oP Mas, an i to yao a Seen rae at ional Science Board Nat 1973 ao Marine Biological Laboratory Library Woods Hole, Massachusetts Gift of Bostwick H. Ketchum - 1976 Science Indicators 1972 Report of the National Science Board 1973 mene National Science Board National Science Foundation For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 Price: $3.35, domestic postpaid, $3, GPO Bookstore. Stock Number 3800-00146 Letter of Transmittal January 31, 1973 My Dear Mr. President: I have the honor of transmitting to you, and through you to the Congress, the Fifth Annual Report of the National Science Board. The Report is submitted in accordance with Section 4(g) of the National Science Foundation Act as amended by Public Law 90-407. In this Report the National Science Board presents the first results froma newly initiated effort to develop indicators of the state of the science enterprise in the United States. The ultimate goal of this effort is a set of indices which would reveal the strengths and weaknesses of U.S. science and technology, in terms of the capacity and performance of the enterprise in contributing to national objectives. If such indicators can be developed over the coming years, they should assist in improving the allocation and management of resources for science and technology, and in guiding the Nation’s research and development along paths most rewarding for our society. Because of present limitations in data and methodology, the indicators in this Report deal principally with resources—funds, manpower, and equip- ment—for research and development and the areas to which the resources are directed. The Report presents relatively few measures of the outputs produced from these resources—the scientific advances and technological achievements, and their contributions to the progress and welfare of the Nation. The present paucity of such indices limits the conclusions which can be drawn concerning the quality and effectiveness of our scientific and technological effort. The Report represents only an initial step toward a system of science indicators. The further development of such indicators is a matter of high priority for future reports in this series. Respectfully yours, Weta H. E. Carter Chairman, National Science Board The Honorable The President of the United States ili 1% - - ema! re ‘we ene aut sii 24) *r ITs a eo eee Nw ene Contents INGER OD UIGTION a cokes reese sons aby ests soap ate bie Seetns ie Geena) ones saoceauter eels INTERNATIONAL POSITION OF U.S. SCIENCE AND TECHNOLOGY. cnuvoknnet oseGa cise Otae gece tee Suet Levelof Research'and Deéevelopmeént: j..0s54 55000550 eens eee« ton Results of Research and Development « .6.50..0..00sccewransssos Productivity, Technology Transfer, and Balance of Trade ........ RESOURCES FOR RESEARCH AND DEVELOPMENT 2440 ¢00%005505 National Resources for Research and Development .............. Federally Funded R&D and National Objectives ...............-- Resources-tor industrial R&D ti520ccc — o . «re ot Oo > 7 oe (AON) Thad ends "9 > Jr Aare due ee : ee eh meer a ry Introduction The National Science Board is charged by the Congress with providing an annual report of the state of science in the United States.! In its first four reports, the Board dealt with selected aspects of this subject, but with this, the fifth report, the Board begins the development of a system of indicators for describing the state of the entire scientific endeavor. These indicators, expanded and refined in the coming years, are in- tended to measure and monitor U.S. science—to identify strengths and weaknesses of the enter- prise and to chart its changing state. Such indicators, updated annually, should provide an early warning of events and trends which might reduce the capacity of science—and subsequently technology—to meet the needs of the Nation.’ The indicators should assist also in setting priorities for the enterprise, in allocat- ing resources for its functions, and in guiding it toward needed change and new opportunities. A system of science indicators which would fulfill these several purposes must include indices of both intrinsic and extrinsic aspects of the enterprise. Intrinsic measures would include the resources used for science; the condition of institutions involved in training, research, and technical innovation; quantity and quality of associated human resources; and advances in science. Extrinsic indices center around the application of scientific knowledge, and the tech- nology it fosters, to the achievement of national goals—in areas such as health, energy, environ- ment, defense, productivity, and foreign trade—and the consequent impacts on that elusive entity, the “quality of life.” Measures of these extrinsic aspects are particularly difficult to devise; the translation of science into tech- nology, the diverse applications of the two, and their myriad impacts are all intertwined with in- numerable economic and social variables. The realization of such a system of indi- cators—or even one which is less comprehen- sive—will be a difficult and long task, requiring o investigation of many potential indices, o expansion of the underlying data base, +1 Section 4(g) of the National Science Foundation Act as amended by Public Law 90-407. o improvement of methods for measuring the impacts of science and technology, Oo experience in interpreting the indices, and ao demonstration of their utility. In view of these problems and uncertainties, the effort to develop indicators is regarded as an experiment—a long-term experiment to determine if a useful system of indices can be devised in the years ahead. A central concept of the experiment is an evolving set of indicators, derived from the continuing exploration, refine- ment, and testing of prospective indices. The set will be expanded, refined, and updated annually, as new data become available and as the enter- prise itself changes. Throughout, the criterion of “usefulness” will be used to judge the value of individual indices, and to gauge the success of the experiment as it unfolds. Quantitative indicators, no matter how use- ful, are not a substitute for the experience and judgment of the scientific community. Indices, at their best, can only supplement this experience and judgment. Indeed, the interpretation of indi- cators—what they mean for the present and future health of the enterprise—requires the judgment of this community. The Report Indicators in this report deal with facets of the entire scientific endeavor, as well as certain aspects of technology. They range from meas- ures of basic research activity and industrial R&D, through indices of scientific and engi- neering personnel and institutional capabilities, to indicators of productivity and the U.S. balance of trade in high-technology products. Such a broad range of indices was included in the initial step of the experiment in order (a) to explore the scope of the effort involved in developing a relatively comprehensive system of indicators, (b) to identify gaps in the data base, and (c) to select specific areas for focused efforts in the future. While many potential indices were conceived for future development and a number of new concepts and even new data collections were Vii initiated, the actual indicators in the report were based largely on readily available data. As a result, the indices deal principally with re- sources—funds and personnel—for R&D, the disciplinary and functional areas to which the re- sources are directed, and the institutions which carry out the teaching and research functions. Relatively few output measures of either an in- trinsic or extrinsic nature are presented, because of the limited data available and methodological problems of separating the distinct contribu- tions of science and technology from those of other factors. Furthermore, the few such indi- cators which are presented (e.g., quantity and quality of scientific publications, patent output, and trade in technical knowledge) are subject to considerable uncertainty as to valid interpreta- tion and significance. These deficiencies limit the conclusions which can be drawn regarding the performance and contributions of the enter- prise. The first five chapters of the report present the initial set of indicators. The indices, wherever possible, are time series, usually ex- tending from the early 1960’s through 1972. Indicators are presented in graphical form and numbered so as to correspond with the numerical data tables in Appendix A. Preceding viii each of these chapters is an “indicator high- lights” section which briefly summarizes the major indices presented in the chapter. These sections, it should be noted, often omit important caveats and discussion contained in the full text. The last two chapters present results from opinion and attitude surveys of topics related to the state of science, which are not amenable to purely quantitative treatment. The first of these is a Delphi survey of the judgments