes toward industrial careers were uni- formly negative, usually because they saw such positions as lacking both the 1 Except for the reduction arising from those candidates who will purposely delay com- pletion of degree requirements until they have passed the critical 26th birthday.

62 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION TABLE 11 Next Anticipated Employer of Immediate U.S. Postdoctorals, by Type of Host Institution Percentage of Immediate Postdoctorals by Type of Next Anticipated Employer Type of Host Institution University College Federal Govern- ment Industry Other and Unknown Total Number (100%) University 73 7 3 8 9 1,749 Foreign 77 7 2 8 6 156 Federal government 66 3 23 7 12 209 Industry 63 6 0 36 6 17 Nonprofit 70 6 3 3 18 101 Total % 71 6 6 8 10 No. 1,597 139 108 170 218 2,232 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. freedom and the student contact of the academic world. "If I had wanted an industrial job, I wouldn't have taken the postdoctoral," said one chemist. "I took a $4,000 cut in salary to come here." Another objection to nonacademic positions is the belief by many that the move is unidirectional: once one leaves the academic world, they feel, it is difficult to return. Some reluctantly ad- mitted that, if no suitable academic position was available at the end of their appointment, they would take one in government or industry. Others, how- ever, indicated that in such a circumstance they would try to prolong their postdoctoral appointments or that they would move down the academic hierarchy. How much their formal responses to this question on the census question- naire reflected the postdoctorals' desires and how much their more realistic expectations is unknown. It is possible that those interviewed were a biased sample, since their unanimity does not correspond to the replies to the ques- tionnaire given in Tables 11 and 12. Nevertheless, 80 percent of the immediate postdoctorals at universities anticipate an academic career and even 58 percent of those who are taking their appointments in industrial or federal laboratories expect to return to a college or university. By field, physics, chemistry, and engineering have the most postdoctorals heading toward an industrial career. In physics, most of those anticipating an industrial position come from the subfields of atomic and molecular physics, solid state physics, and classical physics. Solid state physics, with 107 university-based postdoctorals, has only 15 going to industry. In nuclear and elementary particle physics, with 221 postdoctorals at academic institutions, only 9 are going to industry.

I i E 1 1 Sin CN O CO U) •- o CMOOCM OOCOCM -!. C | in oo co co 05 CM J£ £ co cn CM •- Nf k 9 j; •* 0 CD in CM in oo CO CM oo r^ CD CMCMt^in 0) i 5 in f CO CD CM CM CO C0 1 c n en r^ co oo CM P^ r^ oo CD O •o o .£ a 1 1 E UI •o J s a e 0 9 CD p^ in c N CM r-. «— CO o0 P^ o0 ^ CD r- "(7> '+j "2 CO 1 en p^ M CM CM O CO CN •-. 1 cN 1 1 1 CM 6 r^ § * CM «- "c X D i ts •8 e "S i 1 2 8 i- Oederal I i «- 1 lo- CO CO «- CO O CO CM CM in CO 1 1 1 1 £ a •a 1 1 in CM r^ co 1 co c C M CM c £ 2 1 | i CO ' 6 D w vt a> £ i to « CO O r- CD CM CD O ^ in r^ t^ «- CMOOfOS in co C 3 •- •* «- co CM in t«« 1 K co CD «- co CM eo in «- r^ CD Q) ^ o E 1 "5 E C % M- O •s i 1 fc 1 ,2" c °- O c E & —- 'u CD Q. 8 > co o en co CM r-- CM CD r- * o P^ «- co o en «- co co & C E 1 1 1 1 '£ O5 CM 3 SS CO CD CD r- CM T- co en in en CD p^ in co r- r- 15 o •- f LLJ C0 cn r- CD oo cxi co r- oo CD r- in M 0j 3^ CL a .S "2 S 0 • S '^ » ^° i« '0 0 (O ^ 8 ° ™ "o ^ S a " i • 1 2 1 IBS || « .n .y o ^ ft ? •- 8 '= a c O Postdoctoral O Oarth sciecces Ocgineericg E MP3 Tota Ogricultural sc Biochemistry E ^ cl « F ? Other Totat Total All F> dOcgiceericg, Co.-. CRC, CN Mathematics Ostrocomy Physics Chemistry o»sl >.ciiS •iS&<5j ||*| LU CO f .9! S3 # ° - g S 1 § 1 J fill O bo 5 a! w < S 63

64 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION The situation is similar in chemistry. The subfields of analytic, organic, and pharmaceutical chemistry contribute most of the postdoctoral s to industrial positions, while inorganic, nuclear, and theoretical chemistry tend to retain their postdoctoral s at the university. In engineering, the fields of mechanical and metallurgical engineering contribute 9 out of the 14 engineers of all kinds going into industry. Subfields such as electrical, aeronautical, and chemical engineering contribute only 3 postdoctorals to industry out of the 23 in these fields. Although it has been suggested by some directors of industrial laboratories that the postdoctoral experience weans the young doctorate away from indus- trial careers, it is more likely that the career decision between the academic and the industrial environment is made earlier. Reflecting the attitude of many industrial employers that the postdoctoral experience is unnecessary, faculty members tend not to urge their better students to take postdoctoral appoint- ments if they are headed toward industrial careers. The response of faculty (with and without postdoctorals in their groups) to the question, "How strongly do you encourage your better graduate degree candidates to take an extra year or two of postdoctoral study?" is given below: Encouragement of Postdoctoral Work by Faculty (Percent) Anticipated Career With Postdoctorals Without Postdoctorals of Doctorate Fairly Not Fairly Not Recipient Strong Strong Strong Strong Strong Strong Academic 75 18 7 49 23 28 Nonacademic 15 28 57 9 22 69 Reasons for Postdoctoral Work With this background we can examine the motivations that led the new PhD to his postdoctoral position. The typical postdoctoral in the natural sciences aspires to a lifetime of research in an aca- demic setting where he will have students to train and where he can be a fac- ulty member in the complete sense of the word. However, when he examines the prospect, there are several reasons why he is willing to postpone entering the community as a full-fledged member. The first reason can be stated generally as "I am not yet prepared academ- ically to become a professor." In part, this attitude is realistic in that the young PhD has not undertaken a complete research problem. We asked a group of 16 terminal-year graduate students from a variety of departments in a Big Ten university how many were anticipating a postdoctoral appointment. Slightly over half responded affirmatively. We then asked how many of the group had invented their own thesis topics. The correlation was perfect in this imperfect sample: All who had been assigned a thesis problem by their advisers

65 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS planned to take a postdoctoral appointment; all who had come up with an acceptable research topic on their own did not feel the necessity of the addi- tional apprenticeship. Again, dealing with this same group, we discovered that there was a strong subfield dependence for this phenomenon. The geologists and classical biologists tended to have been more independent during their thesis research, whereas those in the more mathematically complex sciences were dependent on their advisers, at least to the extent of knowing what prob- lems were both significant and capable of being accomplished in a reasonable amount of time. A physics professor has suggested along this line that the transition from being a student to being a professor is too abrupt. In the present system the professor, in addition to his pedagogical responsibilities, is expected to carry out independent research. Postdoctoral s maintain that struggling through only one research problem is not sufficient to create the independent re- searcher who can be a teacher as well. Before facing students, many postdoc- torals would like to shift fields slightly or to change institutions to pick up more breadth and style in their approach to research. They argue that with- out this experience they will tend to work the rest of their lives on their thesis problems. In part, however, the postdoctoral who senses that he is unprepared for full faculty responsibility is less concerned about his research qualifications than about his readiness to undertake the other responsibilities of a graduate faculty member. One young man questioned whether he was ready to guide graduate students in research. He expected to learn how this was done by observing his postdoctoral mentor and by serving as a surrogate faculty member in the re- search group. For him the postdoctoral appointment was more like a medical internship where he would have limited responsibility in the whole scope of professorial activities. In this vein, another response by a postdoctoral expressed the desirability of allowing time to get his first research paper published in order that he might have stature in the eyes of the graduate students. Among the other benefits of a postdoctoral appointment is the time lapse during which one's reputation can become established on the basis of one's thesis research. It is likely that this motivation depends less on the academic realities than on the insecurity of a man who has finished only one project. A second reason for undertaking postdoctoral work that is shared by many postdoctorals is enlightening for the insight it provides into what graduate students perceive to be the life of a professor, especially before attaining tenure. It can be oversimplified by the statement: "I am not yet eager to be- come bogged down like the assistant professor." The assistant professor is understood to be "the low man on the totem pole," burdened with a heavy teaching assignment, faced with creating lecture notes de novo, forced to seek

SB c ^ 9 1 i - C £ i2 § a a fSi CMCO^"^ ^COlD^ Or^C^^ lf)^CMCM «- C0 ^C0 ^C0 O) •o c g S 1 3 1** 1 sij o e o g 6i£ ££"" SP^" £ R «" «" °o§"«- i 1 < | ? "o 0 1 o> c f ?; •o S (OCNJ H S 5 coco coco coco •- oo £ H- i] 43 £ 1 1 .B C >. C 9 | » SO) ^ o o r^' n o ^ ^ C>i o ^ t ^ •- i ^t in ^ ^ i n CM r^ , ; Q. 0) c O • "8 "g 1 D- I f .i fl (/> i 1 .c 16 LO CM CM ^ O) co ^ ^ fo 3 ^ ^ i n PO CM o £ o "o o "o 1 S o. I il r-CM«-o ooi^-ino cnint«- cn(o*«- c int inn ** «-r- c " i ' ( 2 1 •— c c c c ! <5 < o .O o o "o ceo cO era cw c. .2 R £ .2 £ £ .2 s £ .2 a : 1 1 1 fc g 1 i u g 1 c . 1 1 * fc 1 II i0 3 •-• i~ (0 j «^ l^ (0 D •— ^ (D 3 •— >f i • H C S SCfc %^~E <ul-C<V ' o cc ^ <C O cc .E <c O cc -E <c O cc -E <c Q ' : § t PI •« > | 2 ( LU | w « >• 'c 03 'in C O c 2 & £ O CO I ! C 66

67 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS extramural funds to support his research, and expected to be compiling a re- search record that will result in a permanent appointment. Since the teaching must come first and since the ancillary responsibilities of committee work will compete with his research, the postdoctoral seeks to get a running start at his research in the hope that the momentum will carry him through those first critical years. Lacking the confidence to expose himself to these overwhelming pressures and counterpressures, the fresh PhD seeks the intermediate stage of the postdoctoral appointment. From his point of view, the postdoctoral years provide several useful step- ping stones. In the first place, he recognizes that it is easier to do research "piggy-backing" on a faculty member's research grant than to obtain inde- pendent support. He does not have either the research record or the reputa- tion to be able to compete successfully for his own grant. Although some sources, such as the Petroleum Research Fund, have special "starter" grants especially designed for the young new investigator, the size of the grants is seldom sufficient to enable the man to purchase major equipment items. Unless the man's field is "small" science, the various grants and fellowship programs alone are unlikely to provide him with the research environment he seeks. Not only will the postdoctoral period enable the young researcher to estab- lish a research record and a respectable publication list to present eventually for promotion, but that record will also make it easier to obtain a grant of his own when he joins the faculty. Finally, some anticipate accumulating a num- ber of research problems on which they can work while serving as assistant professors. They do not expect ever again to have enough unoccupied time to be able to plot the future. It should be pointed out that the picture drawn above is that perceived by many postdoctorals. If it is incorrect or distorted, it is nonetheless affecting the behavior of these young men. The only information that we collected that bears on the matter is given in Table 13. The chairmen of departments in all kinds of institutions of higher education were asked to describe how the average assistant professor in their departments distributed his time. Under- standing that these are estimates by the chairmen, there is still an interesting shift as one moves from field to field and among the reputations and types of schools. At the top institutions in the sciences, approximately one half of an assistant professor's time is spent in research. At other schools the fraction of research time is much less and correspondingly more time is spent on instruction. The third motivating reason for postdoctoral activity is somewhat more cynical than the others. It is a response to the academic marketplace and takes the form of the assertion that "the establishment requires that I have this experience." By only a very few is this reason given as the primary cause for

K « •s III 1 |!f oor^enin^ CM°°CMCMCM co"~•-* S^°°in H 111 | 1 1 S Q g- c o sis oooT-vr^ tcooCM«- cnn3(0C0 acovcov O 1 «-«-COCM«- CMCMCO«-«- ^^COCM O «-Tf £ 1 E £ o 1 i o c I 1 £• & <— CM CO »- «- «-COCM «- CM«-CO«- CO «-«? § •s U. 1 0 H '£ ^ J3 ^ 2- i toincMen coooocMr^ COfCDOCM CM*rinr^CM I •-inCM «-KcM «-*CM«- CO COCM c 3 'o O 1! o 'c < i ii f •8 Is «-in«— »— ^incM CM^CM CM coco 0 1 § S I .22 E ll "S M-«)COOCM CMr^Mff ^gcococo ^vcnin^ 1 * t- -J ('•« «- CD CM CM ^ ^ CM CO CO X i LU i W5 •M +^ +J *•• reviou i i0 i i0 i i0 i i0 i| i I 1| if ii § 1 |s if Q. iA C ^2oa*-Q 2oflj-Q SQe^ SQ^-n •2 .2 E — "a ^ ^ ro ~"'n^"<o ™~'u^« ~'D^C£ C II '"8 I E | a. uj l2o?Z(5z l2o?ZOZ l2o?ZOZ l2o?ZOZ £ * .y c > s m i i 8| 8! 1 t CD <g a 'w • £ :; <£ CJ £"£.23 H < Q a. O CO I 68

69 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS their taking a postdoctoral appointment, but most will agree that the "system" insists upon it. One man, an organic chemist at a California university, stated that he would not have taken the postdoctoral if it had been possible to get a faculty position in a "good" school without it. A biologist from a New England university of note was an instructor for the first semester of the 1967-68 aca- demic year but became a postdoctoral the second semester not only to allow himself more time for research but also because "it is the done thing." A post- doctoral from Italy admitted that the research he is doing here is similar to what he would have been doing at home but having been a postdoctoral in the United States would increase his chance for a better job back in Italy. The idea that it is not possible to get a faculty appointment in a major in- stitution without a postdoctoral record is only a slight exaggeration in some fields. The rationale of department heads for preferring postdoctoral s for faculty appointments will be examined in Chapter 6 but it is instructive to examine the practice of recruitment in selected fields across the spectrum of institutions. Table 14 gives the distribution of the immediate previous experience for recent appointments to the junior faculty (instructor or assistant professor) in several fields. What is striking in the sciences is the de- crease in the fraction of new appointments who are postdoctorals and the corresponding increase in the percentage who are still graduate students as the reputation of the institution descends. Also of interest is the general tendency for the percentage of new faculty who are appointed directly after earning the PhD to rise as the institution goes down in reputation and then to fall for the weaker colleges. More to the point, however, is the far-from-negligible fraction of new appointments even at the top schools who are fresh PhD's. Although it is clearly advantageous to have had postdoctoral work, it is pos- sible for the most talented young PhD's to be hired without that experience. It is curious that whereas the chemists, both postdoctorals and faculty mem- bers, spoke most often to us of the "requirement of postdoctoral work by the establishment," it is the physics departments at the better schools that tend to require it more often. The fourth reason is obviously more appropriate to some postdoctorals than to others but, with some extension, might be made a valid rationale for postdoctoral study generally. This reason can be stated, "I want to see how research is done elsewhere." One postdoctoral who had obtained his PhD from a small technical school wanted to see what the academic world was like in a large institution. He was aware that the style of research and graduate education at a developing university was different from that at a major univer- sity, and he felt that without the postdoctoral experience he would have had a distorted idea of research generally. Somewhat the same idea was expressed by a postdoctoral in chemistry who took his PhD with a relatively young pro- fessor at a small university but who was taking his postdoctoral with an emi-

70 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION nent scientist at a prestige school. He not only wanted to see how a top scien- tist did his research but he realized that he was much more likely to acquire a good faculty position with the recommendation of the better-known man. Several postdoctorals have pointed out the possibility of upward mobility in the academic world through the postdoctoral mechanism. The final general reason given for seeking a postdoctoral appointment can be phrased, "I finished my PhD at the wrong time (or in the wrong field)." The ideal time for finishing one's doctoral work is in the late summer. Then, with no break in income, the graduate can take employment in the fall. If a man finishes in December, say, the choice positions are filled and the recruit- ing season is not yet open. It often happens that a man will be appointed as a research associate on his mentor's grant for the remainder of the year. From the faculty point of view the situation is ideal; his new associate is entirely familiar with the apparatus. From the postdoctoral's point of view an awk- ward financial situation is resolved. If a suitable appointment does not appear during the year, he might be kept on for another year. Several men have pointed out the utility of the postdoctoral appointment in providing a useful and productive way of waiting until the appropriate position opens up. Another alternative is to make use of the postdoctoral period to change fields. One man did his doctoral work in chemistry and then decided he needed more physics than he had been able to acquire as a student. The postdoctoral appointment made this possible. Another chemist did his work at the predoc- toral level in nuclear chemistry and was taking his postdoctoral in radiochem- istry. He asserted that there was no other way to make the shift unless he re- peated some graduate work. A professor in the field of x-ray crystallography as applied to biological structures pointed out that interdisciplinary fields, such as his, train their students at the postdoctoral level. He prefers to have his advisees complete their doctorates in chemistry or biology before joining his group. In addition to these general reasons, there are more isolated ones. One bot- anist wanted to follow up some peripheral areas of his thesis research that did not appear within the dissertation. He remained at his doctoral institution since that was where his plants were. For married women the postdoctoral position is an ideal one for working in their fields either while waiting for their husbands to finish their graduate work or because their husbands are on the faculty and the nepotism rules do not permit them both to have a regular appointment. The situation in the humanities and in the social sciences is different. As is evident from Tables 12 and 14 (p. 63 and p. 68), the postdoctorals in these areas who seek academic positions-and most of them do-would have had no difficulty in taking a faculty appointment even before finishing their doctor- ates. It is also the case that only a minority of the postdoctorals in these fields

71 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS can be classified as immediate postdoctorals (10 percent in the humanities and 26 percent in the social sciences as compared with 53 percent in biochemistry, 71 percent in physics, and 75 percent in chemistry). Unlike the case in the natural sciences it is not the pattern for doctorates in these fields to seek post- doctoral work or to get it. Consequently, we are dealing here with individual cases rather than with general patterns. The immediate postdoctorals in these areas are people with particular research interests and with exceptional oppor- tunities to exploit them. Almost inevitably they will be back in the classroom within the year. When asked to check the three most important reasons for seeking a post- doctoral appointment, over 70 percent of the respondents in the natural sci- ences selected the following:14 To gain further research experience (1) To acquire additional research techniques (4) To work with a particular scholar (2) To broaden my understanding of the field (3) To carry out a piece of research on my own To put myself at the growing edge of current research (8) To develop further the research I did during my predoctoral training To see work being done at other centers (7). The other options that were checked by less than one in seven respondents were as follows (in no particular order): To sharpen the focus of my research To give me a free period for research before I get saddled with other responsibilities (5) To support myself in the academic world until a suitable faculty appoint- ment becomes available To give me some teaching experience To give myself a breathing spell after my formal training To give me further time to mature (6) To give me a chance to publish something. That these lists should give a different impression from the discussion above is perhaps explainable by the fact that the unstructured interview per- mits more candor than the printed form. The choices by the faculty more closely correspond to the interviews with postdoctorals. The list is arranged in order of decreasing frequency of response. The parenthetical numbers following certain statements represent the order in which at least one out of seven faculty members gave as reasons for promoting postdoctoral study among their better graduate students.

72 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION

73 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS Geographic Mobility Having some idea how many immediate postdoctorals there are and why they seek such positions, we now look at where they are and where they come from. We will concentrate on those with U.S. PhD's; the foreign component will be discussed later. Figure 4 gives a comparison of the geographic location of the doctoral institution of all 1966 U.S. PhD's, of the PhD institutions of immediate U.S. postdoctorals, and of the postdoctoral institutions of these same postdoctorals. It is evident that the northeast and Pacific regions consistently attract more postdoctorals than they produce, whereas the rest of the country has the reverse experience. Moreover, the east- ern and western seaboards produce a larger proportion of postdoctorals than they produce PhD's. The center of the country from north to south, on the other hand, sends a smaller fraction of its doctorates on to postdoctoral work. When we examine the geographic distribution of the immediate postdoctor- als at their various educational levels (Table 15), a general pattern unfolds. As the population progresses from the baccalaureate to the PhD and from the PhD to the postdoctoral, it becomes more uniformly distributed geographi- cally. This is true, almost without exception, in each field. The East and Mid- west tend to send their baccalaureates to postdoctoral appointments in the South and West with the West being the major beneficiary. The East particu- larly is the baccalaureate origin of eventual postdoctorals, to a greater extent than its being a baccalaureate origin of PhD's generally. The situation in the Midwest is just the opposite. TABLE 15 Geographic Location of Immediate Postdoctorals (with U.S. Baccalaureates) at Three Training Levels, All Host Institutions Geographic Area Baccalaureate PhD Postdoctoral Baccalaureate PhD East 40 34 33 32 30 Midwest 26 27 20 32 34 South 17 18 19 20 18 West 16 20 21 16 18 Foreign 1 7 Percentage of Immediate Postdoc- 1960-1966 PhD.s torals by Location at Training Level Total Number (100%) 2,261 2,261 2,261 80,042 80,042 Note: The Eastern area includes New England and Middle Atlantic regions; Midwest: East and West North Central regions; South: South Atlantic, East and West South Central regions; and West: Mountain and Pacific regions. See Figure 4 for states included in regions. Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire and Doctorate Records File.

74 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION It must be remarked that the data presented in Table 15 are for all imme- diate postdoctoral s both in and out of academic institutions. If we restrict our attention to those who hold their appointments at U.S. academic institutions, the picture changes somewhat (Table 16). Except for physics, the overall flow pattern is that the South has a net loss to all other areas, the Midwest to all areas except the South, the East only to the West, and the West gains from everywhere. The major reason for the difference between this pattern and the one for all immediate postdoctorals is that the nonacadcmic host institutions (mainly federal government installations) are heavily concentrated in the South, whereas the South is relatively weak in academic institutions. It re- mains to be seen whether the conscious federal policy of placing federal labo- TABLE 16 Migration of Immediate Postdoctorals at U.S. Academic Institutions from PhD to Postdoctoral Institution for Selected Fields Postdoctoral Field Geographic Area Number of Postdoctorals With PhD from Area Net Flow into Area Net Upward Mobility into Area In Area Physics East 148 171 -23 +9 Midwest 119 114 +6 +4 South 66 66 -1 -24 West 106 87 + 19 +11 Chemistry East 161 142 + 19 +11 Midwest 123 135 -12 +46 South 81 97 -16 -42 West 92 83 +9 -15 Biochemistry East 100 80 +20 +6 Midwest 92 110 -18 +34 South 43 69 -26 -37 West 83 59 +24 -2 Biosciences East 72 63 +9 -21 Midwest 53 66 -13 +25 South 23 29 -6 -9 West 67 57 +10 +5 Total, all fields East 729 674 +55 -4 Midwest 489 551 -62 +140 South 291 353 -62 -151 West 467 398 +69 + 15 Source: N RC, Office of Scientific Personnel, Postdoctoral Census Questionnaire.

76 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS ratories in economically depressed locations will raise the level of the academic institutions there. It is evident from our data that the effect of the policy is to draw substantially more postdoctoral talent into the South than the academic institutions alone are able to attract. Another component of this geographical flow of postdoctorals is the mi- gration among institutions of different reputation. We assign to each postdoc- toral who changes area after his doctorate a positive or negative weight, de- pending on his moving up or down in the reputation of the schools with which he is associated.15 Thus, a man who received his PhD in the East from one of the ten leading institutions and who takes his postdoctoral in the South at an established institution will be given a negative weight. In a similar fashion, a man who received his PhD in the Midwest from one of the 20 other major institutions and who takes his postdoctoral in the West at one of the ten lead- ing institutions will be assigned a positive weight. Finally, a man whose post- doctoral institution has comparable reputation with his PhD institution will carry zero weight. The last column in Table 16 gives the net upward mobility measured in this way. Institutions in the Midwest tend to bring in postdoctorals from institu- tions of lesser reputation, whereas the South does the opposite; East and West show little net change. The following table gives the number of institu- tions in the top 30 schools in three broad fields in each area (the number in parentheses is the number of schools in the ten-leading group): Number of Top Thirty Institutions Area Physical Sciences Basic Medical Sciences Biosciences East 13(5) 13(4) 9(1) Midwest 9(2) 10(3) 11(4) South 3(0) 1(0) 3<1) West 5(3) 6(3) 7(4) The direction of flow tends to equalize the geographic distribution of people with experience at more prestigious institutions. The Midwest is undoubtedly doing more than its share of upgrading, and the East is not helping as much. On the other hand, the East is relatively weaker in the biosciences and the flow in that field is also in the direction to restore the balance. Inhibiting this tendency toward balance in quality is the uneven interest in postdoctoral education among doctoral recipients at institutions of greater and lesser repute. The significance of this variation can be seen in Table 17 in which the percentage of PhD's taking a postdoctoral is given. Chemistry and the basic medical sciences are affected least, but existing problems caused by quality dif- ferences among institutions are likely to persist in fields like mathematics, engi- See Appendix B-2 for the ranking of institutions.

76 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION TABLE 17 Percentage of PhD.s Taking Postdoctoral Appointments in Selected Fields, by Type of PhD Institution Percentage of PhD.s Taking Postdoctoral by Type of PhD Institution Postdoctoral Field Ten Leading Twenty Other Major Established Developing Total Physics 35 31 23 10 26 Chemistry 34 37 36 24 33 Other physical sciences 17 8 4 5 9 Engineering 7 4 5 1 6 Biochemistry 68 72 48 47 58 Other basic medical sciences 41 32 43 29 36 Biosciences 38 30 27 14 26 Social sciences 3 2 2 3 2 Total 17 15 15 11 15 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. neering, and biosciences. These differences among the schools may be a result of corresponding differences in the quality of graduate students attracted to them. The migration by field between institutions of different reputation is illus- trated in Table 18. Overall, there is net upward migration. However, in some fields there is little net change. These are the physical sciences (with the excep- tion of chemistry), engineering, and biosciences. These fields are also the fields in which fewer than half of the postdoctorals make a move involving a change in institutional reputation and they are also the fields showing the least equali- zation through geographic mobility. Of interest in this regard are the tables presented by Berelson16 showing the tendency of faculty members to be hired at institutions of equal or of less reputation than their PhD institutions. Al- though we have no hard data, there is testimony to the ability of a man to up- grade his PhD by taking a postdoctoral appointment at a more prestigious in- stitution. The good PhD from Harvard can expect to have little difficulty in being hired at a top institution; it is probably true that the good postdoctoral at Harvard can do the same regardless of his doctoral institution. Not everyone changes schools after the PhD. However, the differences by field are indicative of significant differences in attitude toward postdoctoral appointments. From Table 18, we can see that chemistry and the basic medi- cal sciences retain only one in six or seven while the other fields keep a third 16 Berelson, Graduate Education in the United States, pp. 113-115.

oaP oor*•p^cooocostf) co coiD^o^^rt«— oo H Z C " 1 ! — CO g; c •a 1 o •o '^ ?*5 0) 1 ? LL a S GC CO«-*o0CDCMtCO O CMCMCM •-CMCM«- CM xt s c Q ! o £ +5 1 3 .1 •g s «43 s ^ M C O) 1 i K S covcncooocooocD us g~ CO«-CMcD«-«-CMCM CM o o o I 3 fi a. C «*• 1 § g TJ o || '5 C § S 1 Q O S« 1 £ H ?! a. 0 tf_ o 1 il > e </) § cNinTfintinocN CM 3 c CMCMCM«-CMCMCMCM CM • •M *J U § I . S 8- i B 2 DC Q. si c 1 "o is £ Cff i ?i > 0 3 V 1 0 il | 1 1 CM O CO ^ CM r^ CO O CM CM^CM«~^COCMCO CO g D_ £ H- "S H i § 1 1 s 1 •s s> i C — 1 2 •s « . ^ u. >hf! 1 o" C0 UJ 1 oc 8«ti|a>8 | z CO 1 •sEO.E-gOS1! ^S ^ 5 f ? .§ € o g § 0) H Q. oToOiutoOoaw c/} 77

in •D i o H m b •o C — «-inoo^co *inr-O)f cNincocnin $ < i^ r- r- oo i- r^ oo oo oo oo oo oo oo oo oo M * E CO n 0 O M- £J 1 Q o 2 • S o |l .2 Q. a •D o •1 Q S 8 - « « (o T- in r^CMoonco CO•-OOCM "S £ a eo £ o ininincoin incou>r^co incor-CDCD 8 \ 0) o CE O BO 1 O E O) 8 2 C s 'E Q i C (D O EC z € LU f* S C s i •§ 1 1 15 I « ro 2 •-cNCMinco COfCM«-CO r^MCMCMin 0 o H Q. to 3 CO tn C £ 0) O £ M I | I !i i 1 H S Q O .n ed (Tota stjtution ntage of P lining at P! ersoccel, 1 o E Ic? a, 5 0 r^in^oi n ^or^of ^r^^r^o 0 HI g O 1 M- 1 |5 S- •o 0 u> O O I § I I O *! TJ 0 1 Postdoctoral Oiel Type of Ocademi Institution 8 fc -5 b 5 "?.£-oo> S?-£T>OI ?£-no) £ •- o » .5 ^ 1 5 0 J S ^ 5 0 J ,| ,, o" uj ^ lltlfl "till l|fl|| cc Z 8 < "o> '»HHujci .yHHuici SHHLUQ ^ !8 .2 1 H iJ. Q. CO CO 78