and opinions of a wide cross section of the scientific and tech- nological community; the topics covered include the future role of science and technology in areas of high public concern, impacts of recent R&D funding changes, and basic research and criteria for allocating resources among scientific fields. The second is a survey of attitudes of the public toward science and technology; topics covered in the survey include the public regard for science and technology, their assessment of its impacts, and their desires for its Future use in coping with national problems. The task initiated with this report is ambitious; the present effort clearly represents only a beginning. The reports to follow in this series will aim to improve the concepts, refine the treatment, and expand the scope to include other facets of science and technology. International Position of U.S. Science and Technology International Position of U.S. Science and Technology This chapter compares the position and performance of science and technology in the United States with that of other major R&D-performing nations, through a variety of indicators. These include comparative indices of the level of R&D, in terms of expenditures for such activity and the number of scientists and engineers involved; results of R&D, as measured by the output of scientific reports and patents for new products and processes; and national performance in areas dependent upon science and technology, such as technical knowledge, productivity, and international trade. International comparisons are confined to general trends and to relative rather than absolute indicators, because of the paucity and limited quality of available information. This applies with particular force to comparisons involving the U.S.S.R. where definitions of R&D and scientific personnel often differ from those of other countries. Specific indicators must be interpreted with considerable caution. Indices of the level of R&D can be misleading because the costs of such activities, and differences in the productivity and functions of scientists and engineers, cannot yet be equated for the various countries. The output indices of scientific reports and patents reflect only a small part of the total output of science and technology, whereas the last group of indicators—those dealing with technical knowledge, productivity, and international trade—include the effects of many factors, science and technology being only one of them. INDICATOR HIGHLIGHTS a The proportion of the gross national product defense between 1961 and 1969, with such (GNP) spent for research and development (R&D) between 1963-71 declined in the United States, France, and the United King- dom but increased in the Union of Soviet Socialist Republics (U.S.S.R.), Japan, and West Germany. By 1971, U.S. expenditures for R&D were 2.6 percent of GNP, as compared with an estimated 3.0 percent for the U.S.S.R., approximately 2.0 percent for the United Kingdom and West Germany, and 1.8 percent for both Japan and France. The number of scientists and engineeers engaged in R&D per 10,000 population de- clined in the United States after 1969 but continued to increase in the U.S.S.R., Japan, West Germany, and France, with the result that by 1971 the number per 10,000 popula- tion for the U.S.S.R. was 37 as compared with 25 for the United States and Japan, 15 for West Germany, and 12 for France. All countries included in the comparisons significantly reduced the proportion of their government R&D expenditures for national expenditures in the United States dropping from 65 to 49 percent of total government R&D spending. Increases in the United States and most other countries occurred in the areas of space, community services, and economic development. In seven of eight scientific areas studied,! the United States produces a larger share of the world’s scientific and technical literature than any of the other major developed coun- tries; the U.S. share remained essentially un- changed between 1965-71. Literature produced by the United States is more frequently cited than that produced by other countries in all the scientific areas studied, with the exception of systematic biology and mathematics where the United Kingdom stands first. 1 The areas were physics and geophysics; chemistry and metallurgy; molecular biology; systematic biology; mathe- matics; engineering; psychology; and economics. ao The United States had a favorable but declin- ing “patent balance” (patents of United States versus foreign origin awarded in each country) between 1966 and 1970; the de- cline was due to a reduced number of patents of U.S. origin in France, West Germany, and the United Kingdom, combined with increased U.S. patents of Japanese origin. o Starting from a higher base, increases in labor productivity in U.S. manufacturing industries between 1960-71 were the lowest of all countries—39 percent—compared with 210 percent for Japan, 86 percent for West Germany, 81 percent for France, and 50 per- cent for the United Kingdom. Productivity gains in the United States offset increased labor costs until the mid-1960’s, but rose less rapidly than such costs during the 1966-71 period. o The United States had an _ increasingly favorable position in the sale of “technical know-how’—patents, techniques, formulas, franchises, and manufacturing rights—dur- ing 1960-71; Japan was the major purchaser @ The basis chosen for appraisal of U.S. science and technology in this chapter is an inter- national one, where the position of the United States is compared with that of other developed countries, particularly the major R&D-per- forming OECD (Organisation for Economic Cooperation and Development) member nations.? For reasons cited previously, compari- sons are usually restricted to measures of rela- tive magnitude and to general trends. LEVEL OF RESEARCH AND DEVELOPMENT The principal indicators of the level of a nation’s R&D effort are the expenditures and manpower devoted to such activities. The general objectives of such efforts, in turn, are suggested by the specific areas (e.g., defense, space, and economic development) to which these resources are directed. 2 The U.S.S.R. is included in the comparisons wherever the available data permit. Limited data and differences in definitions restrict the comparisons which can be made. of U.S. “know-how,” surpassing all of Western Europe after 1967. o The favorable U.S. balance of trade in tech- nology-intensive products grew throughout 1960-71, but was increasingly negative in nontechnology-intensive areas. ao Within the technology-intensive areas, prod- ucts with the fastest rising trade surplus are aircraft, computers, and plastics. Product areas in which the growth of imports ex- ceeds exports include office machinery, chemical elements and compounds, medicinal products, and telecommunication apparatus. ao The favorable trade balance of the United States in high technology products rested primarily on purchases by developing nations (55 percent in 1971) and countries of Western Europe. A deficit balance with Japan, developed in the mid-1960’s and con- tinuing to grow through 1971, exists in elec- trical machinery, scientific and professional instruments, and nonelectrical machinery. Expenditures for R&D Figure 1 presents expenditures for R&D as a percentage of gross national product for the six major R&D-performing countries included in the study. The United States, as well as France and the United Kingdom, show a declining ratio, as compared with increases for the U.S.S.R., Japan, and West Germany. Over the 1963-71 period (the only years for which data are avail- able or can be estimated for all countries), the United States exhibits the largest reductions, the U.S.S.R. and Japan the largest increases. R&D Manpower Another indicator of the level of a country’s R&D effort is the magnitude of the manpower (scientists and engineers) it devotes to these activities. This indicator should be regarded as only an approximation of the true level of R&D in that it fails to account for certain national dif- ferences, such as the designation and training of scientists, engineers, and technicians, and the quality of the research equipment available. 3 Figure 1 R&D Expenditures as a Percent of Gross National Product, by Country, 1963-71 (Percent) 3.2 3.0 ae — 2.6 24 2.2 20 a | | 2 ala 1963 ‘67 69 70 71 (a) 1964 SOURCE: Organisation for Economic Cooperation and Development; National Science Foundation estimates for 1970 and 1971; U.S.S.R. estimates by Robert W. Campbell, Univ. of Indiana. These differences also enter into determining the effective R&D level. The number of scientists and engineers engaged in R&D per 10,000 population in the United States, France, West Germany, Japan, and the U.S.S.R. is shown in figure 2. (Data for the United Kingdom are not available.) Taking the limitations noted above into consideration, these trends indicate that the U.S.S.R. sur- passed the United States after 1967 in the proportion of its population employed in R&D, 4 and that Japan had almost reached the same level as the United States by 1971. Moreover, of the five countries compared, only the United States had a declining ratio of scientists and engineers engaged in R&D. Government-Funded Research & Development Governments fund R&D in the pursuit of national objectives in various areas, such as national defense, economic development, and space. The distribution of funding among these objectives reflects national priorities whereas significant changes in the distribution often indi- cate shifts in national concerns. Government expenditures for R&D, however, represent only a part of the total national investment in R&D; expenditures by the private sector must also be taken into account for international compari- sons. The OECD has attempted to classify govern- ment expenditures for R&D into the following six areas: National Defense, encompassing military- oriented R&D as well as space and nuclear energy activities of a military character; Space Exploration, restricted to space R&D activi- ties of a civil nature; Nuclear Energy, restricted to nuclear energy R&D activities of a civil nature; Economic Development, including R&D in agricul- ture, fishing, and forestry, in mining and manufacturing industries, as well as service sectors such as public works, public trans- portation, communications, and construc- tion; Community Services, including R&D in health, the environment, public welfare (educa- tion, social services, planning, recreation, and culture), disaster prevention, law and order, meteorology, planning, and statistics; Advancement of Science, including government- funded research in universities, both separately budgeted research as well as research from the general funds of universi- ties provided by the government. Figure 2 Scientists and Engineers” Engaged in R&D per 10,000 Population, by Country, 1963-71 Number) eo? a eoonee W. Germany on” peoeee” ; ei a= eee Pirronensonesenen” aa rance PPT t eooe? eee? - o) * 5 OL ae 1963 < " n J (a) Includes all scientists and engineers (full-time-equivalent basis) (b) 1964 SOURCE: Organisation for Economic Cooperation and Development; National Science Foundation estimates for 1970 and 1971; U.S.S.R. estimates by Robert W. Campbell, Univ. of Indiana. The distribution of total government expendi- tures for R&D among these areas is shown in figure 3 for the United States, United Kingdom, France, West Germany, and Japan for the years 1961 and 1969. (Data for the U.S.S.R. are not available.) The chief changes in the United States between the 1961 and 1969 periods were the proportional reductions in R&D expenditures for national defense (down from 65 percent to 49 percent of total expenditures) and increases in the areas of space, community services, and economic development. Relative reductions in defense R&D also occurred in each of the other four countries, while the major increases, al- though differing from country to country, were in space, economic development, and_ the advancement of science. (Expenditures for the latter area are difficult to compare from country to country because of different government practices in the funding of university research; some governments provide support for such research through general grants to universities, whereas others—such as the United States—provide much of their support through mission-oriented agencies for specific research projects.) RESULTS OF RESEARCH AND DEVELOPMENT There are certain relatively direct results of R&D which provide indicators for comparing the scientific and technical performance of nations. Primary among these are reports of research published in scientific and technical journals, citations of reports from these jour- nals, and patents for new products and proc- esses. These provide measures of certain aspects of the output of the scientific-technological enterprise. Journal reports are produced primarily, but not exclusively, by the scientific portion of the enterprise as the result of research, both basic and applied, while new products and processes leading to patents are produced principally by industrial firms and inventors as the result of applied research and development. Scientific and Technical Publications Research reports published in scientific and technical journals are one of the more tangible outputs of the scientific community. Such reports reflect the results of specific research efforts. The results themselves may lead to further research, or be used many times over ina variety of practical applications. Furthermore, the critical review which usually precedes publication ensures that the reports have some degree of scientific or technical significance. Indicators based on research reports, however, have several limitations for the purpose of international comparisons: the quantity of such reports may be influenced substantially by the national customs regarding the publishing of re- search papers, by the availability of funds for 5 Distribution of Government R&D Expenditures Among National Objectives, by Country, 1961 and 1969 United States Percent France Nationa 1961 defense 1969 United Kingdom Fla T T T 1 Nuclear energy Advancement Total government R&D spending (Millions of Dollars) 1961 1969 US. $11,089 $15,847 France $ 601 $ 1,798 Japan $ 235 $ 839 Germany $ 423 $ 1,417 UK. — $ 1,078 $ 1,436 preparing and printing papers, by journal refereeing and publishing policies, etc. These factors provide good reason for caution in interpreting such indicators. The data presented here are the results of an initial attempt to: (1) Estimate the proportion of the world’s significant research and _ technical literature (in selected scientific areas) which is produced by the United States and other major countries, and (2) Assign a figure of merit representing the quality or significance of the litera- ture produced in each area by each country. The scientific areas and countries encom- passed are shown below: Scientific Areas Physics and Geophysics Chemistry and Metallurgy Molecular Biology Systematic Biology Psychology Economics Mathematics Engineering Countries France Japan United Kingdom United States WiSiSeR: West Germany National Origins of Literature. Estimates of the distribution of literature among fields and coun- tries were based upon counts of articles, letters, and notes published in some 500 journals covered by the Science Citation Index (published by the Institute for Scientific Information, Philadelphia, Pa.), supplemented by data from various abstracting services. The results are presented in figure 4. 3 For details of the methodology employed, including validation checks, see Development of U.S. and International Indicators of the Quantity and Quality of Scientific Literature, Computer Horizons, Inc., September 1972. The United States has a larger share of the literature in each of the areas (except for chemis- try and metallurgy) than any other country. The U.S. share, as well as those of other countries, shows little change over the short period (1965- 71) covered by the data. In the area of physics and geophysics, the United States produces some 40 percent of the literature, as compared with about 15 percent from the U.S.S.R., and between 5 to 8 percent from each of the other four countries. How- ever, in the other physical science area—chemistry and metallurgy—the U.S.S.R. has the largest share with some 29 percent, versus the 24-percent share of the United States, and the 5- to 9-percent share of each of the other four countries. In molecular biology the United States produces almost one-half the world’s literature, with each of the two next largest producers (United Kingdom and France) having shares of only 9 percent. The U.S. share of the systematic biology literature is considerably less, at some 30 percent, than molecular biology, but still ex- ceeds that of any other individual country by a wide margin. In psychology and economics, literature of U.S. origin represents the largest share. The available data for both areas, however, were more limited and less reliable than data for other areas. In spite of these limitations, the relative position of the United States in each area is believed to be accurately reflected by the data. In mathematics the United States andU.S.S.R. share the lead as the major literature producers, with the United States having a slightly larger share. The shares of the other four countries range from 4 to 7 percent. In engineering the United States produces about 50 percent of the world’s literature. The next largest producers are the U.S.S.R. with 12 percent, and the United Kingdom with about 10 percent. Literature Citations. In addition to the purely quantitative aspects, an effort was made to esti- mate the relative “quality” or “significance” of the literature. The basis for this indicator was the number of citations to the literature 7 Figure 4 Scientific Literature in Selected Fields Produced by Major Developed Nations, 1965-71 (Percent of Total) Physics & Geophysics 50 40 | 30 20 f- All Other Countries U.S.S.R 10 f— jo ee eee ee ee |) Japan PPPPPPTTerr rt itt teehee CO de Germany France 0 Ik | Molecular Biology 30 -- 20 |- 10 = — a an | re e Peeeee Cc _ W. Germany teeeseee peecceeees 8 Japan = SS eg = ere, USSR 0 | ssc esis | 1965 ‘67 ‘69 a SOURCE: Computer Horizons, Inc. All Other Countries (Percent of Total) Chemistry & Metallurgy 50 40 }— aly U.S.S.R U.S 20 | xaos All Other Countries 10 -— UK a W. Germany pcos tecceeouweeees eewenn.,”.