79 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS or a fourth of their postdoctorals at the PhD institution. Although the num- bers are small, in engineering two thirds of the postdoctorals remain at home for their appointments. With some danger of oversimplification, these results correlate with the impression gained from talking to faculty and chairmen around the country. In those fields where relatively few remain at their PhD institutions, there tends to be more concern about the experience that the postdoctoral receives. In the other fields, the postdoctoral is seen more as a research aid than as a person to be trained. In fact, there is not much enthusi- asm for postdoctoral work for any reason among the engineering faculty. Industrial experience is often seen as a much more important component of a faculty member's background. Another aspect of what we might call the "stay-at-home" is his quality compared with the quality of those postdoctorals who are brought in from the outside. We cannot use the reputation of the PhD institution here, since the stay-at-home at the ten leading institutions will, of course, share that repu- tation with all those classmates who changed institutions. An alternative meas- ure of quality is the years elapsed from baccalaureate to PhD. Although not significant in individual cases and certainly not comparable across disciplines because of differences in curricula and in predoctoral support patterns, it is probably true on the average within a field that the shorter the baccalaureate- to-PhD time lapse, the better the graduate. Table 19 gives some data on this variable for several groups. Although even postdoctorals who remain at their doctoral institutions average a year less in achieving the PhD than graduates generally, the postdoctorals attracted from the outside have spent one year less than the stay-at-home in completing degree requirements. The migrating postdoctoral is likely, therefore, to be of higher quality than the stay-at-home, and postdoctorals generally are signifi- cantly better than the average PhD. Even those who migrate differ, and the complaint is heard that weaker schools cannot attract postdoctorals of as high quality as those the more pres- tigious schools bring in. To measure this effect we have assigned a weight of 1 to graduates of the ten leading institutions, 2 to graduates of the 20 other major institutions, 3 to graduates of established institutions, and 4 to the graduates of developing institutions. Measured in this way, we see in Table 20 the average quality of postdoctorals attracted to various institutions. In every field except biosciences the ten leading institutions attract better students than the other schools. For all fields combined, the quality of the postdoctoral decreases with the reputation of the school, but the individual fields show no such neat regularity. The numbers are sufficiently small that many of the per- centage differences are not statistically significant. The other side of the question is how much the reputation of the school at which one takes a postdoctoral appointment is determined by the reputation

•o . 0) i o 2 11 U- voootr^cNnoo co O) a CM 25 «- «- CM CM tt) Q 1 O '5. 1 a. e .o 1 4-J 0 X D 3 1 { 1 1 Q E •-oomoor^r^vco co e 1 "co 1 CMCM«-CMCMCMCMCM CM m *— O +-J t» S g 1 I "O E z r^ocNrxincMooc^ «- So 8 CO r- CM M CO CM •- O Q_ •s i | l« •s 1 Z X j E «-r-ooooinoo * .o h- £ Ul CMCMCM«-CMCMCMCO CM § £ £ i So i/r o c T" Institution I Other Major Number r^incM^cococ3)t LO ..he icstitutio sible index is inaire. •8 r--«-CM «-COCOCM CD O Jj ^ 4_ Q J* X O 1 S °! i 3 f CMcooco^r^coi n * § » 3 ^ z E •S * o i 1 CMCM«-CMCMCNCMCM CM « o) « 5Z 3 CB CL - c CO X 0) S 0) (A ^ "" JT O CO 1 T= Postdoctor f x Numbe 2?g2gSi? § is based on t cot inc.ded, 1, Postdoctori 9 «- «- Ln .E 1 I •s s g • „ „ « 13 COCMCOCDCOLOtt «- 2 ™ c X C T-CMT-•-CMCMCMCM CM i§ § -S c f! ? .t: oi £ d C f CO «- '" c S, 52 O 18 o i_ <u 'o .£ — T3 a Q) O tn > >- .y < •Q 1 • - • - '5 !^ .S "S ?•£ o Ik rr u - 0 CM 8 > E g CO O §?.£ ^ •« o) h o w c LJJ 1 s 1 1 1 1 1 1 1 a •s 1 S .S | | a I iS | | " e0 L. 2 m i 11 £ E1 .9 £ 1 § Q.oOiiiooOoo(/) tl 1 80

81 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS of one's PhD institution. Because of the nature of the process by which appointments are made, one should expect the correlation to be high. Usually, informal contacts between one's PhD thesis supervisor and prospective post- doctoral mentor precede any formal application. This is true even if the post- doctoral is the winner of a national fellowship. Since the weight of a profes- sor's recommendation depends on his own reputation, and since the school's reputation is related to the professor's, it would be expected that equals tend to speak to equals. Table 21 bears out this analysis. The better the reputation of the institution of one's doctorate, the better the reputation, on the average, of one's postdoctoral institution. Again, biosciences provides the exception.17 These results partially confirm Berelson's18 conclusion that "there is a tend- ency for postdoctoral people to attend institutions like those from which they received their doctorate." It is, as we have seen, only a tendency. Approxi- mately half of the postdoctorals migrate to schools of a reputation different from their PhD institution. Field Migration Another aspect of the transition from predoctoral to post- doctoral status is the migration between fields. One of the major motivations for postdoctoral work is to enable a PhD to shift directions from his disserta- tion. Although this need not involve a change of fields, it often does. As one postdoctoral suggested, a change of institutions without a field change permits a person to get a new perspective, to become broadened, and to gain further experience. Of the immediate postdoctorals, 35 percent change fields and 46 percent change institutions without a change in fields19; 19 percent do neither. 17Since the grouping of schools by reputation is dependent on Cartter's study, which ranked schools by the quality (really reputation) of the graduate faculty, one wonders if our results for biology do not cast doubt on Cartter's results in this field. 18.Postdoctoral Work in American Universities, op. cit., p. 56. 19These percentages are subject to some question. The difficulty lies in determining the point at which a change of research topic becomes a field change. There may be some doubt that a physics PhD whose postdoctoral field is cytology has changed fields, if the nuclear magnetic resonance techniques that he is using on tissue in vitro are those that he used in his thesis research on impurities in semiconductors. On the other hand, his class- mate whose thesis also dealt with the same techniques and the same class of materials and whose postdoctoral research is low temperature physics would probably be considered by most to have changed fields, particularly if he were learning cryogenic techniques anew and were concerned now with the properties of He. Unfortunately, the information avail- able to us forces us to make the opposite decision in both cases. Each respondent was asked to identify both his PhD and his postdoctoral field by means of a three-digit code from a specialties list attached to the questionnaire (see Appendix B-l). We determined a subfield change by observing any change in the three- digit code. Both men in the above example would have indicated solid state physics (code no. 160) for their PhD field. The former would have given cytology (code no. 522)

o 2 8 g Q •s 81 o c .o i co M I £ tB <§ LU CO CMCMCMtcNCMcNCM CM CM n CM' Sin o LD in 01 01 co (o O «- LO ft CM «- CM «- CO CM co en cq q in oq PI CM CM' CM' «-' CM' CM' CM «-' «-^ CM .9- o o c> oo oq co co en CM CM' CM' T-' »^ CM' CM' «-' 8 «-CM«-«- en r-; oo CD oq CM CM co en •-' «-' «-' •-' •-' CM' CM •-' ^ .i >• t o . liltilil EoOuieoOoBc o 2 y .5^ o o 82

83 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS Field changes are especially common in the rapidly developing research areas. A notable example of an investigator who undertook his postdoctoral in a field different from that of his PhD is James Watson. In his vivid memoir, The Double Helix,20 he describes his experiences in attempting to learn bio- chemistry after his doctoral work in genetics. The breakthrough which brought him his Nobel prize occurred in an interdisciplinary field. One of his co-workers with whom he shared the prize, Francis Crick, was a physicist. In Table 22 some data are presented on field changes by immediate post- doctorals. The fourth column contains the numbers of postdoctoral s with PhD's in one of the major fields listed on the left who took their postdoctoral appointments in another of those major fields. The third column gives the number of the postdoctorals who received their PhD's in one of the major fields and who changed subfields within the major field in moving to the post- doctoral. The reason that biochemistry shows no change in this column is that biochemistry is a subfield with no finer structure in our specialties list. (See also Figure 5.) Chemistry, engineering, and the biological sciences (with the pronounced exception of biochemistry) all suffer a net loss in PhD's to other fields. Bio- chemistry is the major gainer from chemistry and the other biological sciences, while physics picks up most of the engineers who change fields. The following table displays the migration of the immediate postdoctorals among gross field groupings; the number in parentheses is the number who have remained in the same subfield: Postdoctoral Field EMP Life sciences Other fields The gross field move is an extremely limited occurrence. Of the 66 making the transition from engineering, mathematics, and physical sciences (EMP) to the biological sciences, 49 are chemistry PhD's and 38 of thesp changed to bio- chemistry. Similarly, of the 13 going in the opposite direction, 11 are moving to chemistry. Finally, of the 25 who received their PhD's in other fields and who are taking their postdoctorals in the biological sciences, 16 are psychology PhD's. for his postdoctoral field while the latter would again have written solid state physics (code no. 160). Since there is a limit to the amount of fine structure one can permit in a list of specialties, we will have to be content with the possible distortions that are intro- duced in this way. 20James Watson, The Double Helix, Atheneum, 1968. PhD Field EMP Life Sciences Other Fields 1,107(867) 66 13 13 721 (451) 66 10 25 211 (122)

s o o X s £ to D E IW S •o •55 I 1 T3 1 U- S LU CQ ilil fe 4f B S I'll =.«s S f E 8 c ||| 1 s 1 .c S £ f££ •= T3 Q « 2 Ml fii A 8 5 £ yt «-+ cNcnooo m tn <- en ^ ^ T- «- CM C5 •- Sr^ co CM en r- a> in oo «- «- r^ oo t- in co ocoor-cococoo •2 8 S § it .§ 3 S i I > E s . 2 gtta$gij2S ?£.£ c".2'£-£5 Q-OOujcoOOO CM CM" 84

86 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS FIGURE 5 Percentage of Immediate Postdoctorals, by Field of PhD, Who Changed from the Field or Subfield of Their PhD.s. Percentage of Immediate Postdoctorals Who Changed from PHYSICS CHEMISTRY Q £ O 0 OTHER PHYSICAL SCIENCES ENGINEERING BIOCHEMISTRY3 iZ OTHER BASIC MEDICAL SCIENCES OTHER BIOSCIENCES OTHER FIELDS 1 0 20 40 60 3Biochemistry is a subfield in the specialties list; therefore all changes are at the subfield level. Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire The predominance of field-changing in the biological sciences is probably related to the specificity of those fields in comparison with the physical sci- ences. In the latter the mathematical nature of their principles allows students and investigators an economy of categorization. Many diverse systems and phenomena can be subsumed under a few laws or mathematical statements. As yet the phenomena with which the biological sciences are concerned have not been resolved to the point that they can be discussed in precise quantita- tive terms. Consequently, discoveries and understanding on one biological sys- tem may not be transferable to another system. One young English geneticist explained that her postdoctoral in biochemistry was not so much a change of fields as a change of proteins. Such considerations are important in making crossdisciplinary comparisons.

86 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Kinds of Support Although there are four major support mechanisms for postdoctorals, only three play any role for immediate postdoctorals. These are the fellowship, the traineeship, and what we might designate generically as the project associateship. The fourth, the sabbatical, relates to the older postdoc- toral on leave from an established position and is usually available only after an extended stay at that position. The immediate postdoctoral, almost by defi- nition, is excluded from this latter opportunity. We shall discuss in Chapter 9 the stipends associated with these mechanisms and their policy implications. Here we shall merely describe the differences and the similarities among them and their distribution by field. Historically, the dominant mode of support and encouragement has been the fellowship. Generally speaking, the fellow has been chosen in a national competition by a select panel. From the beginning, however, there have been locally sponsored fellowships at host institutions. Both approaches are similar in attempting to provide a period of relative financial security for the young postdoctoral during which he might gain increased sophistication in research. Except for the local programs, of course, the fellow may take his appointment at any host institution that is willing to provide him with space and where a suitable mentor is willing to supervise his activities. This provision has almost always (sometimes by the conditions of the program) led the fellow to an aca- demic institution or to a nonprofit, quasi-academic research institute, although not necessarily in the United States. The applicant must propose a plan of research, and this plan, along with letters of recommendation and copies of publications, constitutes the materi- als on which the selection is based. Much leeway is allowed in the alteration of the research plan once the tenure has begun in order to permit local conditions and unforeseen changes of direction in research findings to determine the most fruitful course of the investigation. It is the hope of these fellowship programs that they are providing assist- ance and encouragement to the most promising young scholars and that their programs, like the earlier National Research Council program, which has been acclaimed for its success, will promote excellence in research in this country. Another support mode-limited almost entirely to the life and medical sci- ences-is the traineeship. The competition here is among groups of faculty or even whole departments to obtain a training grant, usually from NIH, for the purpose of creating a cadre of manpower trained in a particular field. The pro- posal to the federal agency from the department describes the national need for people with a particular background; enumerates the facilities, research per- sonnel, and research activities of the prospective training institution; and re- quests funds both for stipends and for training expenses, including research equipment and supplies. Often the proposed program extends from the predoc- toral level through the postdoctoral level, although a man is relatively unlikely

87 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS to stay at the same institution for work at both levels. The postdoctoral trainee is selected by the training institution rather than by an extramural panel; in particular he is selected by the faculty participants in the training grant on the basis of credentials and letters of recommendation similar to those required in the fellowship programs. The third major support mechanism is the project associateship (often called the "research associateship"). In this case the competition is among fac- ulty investigators for support of their research. The postdoctoral enters into the picture when the successful investigator is awarded sufficient funds to per- mit him to hire people at this level. Gaining an appointment as a project asso- ciate tends to be a less formal process than applying for a fellowship. An appli- cation for appointment generally follows an informal decision by the faculty member to make the appointment. This decision is based on correspondence with the PhD adviser of the prospective project associate in which the strengths and weaknesses of the candidate are explored. Papers by the candidate and a re.sumd of his thesis are also examined, but his area of research is established by the faculty investigator who is bound by the specifications of his grant or contract. Any formal application is filled out for the purpose of obtaining approval by the university administration to ensure that the project associate will be paid.21 From the point of view of the granting agency, of the university administration, and often of the faculty mentor, the project associate is an employee. In principle, then, the three mechanisms can be said to support the inde- pendent researcher, the research student, and the research employee, respec- tively. From these descriptions one can understand the fellow and the trainee as two different kinds of postdoctoral s in our sense of the word, but the case is less clear for the project associate. He is included because in practice the dis- tinctions of principle only partially survive. Whatever the motivation of the funding agencies, and however clearly they perceive the particular need that their funds are intended to satisfy, the postdoctoral s and the faculty are rela- tively indifferent to the mode of support. The critical concern of the postdoc- toral is to work with the particular faculty member. The major interest of the faculty is to have junior colleagues. The various mechanisms are used to maxi- mize success for both participants. As seen from the vantage of the terminal-year graduate student who desires to become a faculty member at a major university, there are two principal routes. The first (and less likely) is to be hired immediately after his doctorate as an assistant professor at a prestige institution. This does occur, as can be seen from Table 14 (p. 68), although it occurs infrequently. In physics, chem- 21 The process described here is typical but not universal. In Chapter 6 we will examine the situation in more detail.

THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION istry, or biology, only one in five faculty members appointed as assistant pro- fessors at the ten leading institutions came directly from graduate school. Assuming that each school appoints four men each year at that level in each field, then eight people in each field qualify annually with the PhD alone. For those to whom this opportunity presents itself, the postdoctoral fellowship is probably less attractive. The data in Table 13 (p. 66) indicate that their teach- ing loads are small and their research opportunities are large. Since these for- tunate few are probably the most able researchers in their PhD class, their de- fection from the fellowship applicants means that the fellowship programs are not supporting all of the very best. This loss is only to the fellowship program; both research and higher education are served by their employment. The other route to faculty status at a major institution is to be awarded a postdoctoral appointment. Winning a postdoctoral fellowship gives a man a number of advantages, including prestige in applying later for an academic position or for a research grant. But postdoctoral fellowships are not easy to get. Only one in nine applicants was successful for the 1969 fiscal year in the NSF program. If the faculty member with whom he wants to work has project associate funds, it may be possible to proceed informally through his PhD adviser to a guaranteed position.22 Nothing much is lost if being a project asso- ciate entails much the same experience as being a fellow. Although exceptions exist, the project associate is usually given more free- dom than the employee status would imply, and the fellow has less freedom than the grantors intended. The faculty member is seldom comfortable in the employer-employee relationship and prefers the master-apprentice interaction instead. His research support is seldom so narrow in description that a spec- trum of activities may not be allowed under the terms of the grant or contract. His own interests probably lie in several areas simultaneously. If his project associate has ideas of his own, he is permitted to follow them if they fall within the scope of the faculty member's interest. On the other hand, the fellow will often discover that unless his research interests coincide fairly closely with those of his mentor, he will get little help. Few institutions have free space not assigned to faculty members, and conse- quently, the fellow's research must conform somewhat to the facilities avail- able to his mentor. Since the fellow is not likely to have sufficient funds to pay for his research expenses,23 he is dependent on his mentor for support from the mentor's project grant or contract. Such funds, however, are legally used only when the research is appropriate to the project. Indeed, once the position is guaranteed, he may be urged by his prospective postdoc- toral mentor to apply for the fellowship anyway. If he wins it, the mentor will be able to hire a second postdoctoral with funds released. Some programs provide up to $1,000 for expenses. Not all of these funds are neces- sarily available to the fellow, and even if they were, research costs often exceed this amount.

89 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS The net result of these conditions is that the distinction between the proj- ect associate and the fellow is lessened. When a research group contains both types, a faculty member is especially loath to insist on differences. This does not mean that there are none, however. The project associate, as an employee of the university, usually shares in the fringe benefits of health insurance, park- ing privileges, and even retirement plans. On the other hand, the fellow is granted exemption of tax liability for up to $3,600 of his stipend. The faculty often see the training grant as a means of increasing the num- ber of postdoctorals in the department. Since the award is at their disposal, the traineeships are used to attract able postdoctorals in the department. Once there, the postdoctoral may be urged to apply for a fellowship. If he is suc- cessful, a traineeship is released to bring another postdoctoral to the group. Although this shuffling from traineeship to fellowship or even to project asso- ciateship makes the impact of the training program difficult to measure, the individual continues to receive the experience that he sought, the faculty re- ceive the assistance that they desire, and the manpower pool generally receives another independent researcher. None of the above destroys all differences between the three modes of sup- port. It merely tends to make them less severe. Fellows, after all, have been selected in a national competition and tend, on the average, to be much better researchers. Some faculty want only fellows in their group for just this reason. They argue that the national committees can do a better job of selection than the individual faculty member. As one put it, "I insist that the people who come to work with me be good enough to win in a national competition." Of course, not all faculty members have the reputation to attract fellows. Those who do tend to be at the prestige institutions. The project associate may be a graduate of the host institution who has been kept on since he was offered no suitable outside position. As we have seen from Table 19 (p. 78), he may not be as able as the man from the out- side. It is probably true that, on the average, the project associate is not as promising as the fellow. Even if this were true, the overlap in ability of the two groups is extensive. Not only is there little difference in treatment among the fellow, the trainee, and the project associate once they are at the host institution, but the situation is confused further by the lack of consistency in the use of titles at the host institutions. Respondents to our census of postdoctorals were asked to give their title and, separately, to check the type of appointment they held. The latter options were fellowship, traineeship, sabbatical, position sup- ported by project funds, and other. The following table gives the relationship among their responses24: 24 9,971 out of the 10,740 respondents provided both title and type of appointment.

90 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Type of Title Appointment Fellow Research or Project Associate Other Fellowship 3,572 726 796 Traineeship 316 105 604 Sabbatical 23 43 93 Project associateship 326 2,030 625 Other 188 184 340 Total 4,425 3,088 2,458 It would appear that many postdoctorals neither know nor care what type of appointment they have as long as their checks arrive on time and they are able to do the research they want.25 It is with little confidence, therefore, that we present the data on the types of appointment held by the postdoctorals. Table 23 gives the distribution of all postdoctorals among the types of appointment at U.S. academic institutions as reported by the postdoctorals and as reported by the departments.26 The departmental response is probably accurate and the lack of agreement between the two sets of data reinforces the comments made above concerning the postdoctoral attitudes toward the vari- ous modes of support at academic institutions. It is apparent that regardless of nationality, postdoctorals prefer to consider themselves as fellows, no matter what their real status may be. The reasons for this preference are many. They include the prestige of being a fellow, ignorance of the distinction between the various types, and the confusion of titles. Concentrating now on the departmental response, and realizing that approxi- mately 62 percent of the science postdoctorals are immediates, it is apparent (Figure 6) that postdoctorals in the engineering, mathematical, and physical sciences have fewer opportunities for fellowships than those in the biological sciences and almost no opportunity for traineeships. The burden of postdoc- toral support in the EMP fields is on the research grant mechanism. This ex- plains why current cut-backs in research funding affect the postdoctoral situ- ation in the physical sciences so much more severely than in the other areas. It will also have a serious impact on the foreign'postdoctoral in all science fields. Only in the humanities and social sciences ("other fields") are the for- eigners less often project associates than the Americans. The lesser dependence on the training-grant mechanism in the EMP fields correlates with a lesser interest among the faculty in these fields in the merits 2SOne young biologist told us that he had avoided a project associateship because he thought it would commit him to his mentor. Earlier in the discussion he had complained that his mentor had ignored the project outlined in his fellowship application and had required the fellow to work in an area that interested the mentor. 26 Not all departments were asked for data so that the numbers of postdoctorals as given by the departments need not agree with the numbers from the census. On the other hand, it is difficult to reconcile the change in the ratio of foreign to U. S. between the two sources, unless our response rate from foreigners was better than from U. S. citizens.

1 -ii r^cooo top•-CM r-o«-co inococn incooco COOO•-CO CMCotr-, CMCOOCO CONOCD r^ CM co *!• 03° «- Tf «- O) LO CM CM §~ H ZC 0 c I 1 6 «-«-r-oo r-oocor- inr^oooo 1 t a *" «- CM CM CM CM CM CM 9- a i] £ S < •8 ll S«-CO•- •-CMinCO CMOCOO) t O LO CM g | Q- M, l 'S •*: i 1 1 M a 1 £ — t» .£ o 0 S ? CM«-«-O ooo^in coco«-oo o in r- o I £ CM CM CM «- t «- n "D •s E S g ^ a o. w c 1 Q D g o "D 1 1 o) r^ o) ^ ^ co in o r^« o co in CM co «- «- in ^ (O CM CD r«« CN co co '» en CM in in i- CD i M M 3 S E o a O -§ V EC E E EC c c O trt • .5> .S1 .? .? .2* .? . 'a> • 'a O) O) £ 0•' 5?-O^O i^'O^O ^O'O M § W S » 8 z: D|I — \ i. L|. —J U_ m O o a. •i. s i- 2 2 B 1 » i Appointmen Cource of Dai — eo — <0 — n3 s, acd physical ciectif ic Persoi (Q *J B rB El" oo o « g w o c o c o c o C o C o t ! 1 1 1 jnses ^ (0 o a o a tn Q- tn Q- trt Q- <n O- 0 « 0. Q >*~ Q. £ Q £ S £ S 0 (/J cS C O 0) Q" E o, n 0) 0 K 1 •o i JZ ^ CO O y it s i p di ^ 7 c GC CM -0 Uv> 1 1 ' * M I i 1 o 111 i 5 .2 » C D 1— O £ ui a 2 0 c^ (/> 91

92 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION F1GURES Types of Appointment of Postdoctorals at U.S. Academic Institutions. Citizenship ^. . U.S. Foreign EMP BIOLOGICAL SCIENCES 75 Fellow- Trainee- Project Other ship ship Associate- ship Fellow- Trainee- Project Other ship ship Associate- ship o UI 75- MEDICAL SCIENCES OTHER FIELDS 50 Fellow- Trainee- Project Other Fellow- Trainee- ship ship Associate- ship ship ship Source NRC, Office of Scientific Personnel, Postdoctoral Departmental Questionnaire Project Associati ship

93 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS of the postdoctoral experience in preparing PhD's for research in the university setting. In preparing a training grant proposal the faculty member is forced to consider the manpower needs of the country and to design a program to help fulfill those needs. The research grant proposal, on the other hand, puts the emphasis upon accomplishing a desired research goal. Any support for post- doctorals included in the proposal must be justified by the level of effort that the research requires. Although both mechanisms are research based, the latter recognizes the postdoctoral as a "means," while the former considers him an "end." In actual practice, the situation is mixed; the trainee is also a research assistant and the project associate is receiving valuable experience. Both of these by-products, however, are less intentional than fortuitous. A mechanism is called for that ties these two together. Duration of Appointment Postdoctorals spend varying amounts of time on their appointments. A few remain for an indefinite period, becoming, in our terminology, "long-term" postdoctorals. Most immediate postdoctorals, how- ever, tend to stay three years or less, with the overall average being 1.6 years and with over 80 percent staying less than 2.3 years. Contrary to the general opinion, the foreign immediate postdoctoral does not spend any longer time on appointment than his American counterpart. By field, the humanist spends from 0.6 years to 1.4 years, the chemist from 0.8 to 2.0 years, and the biochemist from 1.3 to 2.5 years. All other fields lie somewhere between the extremes. These figures, not surprisingly, do not dif- fer significantly from those suggested by the faculty as optimum either for the postdoctoral's sake or for the department's. In both cases the duration recom- mended is from 1.4 years to 2.8 years, with biochemistry at the upper end and chemistry at the lower. At one major institution the chemistry chairman as- serted that one year of postdoctoral study was enough. "The second year does not double the benefit of one year of postdoctoral study." Another chemistry chairman echoed this impression and added that "the first year rewards the postdoctoral; the second year rewards the mentor." A third chairman, also from chemistry, introduced the important proviso that the crucial determinant is that the postdoctoral stay long enough "to do something." There is, however, much variation in the departmental attitudes toward establishing limits on the length of time that a postdoctoral may spend in the department. The top institutions tend to have a policy on duration more often than the lesser institutions and the EMP fields more often than the basis medi- cal sciences. Only 77 departments out of 915 that reported having postdoc- torals limit the tenure of postdoctorals and in no field did more than 18 per- cent of the departments report such a policy. Of course there are other constraints on the duration of appointment. Fel- lowships are generally tenable for one year, although some programs permit a