* ee. France jewwonnnn : Japan 0 | | | Systematic Biology 50 40 |— te All Other Countries eg ie 30 f— 208|— 10 [— =, = UK os eR zee: France Japan 0 | 1965 67 69 71 (Percent of Total) Mathematics 50 oT 40 }— 30 U.S All Other Countries 10 Japan = 88 OLE a * nes 3p | W. Germany beeen Sy meng eee sees s=e France U.K a li | Psychology 80 US 70 30 . osee All Other Countries ae REE SLE ssaoF ee ae ae, os UK W. Germany Japan France 1965 67 69 71 (a) Data not available for U.S.S.R. SOURCE: Computer Horizons, Inc. 507-194 O - 73 - 2 (Percent of Total) 50 40 30 20 10 0 50 40 30 Engineering All Other Countries oe ee U.S.S.R — = @UK W. Germany . Japan bessesesessssscssccscccccco sete een eneee France | if sie | | Economics All Other Countries Prey ecccccccccccsceseeree® ee, ° . ce . e °° France fl “deececeescoee, Japan 1965 67 ‘69 ‘71 produced by each country in each scientific area. The general rationale for such an index is the expectation that the most significant literature will be most frequently cited, whereas relatively unimportant research articles will attract few, if any, citations. Support for the validity of this indicator is the high correlation, found in a number of studies, between the significance of papers as judged by researchers in the field and as measured by the number of citations. How- ever, articles may fail to be noticed because of the language or journal in which they are published, whereas others may be heavily cited because of the criticisms they provoked or because they describe a minor improvement in methodology. These and similar limitations of the indicator, however, are minimized by the extremely large number of citations invelved in the present case. The data source for the “significance” indi- cator was the Science Citation Index, which includes all publications cited in the 500 journals used in this study. The total number of citations received by the literature from each country in each area was divided by the number of research articles produced by the particular country in that scientific area. The resulting “cita- tion/publication” ratios were determined for the years 1965, 1967, 1969, and 1971. Since the ratios did not change significantly over the period, the mean annual value was used. These data are presented in figure 5. Overall, the U.S.-produced literature had the highest “significance” ratio in five of the seven fields, with systematic biology and mathematics the two exceptions. (“Economics” was omitted because of the lack of reliable citation data.) Literature of the United Kingdom received the next highest rankings, placing either first or second in each field. Ranking after the United States and the United Kingdom were West Germany, Japan, U.S.S.R., and France, in that order.4 The “Patent Balance” Data on patent applications and awards are also measures of inventive output. Inventions of 4 The particular sample of journals used in estimating the national origins of literature and the relative significance of the literature may have resulted in some bias favoring English-language publications. 10 new processes and products, of sufficient originality to be patented, represent potential technological advances. Patents, however, vary greatly in their technical and economic importance and the basis for their award differs in important ways from country tocountry. Not only does the rigor of the tests for originality vary, but there are also considerable differences in the relative success of litigation involving patent rights; this determines the relative ease and value of obtaining patents in different coun- tries. The absolute number of patent applications or awards in individual countries is not an ade- quate indicator for the purposes of inter- national comparisons. It is more meaningful to compare the number of patents awarded to nationals with those awarded to foreigners in each country.5 This yields an index which reflects the relative success of countries in developing products and processes of sufficient potential significance to warrant international patent protection. Figure 6 presents the total number of patents awarded to U.S. nationals by five countries (United Kingdom, U.S.S.R., West Germany, Japan, and France), those awarded by the United States to nationals of these countries, and the resulting U.S. balance. This shows that the “patent balance” of the United States fell by some 40 percent between 1966-70. The decline is due principally to the reduced number of patents awarded to U.S. nationals by foreign countries. Since patent applications in a foreign country are usually the result of simultaneous patent application in the applicant’s own country, these data indicate that the rate of growth of patent- able ideas of international merit have been expanding at a greater rate in other countries than in the United States. The patent balance of the United States rela- tive to each of the other countries (except the U.S.S.R. which accounts for less than one per- cent of the total patent transactions considered) is presented in figure 7. Overall, the U.S. balance 5 Patents awarded to U.S. nationals by foreign countries minus patents awarded to foreign nationals by the United States. Figure 5 Citation/ Publication Ratio of Scientific Literature in Selected Fields [Mean Values for 1965-71] Physics & Geophysics Ratio 5 1.0 Hy So isa) U.S. UK. Other W. Germany USSR. Japan France Molecular Biology U.S. UK. Other Japan W. Germany US.S.R. France Mathematics UK. US. W. Germany Japan USSR. Other France Psychology ‘” US. UK. Other Chemistry & Metallurgy Ratio 5 1.0 1.5 0 US. UK. W. Germany Other Japan France U.S.S.R. Systematic Biology” UK, W. Germany US. Japan France Other Engineering UIS: UK. Japan Other W. Germany France US.S.R. (a) Data not available for the U.S.S.R. (b) Data not available for France, W.Germany, Japan and the US.S.R. SOURCE: Computer Horizons, Inc. Figure 6 Patents Awarded to U.S. Nationals by Foreign Countries and to Foreign Nationals by the U.S. 1966-70 (Thousands) 35 30 25 10 0 1966 ‘67 68 69 70 SOURCE: World Intellectual Property Organization, Geneva declined in each country, with the major reduc- tion occurring in respect to France, followed by smaller declines in relation to the United King- dom, Japan, and West Germany in that order. Declines in the U.S. balance vis-a-vis France, United Kingdom, and West Germany were due principally to reductions in the number of patents awarded to U.S. nationals; conversely, the decline with respect to Japan was produced largely by an increasing number of awards to that country by the United States. Thus, the U.S. patent position, while still favorable in 1970, was eroding for two reasons: (a) a declining number of foreign patents of U.S. origin and (b) an increasing number of U.S. patents of foreign origin. 12 Figure 7 U.S. Patent Balance with Selected Countries, 1966-70 (Thousands) 12 = i Pe = Se 10 =e UK == —— ees evcccceecossoscs gene *. ee” ° o? e Cha . st *e *, ° ° * France “. 6 *. . Meee. 4 bree. -on Japan .o%e., aL *,°° Peco, = W. Germany 0 a= = ee Pm es ey SS oC he FS x x x ~ x —2 1966 67 68 '69 ‘70 SOURCE: World Intellectual Property Organization, Geneva. PRODUCTIVITY, TECHNOLOGY TRANSFER, AND BALANCE OF TRADE Science and technology play an important role in industrial innovation, productivity, economic growth, and trade among nations. They repre- sent, however, only one component in the com- plex matrix of factors which determine the tech- nological and economic position of a nation. The numerous other variables involved (legal, finan- cial, social, market, etc.) make it difficult to specify the exact role and contribution of any single factor—such as R&D. In discussing productivity, technology trans- fer, and balance of trade, comparisons are limited to those areas and aspects which depend upon R&D in rather direct and obvious ways. Productivity Productivity expresses the relationship between the quantity of goods and services produced (the output) and the resources (e.g., labor, capital, land, and energy) used to produce them (the input). One of the most commonly used indices of productivity is “output per man- hour,” which relates output to the input of labor time. R&D contributes to productivity by providing advances in technology which increase output per man-hour. All studies of the effects of R&D on productivity growth conclude that Figure 8 Productivity in Manufacturing Industries, by Country, 1960-71 [Index 1960 = 100] (Index) 350 300 = Japan : on 250 BS : : | : ° : : ge ie; : 200 L rs : ? W ee one . . costa Pes ee 150 Aer ie : 76° UK 4 1960 ‘61 ‘62 ‘63 ‘64 65 ‘66 ‘67 ‘68 ‘69 ‘70 ’71 SOURCE: U.S. Department of Labor. there is a direct relationship which is “positive, significant, and high.”e Indices now available do not permit compari- son of absolute levels of productivity in dif- ferent countries, except in the case of certain individual industries. Instead, comparisons are limited to changes in productivity which occur over time in individual countries. Normalized data for changes in relative labor productivity in manufacturing are presented in figure 8 for the United States, United Kingdom, France, West Germany, and Japan. © Research and Development and Economic Growth/Productivity, Papers and Proceedings of a Colloquium, National Science Foundation, NSF 72-303, December 1971 Figure 9 Unit Labor Cost in Manufacturing Industries, by Country, 1960-71 [Index 1960 = 100] (Index) 160 al 150 }- U.K i: 140 L ae: ae: ean pe France wey : 30 L s iy 2 7 OF Japan ee, Kp ot . 