94 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION renewal for an additional year and sometimes longer. The postdoctoral's own career interest is a major cause for limiting the appointment. Most are eager to get onto the tenure ladder as assistant professors. They are quite conscious of their artificial status at most universities where they no longer think of them- selves as students but are not faculty members either. The actual duration is likely to be a compromise among a number of forces including the postdoc- toral's desire for faculty status, the mentor's desire for expert research assist- ance, the progress of the particular research problem involving both of them, and the availability of a suitable next appointment for the postdoctoral. Intermediate PhD Postdoctorals The intermediate postdoctoral did not take his appointment immediately after his doctorate. Presumably he was employed elsewhere in the intervening time and then made a decision to pursue postdoctoral study. Postdoctorals in this category are of two different kinds: those who are on leave from their previous positions, and those who have resigned from their previous positions and are making a transition to new employment. The former are in a sense taking an early sabbatical, perhaps to escape the distractions from research of their regu- lar employment and possibly to achieve new competencies in their research fields. In the humanities and social sciences, especially, this is the time when the thesis may be transformed into a book for publication. For the scientist who went immediately to an academic position following his PhD, the tempo- rary leave allows him to pick up his research, which previously had to compete with the preparation of lecture notes and with the other demands on the time of a new assistant professor. For others the postdoctoral appointment is a mechanism for upward mobil- ity in the academic world. Having taken a position in a lesser institution (from which it is difficult to appear attractive to the better schools), the young PhD takes a postdoctoral appointment and essentially starts over again in the em- ployment market. Thus, the postdoctoral position provides for the system a means of individual renewal-a second chance. This is particularly important for the PhD in science who, having tired of being a student, opted for imme- diate faculty status. Without the postdoctoral experience he is unlikely to re- ceive an appointment at an institution of high prestige (see Table 14 on p. 68). Were it not for the opportunities for an intermediate postdoctoral appoint- ment, such a man would be unable to move to a more desirable university. These remarks apply mainly to the U.S. citizen. For the foreign citizen the intermediate postdoctoral appointment, in addition, may be simply a delayed immediate postdoctoral position. The difficulties of arranging appointments from abroad, as well as the problem of acquiring travel funds, may cause a

95 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS year or two to pass before the foreign PhD can finally get to the United States. In the meantime, of course, he has been temporarily employed. In every sense, except for the formal definition, he is an immediate postdoctoral. The number of intermediate postdoctorals is not very large. Table 24 sum- marizes some of their characteristics. In the sciences over 70 percent of the intermediates are foreigners, while in the humanities and social sciences less than 30 percent are from abroad. Altogether, 72 percent of the U.S. interme- diates are at universities; 14 percent are pursuing postdoctoral work abroad. In all fields combined there are only 326 U.S. intermediate postdoctorals; a small number when one realizes the important function postdoctoral study may play at this career stage. Table 25 gives the leave status for intermediate postdoctorals. Since migra- tion to a university is relatively difficult from a nonacademic institution, it is not surprising that a fair number of postdoctorals are on leave when taking their appointments outside the university. Again, those in the humanities and the social sciences demonstrate behavior much different from the scientists. They are much less likely to use the postdoctoral appointment as a means of changing institutions. Their main interest in an appointment is that it tempo- rarily releases them from other time-consuming duties connected with an aca- demic position and that it enables them to devote themselves to research. The importance of postdoctoral study at this time for these disciplines is indicated by the relatively large proportion of intermediate postdoctorals in the social sciences (other than psychology) and in the humanities. Although only 2 per- cent of postdoctorals at the immediate level are in these fields, they are the fields of interest of 18 percent of the postdoctorals at the intermediate level. Eighty-eight percent of the intermediate postdoctorals in those fields are on leave. People who delay their postdoctoral appointment until the intermediate stage have had maturing experiences beyond their PhD training. Consequently, it is difficult to measure their quality compared to the immediate postdoctorals. One would expect that, having tasted regular employment, they have a clearer idea of what they want to achieve during their postdoctoral study. The matu- rity that some years out of graduate school have given them may compensate for whatever initial differences separated them from their colleagues who went immediately into postdoctoral study. When we compare the two groups with regard to their total baccalaureate-to-PhD time lapse, the differences are small but interesting. The mean time lapse for intermediates in the physical sciences is 6.1 years, for the basic medical sciences 7.1 years, and for the other biologi- cal sciences 6.8 years. In each case the intermediate falls midway between the immediate who migrates and the immediate who stays at home (see Table 19, p. 78). In all cases the intermediate shares with the immediate about a 1 %-year advantage over the PhD population generally.

ill fl" LO OO Of CO^ CDO «-t Ooo «- oco CMr^ 1 «- CO •- «- •- CO LO Q. J> r-o ooo ^o ooo to * c 5 N § b 3 c •a i 1 to c c •o •; > .£ I a ooCD t--LO «-o coo coin i-L $ "8 wf ^ .2 1 •a V, "O3 3 I •j-< "g O O O«- O«- OO T-O O«- c ^ 1 a 1 •5 X 1 1 1 O) c o 1 O cot CD«in CMf oo ooin 1 tB CM «- •- «- (A v» E <a i 0 (J i 8> > 2 2 B 'M o t> i I o 1 E CL OCO OO (0(0 COO CMCO $ e0 r-.oo (ooo LOCO coe n r-oo iA W» 0) O 8 °: 1 1 1 E i -S1 lit 1 s c c c c c S.2 | > c o_ O) O) Ol O> O) ^o "^o "'o "? o ^o DLL DLL DLL DLL DLL *! i •5 o c J E " -a «> § '+* vT c 'o | 0 e0 to 'I JS K o 1 fc C 0) OJ CD U b 1 i •& £ 'o ^ g S E^ ° CT'" ° f 1 i ^ 1 •» z CM 0 "en u> 'o "D <u QJ 0> -n LU ! 1 1111 .|1 S c g 5 3 i- £ LU bo 5 O n i 0 ift -Q (/5 EMP* Biological s MedicaI scre Other fields 96

97 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS TABLE 25 Percentage of Intermediate Postdoctorals at Academic and Nonacademic Host Institutions on Employment Leave, by Field and Citizenship Percentage of Intermediate Postdoctorals on Employment Leave U.S. Citizens at Citizens of Foreign Countries at Academic Nonacademic Academic Nonacademic Postdoctoral Field Host Inst. Host Inst. Host Inst. Host Inst. EMP3 19 58 55 68 Biological sciences 18 30 50 67 Medical sciences Other fields6 20 71 0 50 57 0 67 56 Total 34 45 54 66 ^Engineering, mathematics, and physical sciences. "Includes social sciences and humanities. Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. As with the immediate postdoctorals, the usual period of time spent as a postdoctoral is from one academic year to two full years. It is shorter in the humanities and social sciences, where it seldom lasts longer than a year. Bio- chemistry is the longest, with two thirds of the intermediates spending from 1.2 years to 2.8 years on their appointments. For both the immediate and the intermediate the appointment is limited both by the availability of funds (es- pecially outside the natural sciences)27 and by the availability of a suitable position. The intermediates look forward to academic positions even more strongly than the immediates do. Many, of course, are returning to the ones they left; others to better ones. The striking difference is in the proportion heading toward a college rather than a university. Whereas 7 percent of university- based immediates were anticipating colleges as their next employers, 15 per- cent of the intermediates are planning on teaching at a college. It is possible that most of these are on leave from colleges and are simply returning. Gov- ernment and industry are selected less often by intermediates than by imme- diates, which probably reflects the preselection of the entire group of interme- Even in the physical sciences, which are dependent almost entirely on the National Science Foundation for fellowship support, there is little money for the intermediate postdoctoral. Of the 120 fellowships awarded in the 1968 NSF regular postdoctoral pro- gram, 86 went to persons who had not finished their doctorates at the time of their appli- cations. At most, the remaining 34 fellowships went to intermediates in all the fields cov- ered by that program.

98 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION diates. Most were previously in academic positions and are not changing their minds. Senior PhD Postdoctorals Whatever doubts may exist with regard to the necessity of postdoctoral study immediately following the PhD, study or research leaves for mature scholars are universally recognized as important and desirable. After several years of teaching, research, and administration, the senior investigator is often in need of both a change of pace and the stimulation of new surroundings. For some, a leave of absence permits them the leisure to complete a book on which they have been working. For others, it is an opportunity to work with a colleague at another institution. For still others, the absence of regular duties allows them to visit libraries and other sources of original documents to pursue their research. All look forward to the experience to renew their ability to cope with their normal responsibilities. Institutions recognize these needs and support them generally. Often the support is limited to granting leaves without pay to their staff. Many univer- sities and some nonacademic institutions have formal sabbatical leave pro- grams. The usual pattern is to provide a half-year's salary every seven years and to require no services for a period up to a year. The employee has the option of receiving full pay for one semester or half-pay for the entire aca- demic year. The sabbatical leave is seldom automatic and is granted only on the submission of a proposed plan of study and research. It is understood that the professor on sabbatical may supplement his income through research grants and fellowships, but he may not be paid for services during his leave. Although some senior investigators make use of their leave to acquire new skills, more often their motivation is to have free time to exploit their already considerable talents. They do not think of themselves as postdoctorals and it is likely that our estimates of their numbers are low. The formal fellowship programs, such as the NSF Senior Postdoctoral Fellowships, make the identi- fication with postdoctoral study and research. Others, such as the Guggen- heim Fellowships, are designed to support scholars with or without the doc- torate. Humanists supported by a grant from the American Council of Learned Societies (ACLS) or social scientists who have been awarded a grant by the Social Science Research Council probably do not perceive of their activities as being "postdoctoral" in any special sense. Part of the difficulty in estimating the numbers of this group arises from our definition. If a scholar receives an ACLS grant that supports his research part-time during the academic year while he maintains his pedogogical duties, his situation is akin to that of the physicist with support from a research contract. If, on the other hand, he is

99 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS released from teaching responsibilities to pursue his research, he becomes, in our definition, a postdoctoral. Assuming, nevertheless, that our data are representative if not complete, they show that the humanists and the social scientists make much more use of the senior postdoctoral appointment than do the natural scientists. Almost half of the American senior postdoctoral s are from fields outside the natural sciences. By contrast, only one in seven of the foreign senior postdoctorals are not natural scientists. Table 26 gives the distribution of senior postdoctor- als among the host institutions. In striking contrast to the rest of the postdoc- toral population, the American senior postdoctoral is almost as likely to travel abroad as he is to spend his time at a U. S. educational institution. In the physi- cal and biological sciences particularly, he will be a visitor in a foreign coun- try as often as he will be at a different U. S. university. While the American senior postdoctoral is at an academic institution abroad, the foreign senior postdoctoral is at a university here. The distribution of senior postdoctorals among the fields may only par- tially reflect the availability of funds. It is also a consequence of the different nature of research in the different disciplines. Most experimental scientists at universities have their own laboratories at their universities. They are likely to slow down their research if they go on leave for a year or less.28 Unless a humanist is extremely fortunate, his "laboratory" is distributed around the country and abroad. Once he has exploited the resources of the local library and whatever materials may be obtained through interlibrary loan, he has need for extended periods of uninterrupted time to write. It may also be necessary to see original-source documents. In either case, his research requires-theleave of absence to become efficient. As we have seen earlier, the scieritist-often finds it necessary to take a postdoctoral appointment early in his career in order to become a productive investigator. For him a later postdoctoral is an enrichment, but seldom a necessity. The humanist is already competent in the techniques of scholarship when he receives his doctorate. His immediate need is rather for growth and contemplation, often enhanced by classroom confron- tations. The delayed postdoctoral for him is necessary if he is to bring his re- search to fruition. Although the senior postdoctoral may be relatively more important for the humanist than for the scientist, it is still important for the scientist. Especially if he desires to work with or near colleagues abroad, the availability of fellow- ships is crucial. The evidence is that there is not nearly enough money to sup- port postdoctoral activity in the sciences for the mature investigator. All of This is not true for the theoretical physicist or mathematician. The association with colleagues at a different institution can be extremely fruitful, even if the duration is relatively short.

ill (DO) O CM ^ O) 9 Oi CO O £ z c c I Q. £ in ^ o o in o 1 C O M £ C N 3 'a | c o T3 0 CO O •- CM O «- C a U •a x 0> <«• n O £ 1 K w ^ 3 c O CM O O O O ,o If) — '•p a D § v> o C ^5 .ti w .o M & | o 5 1 in cD T- CM in o 05 CM c £ •S C £" tn 2 • g o o 5 1 'E 3S 9i Sg S 1 c c c O) O) CD 1t- ^ D u. D u. D u. 0 o 5 ! 2 8 a> o 8 s 8 .1 o j 1 m '5 S CO •o 0- -2 =0 H £ uj 5 5 ino (Min CM in •* 10 c N OO CDCM inoo Du. com inoo Z) LL .5 *s jg.i Is El ^Ocgiceer Iccludes 100

101 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS the national programs have more applicants than awards. In the 1968 compe- tition the NSF Senior Postdoctoral Program had 384 applicants for 55 awards.29 The NATO program had 462 applicants for 39 awards. The human- ists and social scientists face similar shortages in awards from their respective sources of support. Long-Term PhD Postdoctorals Of major concern to those entrusted with public policy questions as they relate to science education and research is the matter of the long-term postdoctoral. Whatever the values of postdoctoral activity, they seem to some observers to be abused by those individuals who make a career of being a postdoctoral. At a time when both higher education and industry are bemoaning the insufficient supply of trained manpower, the whole of postdoctoral education is open to discredit by the failure of the "eternal postdoctoral" to take a "real" position. A number of considerations must be borne in mind. As we have seen, the average postdoctoral spends less than two years on his appointment before tak- ing a more permanent position. Although the postdoctoral phenomenon intro- duces a delay in the flow of manpower, in a steady-state situation the flow is undiminished for the bulk of the postdoctorals. In fact, it is difficult to isolate the truly-perpetual postdoctoral. As indi- cated earlier, he may not have responded to our questionnaire, since he per- ceives himself to be a permanent employee rather than a temporary postdoc- toral. Furthermore, there may be other factors in perpetuating a postdoctoral career other than the reluctance to leave the academic research laboratory. If we examine the research groups at universities, we find that there are a number of different kinds of people involved, ranging from graduate students (and occasionally undergraduates) through senior faculty. In addition there are immediate and intermediate postdoctorals who are transient members of the group. Occasionally a senior scholar will be a temporary visitor. There are also the more permanent professional research staff. Some of these are techni- cians with varying degrees of formal training and others are holders of the doc- torate who have chosen the academic research environment as their career loca- tion. This latter group is the "unfaculty" mentioned earlier. They occur pri- marily at the major institutions where the level of federal support of research is sufficiently massive, permitting the expectation of uninterrupted employ- ment over an extended period. The long-term postdoctoral may be identified in part with this professional research staff, although he may also exist in less prestigious institutions. 29 In 1969, due to the budgetary stringency, the NSF found it necessary to drop this program altogether.