2 e . eae v, W. Germany 120 EF 110 = 100 —90 1960 ‘61 ‘62 ‘63 ‘64 ‘65 ‘66 ‘67 ‘68 '69 ‘70 ‘Jl SOURCE: U.S. Department of Labor. Japan increased its productivity by 210 per- cent between 1960-71, compared to about a 39- percent increase in the United States. Produc- tivity gains in manufacturing over the same period were 86 percent for West Germany, 81 percent for France, and 50 percent for the United Kingdom. Of course, the United States, which has already reached a high level of productivity, might not be expected to sustain the same high proportional gains as countries starting from a lower productivity base. Increases in productivity can translate into lower product cost, providing the productivity gains offset increases in labor costs. Unit labor cost (hourly labor cost divided by output per man-hour) for manufactured products is shown in figure 9 for the same five countries. It can be seen that productivity gains were sufficient to offset increased labor costs in the United States during the 1960-66 period, but not thereafter. Productivity gains in the other four countries were also exceeded by increases in labor costs, particularly after the mid-sixties. Even in the case of Japan, the large increases in productivity after the middle of the decade were not suffi- cient to offset the growth in labor costs.” Technological Balance of Payments Nations often seek to improve their tech- nological and productive position by purchasing technical “know-how” (e.g., patents, tech- niques, formulas, designs, franchises, and manu- facturing rights) from other countries. Many factors—such as R&D and economic develop- ment policies, ownership and trading arrange- ments, and marketing effectiveness—may in- fluence the actual level and balance of tech- nology transfer between nations. A persisting favorable balance of payments for technical “know-how,” however, is an indicator of a strong position in technology. Data on payments and receipts for technical “know-how” are available for transactions between multinational companies and their foreign affiliates as well as between independ- ent organizations. The latter data were selected for use here primarily because of the greater ele- ment of valuation implicit in transactions between independent enterprises. It should be 7 A detailed and comprehensive study of unit labor cost, productivity, and labor compensation is presented in Competitiveness of U.S. Industries, United States Tariff Commis- sion, TC Publication 473, Washington, D.C., April 1972. 14 noted, however, that the omission of trans- actions between corporations and their foreign affiliates results in a substantial understate- ment of the extent of technology transfer. In addition to transactions involving financial ex- changes, a considerable amount of technical “know-how” is transferred informally between multinational firms and their foreign affiliates. The technological balance of payments with respect to the United States is presented in figure 10. Included are U.S. receipts from the sale of technical “know-how” to all countries, payments by the United States to all other coun- tries for such “know-how,” and the resulting balance of payments (receipts minus payments). The United States had a strong and increasingly positive balance of payments in this area throughout the decade. U.S. receipts from the sale of “know-how” grew exponentially while its payments—which were some four to five times less than receipts—rose linearly. The U.S. balance of payments associated with a country or countries is presented in figure 11. This shows that the principal increase in the U.S. balance is attributable to Japan which more than tripled its purchase of technical “know-how” from the United States between 1966-71. Purchases by the United States from Japan, on the other hand, remained at a very low level (only 2 to 4 percent as much as Japan’s pur- chases from the United States) throughout the period. Western Europe (mainly the Common Market countries) also contributed to the favor- able U.S. balance; however, purchases by the United States from Western Europe, which were much greater than those from Japan, were some 40 percent as large as U.S. receipts from these countries for technical “know-how.” Balance of Trade in Technology-Intensive Products A nation’s balance of trade depends upon many factors including the relative price of its products, the effectiveness of its marketing, its trading arrangements with other countries, and its relative position in industrial technology. The technology position, in turn, is increasingly dependent upon the use of R&D to improve and develop new industrial products, processes, and services. This relationship between industrial tech- nology and R&D provides a basis for dividing — Figure 10 U.S. Payments and Receipts for Patents, Manufacturing Rights, Licenses, Etc., 1960-71 (Millions of Dollars) 700 600 L 500 U.S. Receipts Balance U.S. Payments 1960 62 64 66 68 70 SOURCE: U.S. Department of Commerce Figure 11 U.S. Balance of Payments for Patents, Manufacturing Rights, Licenses, Etc., by Country, 1960-71 (Millions of Dollars) 500 gas U.S.—Total 400 [ 300 200 100 es U.S.—W. e= 1960 62 ‘64 66 '68 LO cai (a) Except U.K SOURCE: U.S. Department of Commerce manufactured products into two groups: tech- nology-intensive and nontechnology-intensive products. The grouping is based upon the rela- tive extent of R&D performed by the industries which manufacture the products. Products from industries with (a) 25 or more scientists and engineers engaged in R&D per 1,000 employees and (b) 4 percent or more of their net sales directed to R&D are designated here as “tech- nology-intensive” products while those with a lower level of R&D investment are regarded as “nontechnology-intensive” products. Based on these criteria, the product areas which are tech- nology-intensive are (a) chemicals, (b) non- electrical machinery, (c) electrical machinery, (d) aircraft and parts, and (e) scientific and profes- sional instruments. All other manufactured products are regarded as nontechnology inten- sive.8 The US. trade balance (exports minus imports) associated with these two categories of products is shown in figure 12. The favorable balance in technology-intensive industries is clearly indicated; the balance doubled over the 1960-71 period. In contrast, the United States had a large and increasing deficit trade balance in nontechnology-intensive products.° The favorable U.S. export position in high technology products results in large measure from the extensive use of scientists and engi- neers as well as other skilled manpower.!° Thus, the U.S. comparative advantage for inter- national trade in manufactured products appears to be in the export of goods that are intensive in the use of highly educated and skilled personnel rather than in low technology mass-produced goods. The favorable U.S. trade balance in the tech- 8 This grouping, of course, is an approximate one. Products and industries, although highly correlated at the gross level, do not perfectly coincide, with the result that not all products manufactured by a high R&D-performing industry are technology-intensive. Further, the criteria used here result in placing “automobile and parts” products in the nontechnology-intensive group, instead of the other group where they are sometimes assigned. 9° The US. trade position may also be viewed in terms of capital and consumer goods. The favorable U.S. export performance is largely accounted for by capital goods, almost all of which are produced by technology-intensive indus- tries, while the unfavorable import situation lies largely in consumer goods. 