102 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION In a study made in the spring of 1967, Kruytbosch and Messinger30 exam- ined the situation, as "observant participants," of the professional researcher at the University of California. The document is not free from bias and was conceived to influence policy, but it yields an interesting and informative pic- ture of both the problems and the activities of the "unfaculty." The report raises important questions about the place of research at a university and, al- though the authors plead for more formal recognition and acceptance by the administration of these "temporary" and somewhat unofficial members of the community, they suggest why some of the long-term postdoctoral s choose that status over being a faculty member at another (and lesser) institution. These people are strongly committed to research and aspire to faculty posi- tions at major institutions. The opportunities to do the kind of research that they desire are limited to a few centers. To leave those centers would require either a change of research emphasis or a diminution of research activity. Given these alternatives, they prefer being unrecognized persons at a research center to having full faculty status elsewhere. Even better, they would like their present status formalized with all the privileges of the faculty at their institution. That they are valuable members of the research groups to which they belong is undeniable. The evidence is strong that they participate not only in the research activity but also in the administration of the grants and contracts that support the research. The longer they stay the more they are able to assist the professors. The fact that the project directors continue to find funds to support them indicates the desirability of their presence. If the object is to produce research, the professional researcher is clearly a most important com- ponent. The question may be raised, of course, as to whether there is a more effec- tive use that could be made of these people. Should the funding of research at universities be such as to encourage the practice of retaining professional researchers for indefinite periods? The formulation of an answer to this ques- tion requires the consideration of several complicating issues. In the first place, it must be decided whether the research being performed is itself sufficiently valuable to be supported at current levels. If so, then the question must be faced as to whether the same research could or would be performed outside of the university setting. Furthermore, except for the important question of the relevance of this kind of research to the university's mission, does it make any difference to the purchasers of the research (ultimately society at large) where it is done? If the same people are doing the same research, the alteration of titles is not a real resolution of the long-term postdoctoral problem. Carlos E. Kruytbosch and Sheldon L. Messinger, Unequal Peers: Professional Research- ers at Berkeley, unpublished report, University of California, Berkeley, April 1967.

103 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS If one should decide that, for whatever reasons, the support of these peo- ple should be stopped, the question of whether they would be as usefully em- ployed elsewhere is the next consideration. The mere possession of a PhD does not necessarily qualify a person for an academic or industrial position. Hopefully, a faculty member wants to teach and has the personal character- istics beyond formal learning and research productivity that enable him to relate pedagogically to students. Industrial laboratories require researchers who are stimulated by applied problems and who are sufficiently self-denying to be productive even when proprietary interests forbid publication and public or professional recognition. Furthermore, there are few industrial applications for elementary particle physics or fruit fly genetics. The long-term postdoctoral has been aware that there are other opportunities and for a variety of reasons, both personal and professional, he has rejected them. It may be that his great- est contribution to society is being made where he is, given his peculiar aca- demic training and personality traits. In lieu of a better criterion, given the data available to us, we have desig- nated as "long-term" those postdoctorals who are in their third or later year of postdoctoral work and who are not on leave from another position. Grant- ing the appropriateness of this definition in this area of postdoctoral activity and accepting the probable bias in the responses to our census questionnaire, it is instructive to examine the details of the group of long-term postdoctorals as we measured them. From Table 9 (p. 59) we see that there is much varia- tion in the proportion of the postdoctorals in a given field who are long term. The physical and biological sciences have a larger share than the medical sci- ences, and in the fields of physics, chemistry, and biochemistry approximately one in seven of all postdoctorals are long term. These three fields also have the largest number of postdoctorals, with the result that they collectively account for sixty-two percent of all long-term postdoctorals. The situation is somewhat more complex, however, since the post-MD component by definition does not contribute to the long-term group. If we compute the percentages on the basis of the number of post-PhD's in the field there are some dramatic changes. The medical sciences have a total of only 175 post-PhD's but 27 (or 15 percent) of them are long term. The fraction of long-term postdoctorals in the biological sciences rises to 18 percent while there is little change in the physical sciences. In addition to variation by field there is a strong dependence on sex and nationality. The following table gives the fraction of each group who are long term: U.S. U.S. Foreign Postdoctoral Male Female (Both Sexes) Field % % % EMP 10 33 14 Biological sciences 9 27 14

104 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Thus foreigners are approximately one and one-half times as likely to be long term as U. S. males, while U. S. females are long-term postdoctorals three times as often as their male counterparts. Stated in a different way, whereas 39 per- cent of all postdoctorals in the biological sciences are U. S. males, only 29 per- cent of the long-term postdoctorals are. U. S. women constitute 9 percent of all postdoctorals in these fields, but 21 percent of the long-term postdoctorals. The fact that U. S. males have a greater chance of obtaining faculty appoint- ments in this country may partially explain the distribution of long-term post- doctorals. Many of the women are either faculty or student wives who are not able to receive faculty positions because of institutional rules on nepotism. There are, of course, some women who find the postdoctoral status to their liking, allowing them to do research part-time while remaining a wife and mother. Nevertheless, it is clear that the majority are simply taking the best position that is open to women who want to do research and to live with their husbands and children. This is especially true at institutions not near other re- search opportunities. The foreign component shares some of the same constraints. Language diffi- culties as well as lack of faculty opportunities at research oriented universities for all but the very best foreign postdoctorals probably account for the attrac- tiveness of postdoctoral appointments for those who want to prolong their stay in the United States. If we examine the fraction of postdoctorals coming from countries in the various GNP categories31 (Table 27), we can see that it is a vast oversimplification to speak of the foreign postdoctoral as though he were member of a homogeneous group. Because they constitute more than nine- tenths of the postdoctorals from very low income countries, the Indians have been considered as a separate GNP category. Indians are twice as likely to be long-term postdoctorals as other foreign groups and they account for 27 per- cent of all foreign long-term postdoctorals, while constituting only 13 percent of all foreign postdoctorals. On the other hand, the postdoctorals from coun- tries with fair per capita GNP become long-term postdoctorals even less often than U. S. males. We will examine the foreign postdoctoral in more detail in Chapters. One final comment about the long-term postdoctoral is in order. As one examines Table 27, it is clear that the number of people involved is not large considering that all fields are combined. In the fields with the highest concen- tration of postdoctorals-physics, chemistry, and biochemistry-there are only For the purpose of comparison among countries, per capita gross national product (GNP) is a better (although not perfect) measure of the degree of development of a country than geographic location. Japan, for example, is better grouped with Great Britain than with the rest of Asia, if one wants to measure the sophistication and rela- tive adequacy of higher education in the countries of the world. The countries in each group are listed in Appendix B-3.

105 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS TABLE 27 Number and Percentage of Long-Term Postdoctorals at U.S. Academic Institutions by Sex and Citizenship Postdoctorals at U.S. Academic Institutions Male Female Total Long-Term Long-Term Long-Term Citizenship Total N % Total N % Total N % Foreign (grouped by per capita GNP of country of origin) High 2,587 256 9.9 203 28 13.8 2,790 284 10.2 Fair 275 13 4.7 42 2 4.8 317 15 4.7 Low 450 36 7.8 91 7 7.7 641 42 7.8 Very low 37 4 10.8 1 0 0.0 38 4 10.5 India 520 118 22.7 47 12 25.5 567 130 22.9 Foreign Total 3,869 426 11.1 384 49 12.2 4,253 475 11.2 U.S. Total 3,916 254 6.5 485 113 23.3 4,401 367 8.3 Total 7,785 680 8.7 869 162 18.6 8,654 842 9.7 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. 61,35, and 42 U. S. male long-term postdoctoral s respectively. The national manpower picture would not change significantly if they were otherwise em- ployed. Post-Professional-Doctorates The postdoctoral in the medical, dental, and other professional fields is at once different in his motivations and background from the post-PhD and also much less well defined. Since professional doctoral training is generally limited in research participation, the post-professional-doctorate is not as useful to the faculty as a research associate. It is, in fact, the purpose of postdoctoral activity in these fields more to instill the methodologies and techniques of research than to expand or to sharpen tools already possessed. Unfortunately for the purposes of our study, the definition of postdoctorals in these fields (generally the ones supported by the National Institutes of Health) is not the same as that found appropriate by NIH. Whereas we have restricted our study to those post-professional-doctorates involved primarily in research, the NIH programs are appropriately designed for physicians, surgeons, dentists, and others who desire additional training for a much wider range of activities. Thus, their "postdoctoral" fellowship and traineeship programs include indi- viduals interested in acquiring additional clinical experience in their specialties, working toward specialty-board examinations, and receiving special residency

106 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION experience, as well as those seeking research training. Since some of the activi- ties may also include an exposure to research during a portion of the appoint- ment, the question of inclusion or exclusion from our study becomes prob- lematic. As mentioned in the introduction, our return rate in medical sciences is probably not only low, but more indefinite. Post-MD Notwithstanding this caution with regard to the accuracy of the ab- solute numbers, we believe that the relative data may be sufficiently precise to describe adequately the post-MD. This confidence arises not only from the consistency of our data with the comments and opinions given in a number of interviews but also with the agreement of percentages between our census and data developed annually by the American Medical Association (AM A). Table 28 gives these data for nine leading medical schools32 and for all others. Because of the internship and residency requirements there is no "immedi- ate" postdoctoral in a real sense among the post-MD's. It is difficult, therefore, TABLE 28 Comparison between Office of Scientific Personnel (OSP) Census and AM A Data on Postdoctorals in the Clinical Specialties at U.S. Medical Schools MD-Postdoctorals in U.S. Medical Schools Clinical Specialty and Type OSP Census Data AM A Data of School Number Percent Number Percent Internal medicine Nine leading schools 372 30 500 33 All others 628 70 1,003 67 Total 900 100 1,503 100 Other clinical medicine Nine leading schools 345 26 749 28 All others 962 74 1,934 72 Total 1,307 100 2,683 100 Total Nine leading schools 617 28 1,249 30 All others 1,590 72 2,937 70 Total 2,207 100 4,186 100 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire; AMA. 32The division into nine leading medical schools and all others is admittedly arbitrary, but it is interesting to note that the same mobility picture that was produced by the reputation grouping of the graduate schools is reproduced here. Although the nine lead- ing medical schools produce only 13 percent of the MD's, they attract 28 percent of the postdoctorals.

107 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS to determine precisely what fraction of the MD's produced take a postdoctoral appointment. In 1965-66 there were 7,574 MD's produced by medical schools in the United States.33 We collected questionnaires from 833 medical science postdoctorals who are U.S. citizens and who had received their MD's within the last seven years. Assuming an average of three years for their postdoctoral experience and estimating that our returns represent half of the total, 555 MD's per year seek postdoctoral appointments. This is only 7 percent of the MD's produced, as compared with 20 percent of the PhD's in the natural sciences. The rationale for postdoctoral education in the clinical sciences is simple and agreed upon by all participants, both postdoctoral and mentor, as well as by the medical school administration and supporting agencies: to create faculty for medical schools. It is the general consensus that a faculty member must be involved in research if he is to be in a position to pass on to medical students the latest developments. Consequently, it is imperative that, following a long period of didactic training and supervised practice of medicine, the potential faculty member be not only introduced to research but raised to a level of proficiency and self-sufficiency. Some achieve this goal by seeking a PhD. Table 29 demonstrates that, compared to the post-PhD, the post-MD is much TABLE 29 Enrollment of Postdoctorals at U.S. Academic Institutions in Regular Courses and in Degree Programs by Field and Citizenship Percentage of Postdoctorals Postdoctoral Field Citizenship Taking or Auditing Courses Candidates for Second Doctorate Physical sciences and engineering US. Foreign 41 31 0 1 Biological sciences Post-PhD US. 46 36 0 Foreign 2 Post-MD US. Foreign 78 47 45 19 Medical specialties US. Foreign 46 36 10 15 Humanities and social sciences US. Foreign 46 56 1 6 Source: NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire. "Medical Education in the United States: 1956-66, Journal of the American Medical Association, Vol. 198, No. 8, November 21, 1966.

108 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION more likely to be a candidate for a second doctors degree. Others choose not to undertake such formal training and instead undergo a more or less informal series of experiences both in the classroom and in the laboratory. Neither mode is entirely satisfactory and a number of medical schools are considering radical changes in the entire medical curriculum, partly motivated by the desire to accelerate the training of future medical faculty. The problem of creating faculty members for the medical schools is an acute one and one that is felt by the current faculty and administration in a way that similar shortages in the arts and sciences are not felt. At the end of the 1965-66 academic year there were 672 faculty vacancies in clinical depart- ments in existing medical schools34 and since that time several new schools have been created or planned. In the spring of 1968 one out of six budgeted faculty positions in pediatrics across the country was unfilled, according to Dr. Ralph J. Wedgewood35 of the University of Washington. Although there are 155 unfilled budgeted positions, only 80 pediatric faculty are trained each year. The traditional lockstep character of medical education militates against satisfying the need for faculty. After a student has piled up debts and has acquired a family during four years of medical school, one year of internship and two years of residency, two years of a clinical fellowship and two years in the military, the prospect of two more years as a research postdoctoral (and thus an academic career) must compete with the financial advantages of pri- vate practice. Existing programs of postdoctoral study in the clinical fields comprise both individual fellowships and varying degrees of formal traineeship activities in- volving groups of postdoctorals. Because the postdoctorals enter their research appointments at various stages of their medical careers (ranging from directly out of medical school, through interruption of their residency experiences, to following a year or two as assistant professors in a medical school), their back- grounds are extremely diverse. Consequently their training must be tailor-made. Some will require additional course work; others will require more clinical experience; all will require research training. In spite of their awareness of the need, however, most medical schools have not integrated their postdoctoral activities with their other responsibilities. Faculty involvement in training postdoctorals is almost inevitably on an over- load basis; there is often no lessening of their other responsibilities if faculty desire to participate in the training program. This is particularly critical when special courses are needed that are not in the regular curriculum. An example was cited by Howard Hiatt of the Harvard Medical School. He points out that 3*Journal of the American Medical Association, loc. cit. 35Private communication.