10 W. H. Bronson and H. B. Junz, “Trends in U.S. Trade and Comparative Advantage,” Brookings Papers on Economic Activity, 2, 1971. r 15 Figure 12 U.S. Trade Balance in Technology- and Nontechnology-Intensive Manufactured Products, 1960-71 (Billions of Dollar 1960 ‘61 SOURCE: U.S. Department nology-intensive area is shown in figure 13 for ea ch of the five industry groups. @ Nonelectrical machinery declined int net exports during 1971. The decline was pri- marily due to the continuing rise in imports of office and textile machinery. Computers and construction and mining equipment account principally for the net export posi- tion. e Aircraft and parts had a 9:1 export/import ratio in 1971. Although this has been the fastest growing exporting industry, there are signs that the favorable ratio may de- cline in the near future. e Chemical net exports declined: in 1971 chiefly due to increased imports of chemical elements and compounds as well as medici- nal and pharmaceutical products; major net export products were plastics and resins and radioactive materials. e Instruments maintained a steady but small growth in net exports. Figure 13 U.S. Trade Balance in Technology- Intensive Products, 1966-71 By Product (Billions of Dollars) 6 Cod “Us ¢? 5 ¢ ? Nonelectrical machinery got & ae 4 3 }-— Aircraft rd oe” Chemicals 1966 ‘67 68 69 70 71 With Other Nations (Billions of Dollars) 6 Developing t 5 na On on aco” ad Pg 4 e PT Western ~ 3 ? ex Europe # 0 ee —j| —____— 1966 67 68 169s ‘70 yal SOURCE: U.S. Department of Commerce. e Electrical machinery continued its rapid de- cline in net exports, principally as a result of imports of telecommunication apparatus. This mixture of growing and declining exports illustrates the complexities of the present U.S. trade position. The underlying dynamics of the position, however, are partially explained by the “product cycle” concept.!! Trade in manu- factured goods, according to this concept, typi- cally follows a cycle in which the United States initially establishes a net export position with the introduction of a new product, maintains this position until the technologies and skills necessary for manufacturing the product are developed elsewhere, and then becomes an im- porter as the production is standardized and moves abroad to minimize costs. The concept implies that the product structure of U.S. exports must have a continuous infusion of new products in order for the United States to main- tain a favorable trade position. The favorable position of the United States in high technology areas is based primarily on exports to developing nations and countries of Western Europe (figure 13). In 1971, developing nations accounted for 55 percent of the net exports in these areas, and Western Europe almost 30 percent. In contrast, the deficit balance in the high technology area developed with Japan in the mid-1960’s and persisted in the following years, with the largest increase (almost 120 per- cent) occurring in 1971. This deficit lies pri- marily in electrical machinery (particularly in consumer electronics), and to a lesser extent in the instrument and nonelectric machinery areas. Only in the aircraft and parts area does the United States have a significant net export posi- tion with respect to Japan. Although the United States still retains a 11R. Vernon, “International Investment and_ Inter- national Trade in the Product Cycle,” Quarterly Journal of Economics, v. 80, May 1966. strong position as a net exporter of technology- intensive products, various indicators suggest that the position may deteriorate in the near future. Not only did the overall trade balance for high technology industries level off between 1970-71, but the two industries most responsi- ble for the favorable balance in previous years (nonelectrical machinery and chemicals) had their first decline in net exports in 1971. Furthermore, net exports of electrical machinery appear to be declining at a faster rate than in recent years; the shift of exports in this area from developed to developing nations is re- garded as a further indication that these prod- ucts are moving from the export to the import stage. In addition to these relatively direct indi- cators of the U.S. trade position, other indices suggest that R&D trends in these industries may contribute to a deteriorating position in the future: the “R&D intensiveness” of the five industries as a whole declined by some 25 per- cent in recent years and expenditures for R&D fell by almost 10 percent between 1968-70, as noted elsewhere in this report. The preceding examination of foreign trade was restricted, for the purposes of this report, to those aspects which provide relatively direct indices of the position and performance of U.S. technology. As a result, the whole area of foreign direct investment and sales of U.S. sub- sidiaries abroad was neglected.!2 Such sales, mostly in technology-intensive products, exceed exports by some 2.5 times, but are highly cor- related with the export position of the individ- ual industries. The large and growing invest- ment income from foreign subsidiaries of the United States helps to finance U.S. imports of nontechnology-intensive products. Further- more, such investment in developing nations probably generates a considerable market for the export of U.S. technology-intensive products. 12 A comprehensive discussion of these and related topics is presented in: P. G. Peterson, The United States in the Changing World Economy, U.S. Government Printing Office, 1971. 17 oar =m $ +i) SS al ew oS 1; (ope +6 a. tee | ids 0 eae es it "Ai 7 are ~¢ . _ “oe ia 4 ee : _ Resources for Research and Development Resources for Research and Development Indicators in this chapter deal with the financial and human resources employed in research and development. These include measures of the total national R&D effort, in terms of the level and sources of funding; the character of R&D (basic research, applied research, and development); and the scientists and engineers engaged in these activities. The general areas of R&D and the institutions involved are indicated, although these aspects are more fully treated in subsequent sections of the report. In this chapter resources are viewed both as inputs to the scientific-technological enterprise and as indicators of the level of its R&D. The use of financial resources as a surrogate for level of R&D activity requires that the purchase value of the dollar be adjusted to reflect rising costs. In the absence of an “R&D price index,” the implicit price deflator for the gross national product is used to convert R&D expenditures from current to constant dollars; this conversion, it is recognized, may not fully account for the increasing costs associated with R&D. INDICATOR HIGHLIGHTS o National expenditures for R&D increased Oo throughout the 1961-72 period when expressed in current dollars; in terms of con- stant 1958 dollars, however, expenditures declined 6 percent between 1968 and 1971, but increased slightly in 1972 to a level equivalent to that of 1966-67. Total R&D expenditures as a proportion of Most affected by the funding reductions were development activities which leveled off in 1970 before rising again in 1971 and 1972; in constant 1958 dollars, however, expenditures for development declined after 1969 and remained at the lower level through 1972. The fraction of total Federal outlays devoted Qo the. gross national product declined to 2.5 to R&D fell from 12 to 7 percent between percent in 1972 from a high of 3.0 in 1964; 1965-72. The decline was due in large part to the decline was due to continued growth of the growth of Federal expenditures in areas the GNP coupled with the reduced growth of which have small R&D outlays, such as Federal R&D expenditures. income security, and to reductions in space R&D. o Federal Government expenditures for R&D : in current dollars leveled off after 1968 and Some 73 percent of all Federal R&D expendi- declined slightly thereafter—primarily as the tures in 1972 went for national defense and result of reduced expenditures for space space exploration. National defense received R&D—before rising in 1971 and 1972; the 54 percent of total Federal R&D funds in result in constant 1958 dollars was a reduc- 1972 and space exploration received some 19 tion which continued through 1971 and percent of the total. amounted to a 12-percent decline. Federal expenditures for R&D in civilian o The number of scientists and engineers areas (areas other than national defense and 20 engaged in R&D reached almost 560,000 in 1969 before declining each year thereafter for a total reduction of some 35,000 by 1972; almost all the decline occurred in the indus- trial sector. space) increased throughout the 1963-72 period, rising to 27 percent of the total in 1972, up from 14 percent in 1963. Areas re- ceiving the bulk of funds in 1972 were health (8.7 percent), advancement of science and technology (4.4 percent), transportation (3.8 percent), environment (3.2 percent), and energy conversion and development (2.5 per- cent). o Total expenditures for industrial R&D in current dollars increased until 1969, de- clined in 1970, and then rose in 1971 and 1972; the trend in constant 1958 dollars, however, was one of considerable declines after 1969 and a small increase in 1972, leaving expenditures at their 1965-66 level. ao Industry-funded R&D, which rose in-cur- rent dollar expenditures throughout the 1961-72 period, is devoted to applied re- search and development in the electrical equipment, aircraft and missiles, motor vehicles, chemical, and machinery indus- tries. Some 80 percent of Federal expendi- @ The Nation devotes a sizable share of its human, institutional, and financial resources to research and development. The largest propor- tion of these resources is directed toward the achievement of national objectives in areas such as defense, health, space, energy, and the environment. A somewhat smaller share of the resources goes for developing the technological basis for new and improved industrial products and services. And a considerably smaller share is used for improving the fundamental under- standing of man and nature. NATIONAL RESOURCES FOR RESEARCH AND DEVELOPMENT Total U.S. expenditures for R&D are shown in figure 14 for 1961-72 in both current and con- stant (1958) dollars. Current dollar expendi- tures rose throughout the period. As measured in constant dollars, however, expenditures peaked in 1968, and declined by 6 percent over the next 3 years to a level equivalent to 1966. Slightly increased expenditures are estimated for 1972. (Coincident with the constant dollar declines were nearly equivalent proportional reductions in the total number of scientists and engineers engaged in R&D, as shown in figure 15s) tures for industrial R&D went to the first two industries in 1970. a Federal funds for industrial R&D leveled of f in the mid-1960’s and declined in current dollars after 1969—primarily because of re- duced expenditures for space R&D—while industry continued to increase its expendi- tures, with the result that in 1968 industry replaced the Federal Government as the prime source of support for industrial R&D. o Universities and colleges, which provided 4 percent of the Nation’s R&D funds in 1972, concentrate their expenditures on basic and applied research in the life sciences (almost 50 percent), physical sciences and engi- neering (20 percent), and the social sciences (16 percent). As a part of its gross national product, the United States is estimated to have devoted 2.5 percent of GNP to R&D in 1972. This ratio, which reached its highest level of 3.0 percent in 1964, has declined steadily since 1967 (figure 14). The reduction is attributable to the con- tinued growth of the GNP coupled with de- clines in R&D expenditures by the Federal Government; non-Federal expenditures for R&D remained at approximately 1.2 percent of GNP between 1967-72. The principal sources of R&D funds are the Federal Government which provided 55 percent of the nation’s total R&D expenditures in 1972, private industry 40 percent, and the universi- ties and colleges 4 percent. Other nonprofit institutions contributed the remaining 1 per- cent (figure 14). Government funding in cur- rent dollars declined slightly between 1968 and 1970 and increased in 1971 and 1972; in con- stant dollars, however, Federal funding de- clined by 12 percent between 1968-71 before in- creasing slightly in 1972. Federal expenditures for R&D in FY 1973 are estimated at approxi- mately $16.5 billion, a 3-percent increase over expenditures for FY 1972, which were in turn 6 percent higher than FY 1971. Expenditures for R&D in terms of the charac- ter of work—basic research, applied research, 21 N N Figure 14 National R&D Expenditures, 1961-72 Total (Billions of Dollars) 30 Current dollars Constant 1958 dollars 10 5 — 0 [pests as [eae ef ea ee 1961 63 65 67 '69 Ti 5 (a) GNP price deflator was used to convert current to constant dollars By Character of Work (Billions of Dollars) 20 Current dollar Basic research se ee es 1961 '63 65 ‘67 ‘69 gh eet (2 (a) GNP price deflator was used to convert current to constant dollars. SOURCE: National Science Foundation As a Percent of GNP Percent) 3.20 f 3.00 & uz Total 1961 '63 ‘65 ‘67 69 Fle 2: (est.) By Source (Billions of Dollars) 16 Current dollars Federal Government 12 el IL & Constant 1958 dollars ~ Industry “a Universities & Colleges Ss 0 sie dele ee 1961 63 65 67 69 ATL (est.) (a) GNP price deflator was used to convert current to constant dollars NOTE: Other nonprofit institutions R&D expenditures increased from $110 million in 1961 to $235 million in 1972 Figure 15 Scientists and Engineers” Employed in R&D, by Sector, 1961-72 (Thousands) 600 550 500 450 400 350 300 250 200 150 100 50 0 1961 '63 65 ‘67 69 7A 72 (est.) (a) Includes all scientists and engineers (full-time-equivalent basis) SOURCE: National Science Foundation. and development—are shown in figure 14. The most salient change appears in the development area in which constant-dollar expenditures de- clined during 1969-71. This decline accounts in large part for the overall R&D reduction noted in figure 14. In 1972, development activities accounted for 64 percent of total R&D expendi- tures, applied research 22 percent, and basic research 14 percent. A part of the nation’s human resources devoted to R&D are the scientists and engineers who are engaged in performing research and development. Their total number reached almost 560,000 in 1969 before declining in each subsequent year for a total reduction of some 35,000 by 1972. Almost all the decline occurred in the industrial sector (figure 15). Industry had two-thirds of the nation’s total scientists and engineers engaged in R&D (on a full-time- equivalent basis) in 1972 as compared with the universities and the Federal Government, each of which had some 13 percent. FEDERALLY FUNDED R&D AND NATIONAL OBJECTIVES R&D resources and activities can be related to the national functions they serve, such as defense, space, natural resources, commerce and transportation, and health. Federal expendi- tures for R&D! in these functional areas reflect the extent to which R&D is used by the government in the pursuit of national goals. Total Federal Outlays and R&D Expenditures Federal expenditures for R&D, expressed as a percentage of total Federal outlays, declined appreciably after 1965, as shown in figure 1o. The reduction resulted from a mixture of (a) rapid growth of Federal outlays in areas which have small R&D expenditures, (b) diminished expenditures for space R&D, and (c) relative de- cline in expenditures for national defense, as a proportion of total Federal outlays. Federal expenditures for retirement, disability, and un- employed, for example, rose from 20 to 29 per- cent of total Federal outlays between 1968 and 1972; R&D expenditures, however, were less than 1 percent of the total outlays in this area. In the space area, R&D expenditures declined both in absolute terms and as a percent of total Federal outlays. Total outlays for defense, which has been the major source of R&D funds, fell from 49 percent of all Federal outlays in 1963 to 35 percent in 1972. (For further detail, see An Analysis of Federal R&D Funding by Function, National Science Foundation, NSF 72-300.) R&D Activities in Functional Areas Expenditures for the 10 major areas of federally funded R&D are presented for the years 1963-72 in figure 17.2 These areas 1 Expenditure data for other funding sources (e.g., industry) are not available for these functional areas. 2 Comparable data are not available for earlier years. 23 NR Figure 16 Federal R&D Expenditures as a Percent of Total Federal Outlays, FY 1961-72 (Percent) 13 1961 '62 ‘63 ‘64 ‘65 ‘66 ‘67 '68 OOS 70k ieee 72 Fiscal year (est.) SOURCE: National Science Foundation eee scneees accounted for 99 percent of all Federal expendi- tures for R&D in 1972. The most salient features represented in the figure are: e The large role of defense R&D throughout the period e The rise and fall of space R&D e The relatively rapid growth in civilian areas. Defense R&D expenditures between 1963-72 ranged from 48 to 64 percent of total Federal expenditures for R&D. In 1972 they accounted for 54 percent, the highest proportion since 1964. Current dollar expenditures for 1972 were the highest of the 1963-72 period. The 1972 R&D expenditures in this area were directed in the main to development of missiles, aircraft, equipment, and to defense-related atomic energy activities, military sciences, and astro- nautics. R&D expenditures for space in 1972 were at their 1963-64 level after declining by more than 40 percent (in current dollars) since the peak year of 1966.3 The area, however, still received 19 percent of all Federal R&D funding in 1972. 3 The entire activities of NASA are reported as R&D or related support; the R&D component was reported as 98 percent of the agency’s total expenditures in 1972. 