109 THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS most research problems in clinical medicine eventually lead to problems in bio- chemistry. Because of the long time span of medical training it is possible for a post-MD to have studied biochemistry as a freshman medical student before James Watson and Francis Crick unraveled DNA. There is a need for refresher courses. Under present circumstances, a biochemist must develop and teach a new course not for his own students, but for postdoctorals from a different department and almost always without special recognition. The situation is aggravated by the fact that the medical postdoctoral is usu- ally less useful as a research associate than the post-PhD is to his mentor. Only after several years is he able, ordinarily, to contribute actively to the research productivity of the faculty member. Whereas the major rationale given by fac- ulty in the natural sciences for having postdoctorals in their groups is to en- hance the quality and quantity of research, medical faculty seldom mention this reason unless pressed. Then they describe the assistance more in terms of that received from a graduate student than that from a colleague. Most post-MD's, even when intending to do research eventually in a clini- cal field, will take their postdoctoral appointments in one of the basic medi- cal sciences. In this setting they are clearly not as qualified as the post-PhD who probably obtained his doctorate in a basic medical science field.36 In view of the many courses that they must take to arrive at proficiency, the additional requirements for the PhD do not seem as onerous. This, perhaps, explains why 45 percent of the U.S. post-MD's in these circumstances seek a second doctorate. Whether the long additional expenditure of time that this path requires is necessary for the eventual clinical researcher is a matter of discussion and concern among the clinical faculty. Post-DDS In dentistry the pattern of research training differs from both that of the basic medical sciences (PhD) and that of the postdoctoral in medicine. H. W. Magoun37 has gathered statistics on these patterns, which are summa- rized in Figure 7. The typical individual interested in dental research com- pletes the work for his DOS or DDM degree and then pursues a graduate pro- gram leading to a master's degree. Although some schools have PhD programs, these play a minor role. Of the 1,337 persons engaged in graduate and post- doctoral study relating to dentistry in 1966-67, 82 percent were in master's programs, 5 percent were pursuing the PhD, and 13 percent were engaged in postdoctoral study. Magoun suggests that the emphasis on master's programs in dentistry may in part by related to the educational preparation of the dental faculty. In However, the post-MD is generally more familiar with human biology. H. W. Magoun, Graduate Education for Career Teaching and Research in Dentistry, paper presented at Workshop on Graduate Education in Sciences Related to Dentistry, Chicago, 1968. Journal of Dental Education, in press, 1969.

110 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION FIGURE? Percentage of Enrollees, by Level, in Dentistry, Medicine, and Basic Medical Sciences, 1966-67. 100 Master s Doctoral (PhD) Postdoctoral DENTISTRY (N = 1.337) MEDICINE (N = 4,087) BASIC MEDICAL SCIENCES (N = 10,055) Source: H. W. Magoun, "Graduate Education for Career Teaching and Research in Dentistry." paper presented at Workshop on Graduate Education in Sciences Related to Dentistry. Chicago. 1968. 1965-66 only 10 percent held the PhD with or without the professional doc- torate, 21 percent held a master's degree in addition to the professional doc- torate, while 69 percent held the professional doctorate only. He further points out that only half of the dental students in the United States possess the bac- calaureate degree on admission to dental school. The present emphasis in den- tistry on post-professional master's degree programs may rest in part on the limited preparation of many dental graduates for more advanced graduate work.

Ill THE POSTDOCTORAL IN U.S. ACADEMIC INSTITUTIONS Although the National Institute of Dental Research has recently supported the establishment of a number of dental research institutes in universities over the country, the situation today is that the post-DDS is a minor participant in postdoctoral study. The statistics that follow will include him with the post- MD without altering significantly the meaning of the results. Post-professional-Doctorates Combined (Post-MD's)38 The post-professional- doctorate does not differ significantly from his PhD counterpart in his choice of postdoctoral host institution. The university attracts three quarters of the post-MD's and the government and private hospitals account for most of the remainder. Figure 8 shows the distribution of post-MD's by host institution, by field, and by degree level. Virtually all (95 percent) of the post-MD's are in the medical sciences and in the biological sciences, although the foreign MD is more likely to be in the biological sciences than his American colleague. The recent American post-MD (within seven years of his doctorate) is almost four times as numerous as the senior post-MD, and only a few of the Ameri- cans hold both the MD and the PhD. This picture is in contrast to that for the foreign component, where one fifth of the postdoctorals hold both degrees and the older postdoctoral is almost as frequent as the younger. Again, this latter pattern is similar to that for the post-PhD population. Table 30 gives the field distribution of the post-MD's in more detail. Among the medical sciences internal medicine and surgery are the major fields, while in the basic medical sciences biochemistry and physiology are the most attrac- tive. Pathology, which has historically been a bridge field, is also popular. Since the postdoctoral programs for the post-professional doctoral are the most self-consciously career motivated, the data on anticipated future em- ployment are particularly interesting. Table 31 gives the choices of the post- MD's by level of degree. For all fields combined there is little difference be- tween the regular and senior postdoctoral. Approximately 60 percent of both groups plan academic careers. Those who hold both the MD and the PhD are more likely to continue in academic medicine and are similar in this regard to their post-PhD associates. The column headed "other" usually describes for the post-MD an intention to enter private practice. A third of the post-MD's do not anticipate a research career. If a man takes his postdoctoral in one of the basic medical sciences, he is much more likely to seek an academic career. Table 31 gives the choice for both biochemistry and internal medicine. Even though an MD does postdoc- toral study in biochemistry, he usually returns to medicine for his research, using biochemical techniques. Presumably the prior commitment to research implied in the selection of biochemistry as a postdoctoral field enhances the likelihood that the man will remain in a research environment. 38Hereafter we shall use the term post-MD.s to refer to all post-professional-doctorates.

112 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Distribution of U.S. and Foreign Post-MD.s by Degree Level, Postdoctoral Field, and Host Institution. U.S. POST-MD.s RECENT MD 78% DEGREE LEVEL FOREIGN POST-MD.s SENIOR MD 20% MD& SENIOR .^ ^*Z- .."-" 1 PhD 2% MD 35% RECENT MD 45% MD& / PhD \ / 20% POSTDOCTORAL FIELD EMP* BIOSCIENCES 22% -^— "" BIOSCIENCES OTHER -.v FIELDS MEDICAL SCIENCES MEDICAL SCIENCES 4% 64% 73% EMP 1% ^ i / X. S y OTHER FIELDS 4% ../ HOST INSTITUTION GOVERNMENT 12% GOVERNMENT ABROAD 2% 2% ^NONPROFIT 11% "Engineering, mathematics, and physical sciences. Source NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire UNIVERSITY 82% NON- PROFIT 15%

f in «- «- 1 £ oo en 5 «- a D a 1 •« U .1 1 1 B 0) o 0 CD T- «- 0 S 3 u. 1 O I "+- 3 *• § 1 1 CO O O) 03 5 CM i 6 3 C * c o 8 I 8 •s * > e I i i 1 ?: SS U- •• i to O 1O * » u. 1 ' * 1 iZ in oo o tn o t*- ^ «- § * 3 3 i M . a : CM CO CM co CM in CM ^ r- in r^ CO CM CM« § 'i 00 CO CM co in in T- CM CM 1 r: co CM jr CO CM ^ CO oor-cN > a O) Q. 2 = X • 03 r^ o co a> 2 <D >.E CD 10 r*« p*• - c 5? u O ^ w ™ t .8 IN .o o-^jC ± o "o 0£ c s LU CO u .2 S .1 -s .1 I •* a - ij ^ {/> ._ Q - >>llf> l« I111I fillI I c 5 -- H |1 •D "D <0 <B » c;? . E^t& ilii :I4| Z Q- ULP- l« * OO « 8 !E it E o o oc z § w 113

s 1 1 incoco r- in ooo* inoio T-cs^ cor^ (oinin ^ooco •— in r^* ^ co <j~ <N 0 3 ° *- z ~ 6 ocoin inr^ cocot^ *«-o 1 •-•-«- COCM COCO^ CMCM^ 3 | - •-OO O«- ^«-CM •-O«- 1 'E O) | C fl) U J 1 "5> 15 | m "+-' (9 LO O) CO r'• O CD ^• CD CM ^ ^ " 1 "i i <o t! V £ a Q O 2 1 I e .2 Q- " •-OO «- M «-cNCM CMCOCO $ Questio 5 •8 > 1 £ 3 •° * 'M i .*• s I o Q i .2 D 1 cooor^ ino ocov CMCMtD oor^r- LO CD coinr^ P^P-CO 0 a. - 0 — Q a Q £ ployment j §Qf §Q ||f °Qf Persoccel, B w ? *^? S w ? S w "? S .2 « S .2 » .2 • £ o to 8 g 9 8 1 8 1 9 8 g Q c DC c o 2 ec (/) DC <A 2 oc (O 5 M LU * C i ? ! Office of « i 1 s 1 1 1 1 tL o" oc T— 75 o n 1 S 5 co I HI ^J 01 <5 H H •1 CO t| § I a « t- £ a I s i2 i « 114

115 THE POSTDOCTORAL IN NONACADEMIC INSTITUTIONS The foreign post-MD is much more likely to seek an academic career. In part this is a reflection of his preference, compared to the American MD, for the basic medical sciences. But even in the clinical fields almost three fourths of the foreign postdoctorals indicate a university as their career location. The Postdoctoral in Nonacademic Institutions Percentage of Postdoctorals at Nonacademic and Academic Host Institutions 100 80- 60- o 10 40- I UJ ^ 20 - 2 - f.\t——— o £" z I NONACADEMIC INSTITUTIONS ACADEMIC INSTITUTIONS Although allusions have been made to that portion (19 percent) of the post- doctoral population not in U. S. academic institutions and occasional com- parisons between the two segments have been made, the nonacademic postdoc- toral activity deserves special consideration. Outside the universities, postdoc- torals can be found in nonprofit institutions, in industrial laboratories, in fed- eral government installations, and abroad. With the exception of industry, none of the above categories is homogeneous; each includes a variety of envi- ronments. Nonprofit institutions encompass hospitals, research institutes, pri-

116 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Nonacademic Host Institutions: Percentage of Postdoctoral s by Field of Postdoctoral, Sex and Citizenship, and Level of Appointment. 100- FIELD OF POSTDOCTORAL 75- l °-50- EMP Medical Sciences Biosciences Other Fields UI * 5 25 NONPROFIT INDUSTRY GOVERNMENT ABROAD LEVEL OF APPOINTMENT 100- Immediate PhD Intermediate PhD Senior PhD Long-Term PhD - Post-MD NONPROFIT INDUSTRY GOVERNMENT ABROAD

117 THE POSTDOCTORAL IN NONACADEMIC INSTITUTIONS SEX AND CITIZENSHIP 100- K O 75- U.S. Male U.S. Female Foreign, both sexes NONPROFIT INDUSTRY GOVERNMENT ABROAD Source NRC, Office of Scientific Personnel, Postdoctoral Census Questionnaire vate laboratories, libraries, museums, and state or local government offices. The federal government installations range from the quasi-academic laborato- ries, such as the Lawrence Radiation Laboratory at the University of Califor- nia and the Ames Laboratory at Iowa State University, through the National Bureau of Standards and the Los Alamos Scientific Laboratory, to the mission- oriented Fort Detrick Biological Laboratories of the Army and Houston Manned Spacecraft Center of NASA. Of major importance is the Bethesda campus of the National Institutes of Health. Postdoctoral activity abroad includes both appointments at foreign academic institutions and archeological field trips in uncharted territories. Other host institutions out of the country are libraries and museums. With such a variety of institutions, little can be said that applies to all of them. Figure 9 shows the differing patterns of fields, of levels of appointment, of citizenship, and of sex among the types of nonacademic host institutions. The lack of uniformity is the most obvious feature of these charts. There are, nevertheless, some important trends and each category of host institution dem- onstrates interesting characteristics. The behavior of U. S. male postdoctorals can be taken as a standard against which both the U. S. females and the foreigners can be measured. Each of the

118 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION latter groups must contend with special restrictions and attitudes that modify their postdoctoral opportunities. For the U. S. female, marital ties and linger- ing prejudice limit her freedom of movement. The foreign postdoctoral con- tends not only with language problems and scarcity of support in some fields, but also is differentially attracted to the United States as one moves from field to field. Especially for more senior scholars in the humanities and in the social sciences, only those concerned mainly with American studies would find the United States a particularly fertile research environment. Similar situ- ations, though sometimes more subtle, face the natural scientists. Although in some fields American science is preeminent, this is certainly not the case in all. The European Organization for Nuclear Research (CERN) or the Niels Bohr Institute in Copenhagen are certainly as attractive for physicists of what- ever country as their American counterparts. Academic institutions can also be used as a standard against which other host institutions may be compared. This is not to imply that the universities have the "proper" distribution of fields, sex, citizenship, or level of postdoc- toral activity, but rather that, as the largest category, they represent the choice that the bulk of the postdoctorals have made. The other categories of host institutions are important for the participants but are seldom statistically sig- nificant in the total postdoctoral picture. Nonprofit Institutions As indicated above, this category comprises several different kinds of institu- tions. In terms of numbers of postdoctorals, rather than numbers of institu- tions, the composition of the nonprofit group is 35 percent at hospitals, 14 percent at research foundations (usually medical), 40 percent at research insti- tutes and laboratories, and the remaining 11 percent at libraries, museums, and assorted agencies and nonprofit corporations. There are 817 postdoctorals in this group, of whom 50 percent are U. S. males, 7 percent are U. S. females, and 43 percent are foreign. By field, the proportions follow the general trends. The number of foreigners decreases as one moves from the BMP fields through biological and medical sciences to the other fields, and women are more likely to be found in the biological sciences and the other fields than in the BMP fields or the medical sciences. These patterns hold for all categories of host institutions. The medical sciences are more predominant in nonprofit institutions than in the universities, as are the humanities and social sciences. Of course, these fields are not represented at the same institution. The heterogeneity is caused by the variety of types of institutions subsumed under the category "non- profit." Nevertheless, some quasi-academic institutions do have several fields

119 THE POSTDOCTORAL IN NONACADEMIC INSTITUTIONS represented. Prominent among these are The Institute for Advanced Study on the East coast and Center for Advanced Study in the Behavioral Sciences on the West. Both are purely postdoctoral institutions offering no formal course work. Although informal seminars are regularly held, scholars work independ- ently except for the serendipitous collaboration that each institution attempts to foster through careful selection of its scholars. The invitational nature of the nonprofit institutions accounts for the rela- tively small proportion of immediate postdoctoral s and the larger numbers of intermediate and senior postdoctorals. Generally, the nonmedical institutions are concerned with research rather than training. Consequently, their limited resources are reserved for established or at least budding scholars who can be expected to be productive over the short period of the appointment. The immediate postdoctorals who are at some nonprofit institutions are there for the same reasons as those at universities, both from their own point of view and from the point of view of the institution. The president of a medi- cal research institute states, "Nonuniversity research institutions need the serv- ices of postdoctoral scientists to the same degree that university research pro- grams do." Over four-fifths of the post-PhD's at nonprofit institutions are again either returning to or seeking academic employment following their postdoctoral appointments, and even 43 percent of the post-MD's are headed for the uni- versity. The nonprofit institution (whether a research institute or a hospital) is, therefore, an alternative place to do research but it is not really different from the university as a place of postdoctoral study.39 It often has its own advantages for postdoctoral study, including special equipment or library col- lections and fewer distractions than a university. Industrial Laboratories We have been able to locate a total of 47 postdoctorals at three industrial labo- ratories. The three firms are Bell Telephone Laboratories, Avco-Everett Re- search Laboratory, and The Mitre Corporation. We know that other industrial laboratories have postdoctorals, but the number is small. Of 42 spokesmen for industry who responded to our inquiries, 17 indicated that they had formal or informal postdoctoral programs. It is characteristic, however, that even the largest corporations offer only a handful of such positions. Except for the Bell 39The director of a nonprofit laboratory engaged in research in the life sciences says: "I think that this laboratory behaves more like the appendage of a university than an ortho- dox nonprofit institution. ... All our research personnel have had university postdoctoral experience." (It is not clear from the evidence that the "orthodox" nonprofit institution exists.)

120 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Laboratories (where there are 36 postdoctorals out of the 47 who filled out our questionnaire), no firm mentioned any larger number than two or three. Most of the appointments are offered on an informal basis. One company, North American Aviation, announces the postdoctoral appointments available at its Science Center in the same fashion as a university. A major reason why these firms have postdoctorals is the competition with universities for doctoral talent. The argument is made by the vice-president of a research corporation in the following way: "To the extent that the young PhD is strongly attracted to the university environment for postdoctoral stud- ies, other organizations in need of PhD's must either find ways to bid compet- itively for their services or provide themselves by other means with equivalent learning and capability." Another vice-president says flatly: "With the advent of more industry-like research going on in universities, it becomes necessary for industry to become more university-like to attract research scientists." Most industrial firms admit that offering postdoctoral appointments is a useful recruiting device. Only a few speak of the need to educate young PhD's in their area of research or point to the stimulus that postdoctorals can give their firm's research programs. One respondent states as a matter of course that "one purpose" of the firm's postdoctoral program is "to attract interested and promising individuals to the laboratories, with the expectation that if we feel they are outstanding, they may become interested in our work and choose to remain with us." A company spokesman who mentions another purpose first quickly lists recruiting second: The prime motivation for establishing the postdoctoral program was the desire to increase in our laboratory the number of young, high-class research men above the number we could afford as permanent employees for the purpose of increasing the infusion of. new ideas, experiences and techniques into our research organization. In addition, we expect to hire a few of these people just as we hire postdoctorals from other establishments. Then the appointment is also a trial period for the laboratory and the man, which can be terminated by either party without prejudice. That such a motivation is reasonable is supported by the data in Table 11 (p. 62). Thirty-five percent of the immediate postdoctorals in industry will remain in industry. This is a larger percentage by far than that from any other source. Nevertheless, only a minor fraction of the nation's industrial firms offer postdoctoral programs. It is instructive to consider why the vast majority do not. For many firms the idea of offering short-term appointments raises serious difficulties. The research director of a major steel company argues: The very nature of industrial research including the possibility of involvement with pro- prietary matters, the dependence of fringe benefits on length of service, and other con- siderations militate against temporary opportunities being offered in industrial research. It is my feeling that such an arrangement would tend to encourage "floaters," employees

121 THE POSTDOCTORAL IN NONACADEMIC INSTITUTIONS who move at frequent intervals from one organization to another motivated solely by the possibility of a higher salary as a result of each move. The president of a consulting firm writes: I find it difficult from my own experience to make a case for offering postdoctoral edu- cational opportunities within very many industrial organizations which I have seen. The reasons for this are first, from management's point of view, I doubt that a cost effective- ness justification could be made for it; and second, from the student's point of view I doubt that he would find the climate and other motivational factors adequate. This is not to say that PhD's coming into industry do not have learning opportunities, but rather that the opportunities are too "real world" and, by definition, are therefore distracting and diverting. It seems to me that most PhD's interested in postdoctoral education are inter- ested in acquiring greater depth rather than greater breadth, and the last thing in the world they want is distraction and diversion. An oil company that has received many inquiries from young PhD's seeking postdoctoral experience has nevertheless felt compelled to turn them down: For reasons that appear obvious to us we are interested in hiring "permanent" employ- ees. An equally strong point is the great proprietary interest we seek to develop from our applied research, which represents about 90 percent of the total. A similar statement comes from the vice-president for research of a pharma- ceutical company: We have not attempted to offer postdoctoral opportunities in the sense that the candi- date would work for us for only one or a very limited number of years to enlarge his doc- toral experience, and then move on. Almost without exception we select our people with the intention that they will become "permanent" members of our research organization. . . . our laboratories operate on the open-door approach, with relatively free discussion of our objectives, and our successes and our failures. This community spirit flourishes best with employees who have made more than a temporary commitment to our organization. We shall return in Chapter 7 to the relationship between postdoctoral edu- cation and the industrial world. For the present we will content ourselves with commentary on the census data. Figure 9 shows that over half of the postdoctoral s in industrial laboratories are foreign. Although the numbers are small, these postdoctorals from abroad are almost entirely from developed countries, a pattern that is significantly different than at other types of host institutions. It is also evident that most of the foreign postdoctorals are not fresh PhD's; the contrast with the Ameri- can postdoctoral, who tends to be younger, is most acute in industry. The industrial postdoctoral is also likely to be in the physical sciences and engineering. The small fraction of life scientists probably reflects the proprie- tary nature of the health products industry (mainly pharmaceuticals), which is particularly adverse to the "temporary employee."

122 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION Federal Government Laboratories One way of characterizing the postdoctoral population in the federal labora- tories is to indicate the agency that supports the laboratory. If we do so, we find that 47 percent of the postdoctorals are supported by the National Insti- tutes of Health and virtually all of them are at the main campus of NIH in Bethesda, Maryland. Thirty-two percent of the federal postdoctorals are at one or another of the Atomic Energy Commission's laboratories such as Brookhaven, Los Alamos, Oak Ridge, Argonne, or the Lawrence Radiation Laboratory at the University of California. Eight percent are at installations of the National Aeronautics and Space Administration such as the Goddard Space Flight Center, the Houston Manned Spacecraft Center, or the Jet Pro- pulsion Laboratory at the California Institute of Technology. Five percent are at the several laboratories of the Department of Defense or of the three serv- ices. Among these laboratories are the U. S. Naval Research Laboratory, the Fort Detrick Biological Laboratory, and various laboratories of the Air Force Systems Command. The Department of Commerce supports almost 4 percent of the federal postdoctorals at its National Bureau of Standards, while the remaining 4 percent are distributed among installations of the Department of Agriculture, the National Science Foundation, and the Food and Drug Admin- istration. While many of the postdoctorals at NIH are similar to university project associates working on intramural research under the direction of the resident scientists, the majority are Public Health Service officers who are fulfilling selective service obligations. They are, so to speak, involuntary postdoctorals and might not properly be included in our census. The situation at the national laboratories of the AEC is strongly university oriented. Since the Manhattan Project, the government's activity in nuclear science has been dominated by academics, and the structure of the national laboratories reflects this heritage. With the exception of the Oak Ridge Na- tional Laboratory, each of the major installations is governed by either a sin- gle university or a corporation of a group of universities. The multibillion-volt accelerators are operated predominantly for university-based physicists and the flow of people back and forth is continuous. Perhaps for this reason the AEC laboratories are highly desirable locations for postdoctoral study and do not have the problem of other government and industrial laboratories in that appointments there impede a return to the academic world. The uniqueness of the facilities, the academic atmosphere of the activities, and the abundance of basic research in fields ranging from nuclear engineering to genetics more nearly duplicates the university than most nonacademic laboratories.40 40It has been suggested that the identification with universities be made closer by allow- ing the laboratories to grant graduate degrees. See Alvin Weinberg.s "The Federal Labo- ratories and Science Education" (Science, Vol. 136, April 6, 1962, p. 29).

123 THE POSTDOCTORAL IN NONACADEMIC INSTITUTIONS Postdoctoral appointments at the other federal centers are awarded for the most part by the Research Associateship Programs of the National Research Council. Since 1955, a number of federal laboratories have been hosts to post- doctorals selected by the NRC. The NRC, as well as the individual laboratories, advertises the availability of appointments at universities and elsewhere. Appli- cations from candidates are received by the Council and its selection panels prepare rank-ordered lists of candidates approved for awards.41 There are actually two separate programs under this rubric. In one, the NRC makes the awards and pays the stipends out of funds supplied by a con- tract from the participating laboratories. In the other, appointments are made under Civil Service regulation to as many candidates as the laboratory has funds for, without departing from the rank order as determined by the NRC panels. In the latter program each laboratory has had to receive prior approval from the Civil Service Commission to participate; however, since 1967 the Commission has permitted any laboratory to make one-year postdoctoral appointments through the NRC, if the NRC approves the laboratory's research program and environment. The Commission has also authorized extensions of appointments for a second year if the laboratory determines that the extension would benefit both the individual and the laboratory. The better-known laboratories, especially those engaged in basic research in fields of current interest, e.g., the National Bureau of Standards, have at- tracted increasing numbers of applicants of high caliber. Candidates are less attracted to laboratories where the emphasis is on applied research or develop- ment. Such laboratories appear to have several disadvantages: they publish less in the scientific journals, they are usually less well known, and candidates who might be attracted to them can get better-paid positions of the same sort in industry. The federal laboratories and the National Research Council recognize a double purpose in the associateship programs: to enlist the scientific resources of the laboratories in the development of talented individuals and to contri- bute to the research programs of the laboratories. Care is taken to keep these purposes in balance. If, over the years, for example, more than a third of the associates in a laboratory's postdoctoral program choose to continue with the laboratory as permanent employees, this is viewed as cause for concern. It is felt that a program is failing in its educational purpose if too many of its appointees close their career options in this way. Some ambivalence exists in the attitudes of the participating laboratories. There is a certain amount of reluctance on their part to releasing 100 percent of the exceptional talent they train. Table 11 (p. 62) indicates that almost a 41 In spite of possessing all the characteristics of fellows in the selection process, these "research associates" are subject to full taxation. As in the university the distinction between fellows and research associates is more a function of legal language than opera- tionally different treatment.

124 THE DEMOGRAPHY OF POSTDOCTORAL EDUCATION quarter of the federal postdoctoral s choose to remain in government employ following their appointment. However, graduates of the program who return to the university (and 55 percent do) often motivate their students to become employees and associates in the participating laboratory. Except in the physical sciences, the foreign postdoctoral plays a much less important role in federal laboratories than at other types of host institutions. Only 26 percent of the federal postdoctoral s are foreign. In part this is a reflec- tion of the dominance of the life sciences and the draft alternative that posi- tions in the Public Health Service represent. Obviously only Americans are concerned with the latter and Public Health Service officers are a large frac- tion (approximately half) of the federal postdoctorals. Postdoctorals Abroad Compared to the postdoctoral at an American university, the postdoctoral abroad is much more likely to be a mature scholar on leave for a year or less to make use of the unique resources overseas or to discover what is happening in foreign laboratories. In fact, as we have seen, the senior postdoctoral is as likely to be abroad as at home. The younger man is not as ready to leave the country, since his visibility for subsequent employment is less at a foreign establishment than at a domestic one. These behavior patterns are easily dis- cernible in the NSF postdoctoral programs, since the awardee may select his own fellowship institution. The fact that only 44 of the 120 regular postdoc- torals in 1968 chose to take their appointments abroad, while 42 of the 55 senior postdoctorals did so, illustrates the behavior. Some (10 percent) of the immediate postdoctorals abroad have already been appointed to the faculty of a university, but have delayed the actual beginning of the faculty appoint- ment to accept the fellowship. Not having to worry about their post-appoint- ment employment, they are free to leave the country. For comparison, only 2 percent of the immediates at U. S. universities are on leave from another posi- tion. Few object to the idea that the senior scholar should travel abroad, not only to represent United States science and learning abroad, but also to see his subject approached from another point of view and to become as familiar with foreign centers as the foreign scholars are with ours. Only the severest chauvinism assumes that the best in all fields is here and that nothing can be learned from others. The problem is whether the same values prevail for the immediate postdoctoral. Those in favor of postdoctoral opportunities abroad for the new PhD point out that for some fellows the foreign laboratory may be the best place to go because techniques and ideas there are more advanced than in the United States. Others, recognizing the indifference of science and

125 THE POSTDOCTORAL IN NONACADEMIC INSTITUTIONS scholarship generally to national boundaries, say that travel per se is not a justification but that the determining factor is where the postdoctoral can . receive the best research experience. If that laboratory is not in this country, so be it. Those opposed to the postdoctoral appointment abroad make their objec- tion on relative grounds. Granted that in some areas superior experience can be found in foreign centers, the question is whether the additional cost is justi- fied. If the man can receive almost as good an experience in this country, why not extend the funds by restricting the travel? Underlying these arguments is the suspicion that the move overseas will involve such a change of environ- ment that the research will not be efficiently pursued. There are problems involved in changing institutions in this country; for the American who goes, say, to Europe there are the additional difficulties of language and custom that must be mastered. Over 97 percent of the immediate postdoctorals abroad are supported on fellowships. The implication of this fact is that, on the whole, they are of higher quality than postdoctorals generally. They have been highly screened and are selected for their probable achievement of research leadership. On the basis of baccalaureate-to-PhD time lapse they are better than all other groups of postdoctorals. The average time lapse in the physical sciences for the im- mediate postdoctoral abroad is 5.0 years and for the basic medical sciences it is 5.9 years. Each is significantly below the time lapses given in Table 19 (p. 78) for the postdoctorals at U. S. institutions. A significant point is that we are not talking about very many people. Only 7 percent of the immediate U. S. male postdoctorals are overseas—a total of 145 people by our count. What might be inappropriate for the entire group of immediate postdoctorals could be valid for a highly select subgroup of them. The subtle influences that produce the creative researcher are not understood. It would seem prudent not to foreclose the foreign experience for a few in the name of economy, as the marginal cost probably does not begin to match the value of the work that one future Nobel prizewinner among them might accomplish.

Next: IMPLICATIONS FOR THE POSTDOCTORAL »
The Invisible University: Postdoctoral Education in the United States. Report of a Study Conducted Under the Auspices of the National Research Council. [Richard B. Curtis, Study Director] Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!