24 The principal programs, in terms of magnitude of expenditures in 1972, were manned space flight, space science and applications, support- ing activities, and space technology. Recent de- clines in the space area occurred largely in the manned space program. Expenditures for R&D in civilian areas—areas other than defense and space—grew sub- stantially throughout the 1963-72 period, rising from 14 percent of total Federal R&D expendi- tures in 1963 to 27 percent in 1972. The func- tional areas accounting for most of the civilian- oriented R&D in 1972 were: (1) Health, which consists of the development of health resources, the prevention and control of health problems, and the delivery of health care. The first category, which accounts for some 90 per- cent of all Federal expenditures for health- related R&D, includes activities of (a) the 10 National Institutes of Health which deal with specific chronic and communicable diseases as well as general medical sciences, develop- ment of health manpower, and establish- ment of biologic standards; (b) the mental health, health statistics, and overseas re- search activities of the Health Services and Mental Health Administration (HSMHA); (c) the medical and prosthetic research of the Veterans Administration; and (d) the health- related activities of the Atomic Energy Commission. The second category consists of the R&D activities of the Food and Drug Administration, Bureau of Mines, and the preventive health services of HSMHA. The delivery of health care category is comprised of the HSMHA programs in health services planning and development, health services delivery, andIndian health services. Expendi- tures for R&D in the entire health area, as a fraction of total Federal R&D expenditures, rose from 5.2 percent in 1963 to 8.7 percent in 1972. (2) Advancement of Science and Technology, which is aimed at strengthening the Nation’s scientific base and at application of science and technology to problems of national concern. The largest category is general science, comprised principally of basic research proj- ects in the various scientific disciplines supported by the National Science Founda- tion and most of the physical science re- search programs (except for controlled thermonuclear research) of the Atomic Energy Commission. A second category is es Figure 17 Federal R&D Expenditures for Selected Functions, FY 1963-72 wus Constant 1958 dollars” eumemeeee Current dollars Fiscal 0 1,000 2,000 3,000 4,000 year T 6,000 7,000 8 C National ae defense 364 Space 1387 Health Millions of Dollars 0 100 200 300 400 500 600 700 800 | | | | | | l Advancement of science and technology Environment Transportation Energy ce conversion 267 and a development — 4975 Agriculture }3e Economic 967 security 6 Education 3 (a) GNP price deflator was used to convert current to constant dollars. SOURCE: National Science Foundation. ee eee eee 25 507-194 O- 73 -3 (3) (4) (5) 26 the technology improvement and — innovation programs of the National Bureau of Stand- ards. As a fraction of total R&D expendi- tures by the Federal Government, this area rose from 2.6 percent in 1963 to 4.4 percent in 1972. Transportation, which consists of R&D in air, ground, and water transportation. Air transportation R&D (which accounted for 70 percent of Federal expenditures for all transportation R&D in 1972) includes NASA's aeronautical technology program, and the activities of this agency and the Department of Transportation in the areas of system safety and future generations of aeronautical vehicles. Ground transportation R&D is aimed largely at improved highway and automotive safety and at rapid transit systems. R&D in water and multimodal transportation includes programs of the U.S. Coast Guard, Maritime Administration, and others. Expenditures for transportation R&D, as a fraction of all Federal R&D expenditures, increased from 1.0 percent in 1963 to 3.8 percent in 1972. Environment, which encompasses the pollu- tion control and abatement programs of the Environmental Protection Agency and the environmental research programs of the Atomic Energy Commission; resource develop- ment and management which includes programs of the Forest Service, National Oceanic and Atmospheric Administration (NOAA); Office of Saline Water, and others; and resource monitoring, measuring, and forecasting con- sisting of the R&D activities of the Geologi- cal Survey and NOAA. As a fraction of total R&D expenditures by the Federal Govern- ment, R&D spending in this area increased from 1.5 percent in 1963 to 3.2 percent in 1972: Energy Conversion and Development, which consists mainly of development of nuclear energy capabilities (85 percent of R&D expendi- tures) and the development and utilization of non- nuclear energy resources. Nuclear energy activi- ties are concentrated on development of the liquid-metal fast breeder reactor; major efforts in the nonnuclear area—which are rising in both absolute and relative terms—center on coal gasification, oxide control technology, and advanced under- ground electric transmission lines. R&D expenditures in this area, as a proportion of all Federal R&D outlays, rose from 2.3 per- cent in 1963 to 2.5 percent in 1972. (6) Agriculture, includes R&D activities aimed at increasing the quantity and improving the quality of agricultural products and expanding the utilization of agriculture resources. The first category, which comprised more than 90 percent of total R&D expenditures throughout 1963-72, includes the efforts of the Agricultural Re- search Services and the Cooperative State Research Service of the Department of Agriculture; R&D in the second category in- cludes activities of the Economic Research Service and the Farmer Cooperative Serv- ice. As a proportion of all Federal R&D expenditures, those in this area were 1.2 per- cent in 1963 and 1.8 percent in 1972. (7) Economic Security, which consists of man- power resources development, reduction of poverty, and income maintenance. R&D in this area is aimed primarily at improving the employability of individuals, promoting equality of opportunity, providing systems of income maintenance, and _ alleviating poverty. Expenditures for such R&D—provided pri- marily by the Department of Health, Educa- tion, and Welfare and the Office of Economic Opportunity—increased from 0.2 percent of total Federal R&D expenditures in 1963 to 1.0 percent in 1972. (8) Education, which consists of the R&D activi- ties of the Office of Education, the National Institute of Education, and the Office of Child Development, all of the Department of Health, Education, and Welfare. R&D is spread among a wide range of efforts, in- cluding the development of improved curriculums and individualized instructional materials, better understanding of the learn- ing process, and the motivation of dis- advantaged children. The fraction of total Federal R&D expenditures for this area rose from 0.1 percent in 1963 to 0.8 percent in 1972. RESOURCES FOR INDUSTRIAL R&D Total expenditures for industrial R&D, which include expenditures of both government and private industry, are shown in figure 18. The separation of these two funding sources indi- Figure 18 Industrial R&D Expenditures, 1961-72 Total (Billions of Dollars) 20 Current dollars Constant 1958 dollars ‘” o=s = x oF By Source (Billions of Dollars) : Ve Federal Government er @=anF Constant a NS sss cotars ¢ x 5 ¢? ¢ X 7 eas | 5 a 4 0 [sees ieee [eee oan oT 1961 '63 65 67 '69 a 1k (est.) (a) GNP price deflator was used to convert current to constant dollars SOURCE: National Science Foundation cates that the decline in total R&D expenditures in 1970 was due entirely to reductions in the level of Federal support. Federal funding actually leveled off in 1966 while industrial support rose more rapidly than in previous years, with the result that industry replaced the Federal Government in 1968 as the principal source of support for industrial research. By 1972, industry funded 58 percent of allindustrial R&D compared with 43 percent in 1961. Although Federal funding for industrial R&D did not start its decline until after 1968, the effects of a relatively slow rate of growth in funding, compared with the increasing cost of R&D, were apparent as early as 1964 in terms of the source of support for R&D scientists and engineers (figure 19).4 The figure shows that the number of scientists and engineers supported by Federal funds started to decline after 1964, al- though the largest reductions did not occur until after 1969, which coincides with the onset of larger constant dollar funding reductions that are shown in figure 18. These funding changes did not affect appreci- ably the relative distribution of funds among basic research, applied research, and develop- ment activities. In 1972 as in 1961, industrial R&D was concentrated in development (78 per- cent), while applied research received some 19 percent and basic research declined from 4 to 3 percent. (The absolute level of basic research, however, declined considerably as shown else- where in this report.) Despite these funding changes, industrial firms still perform the bulk of the Nation’s R&D. In 1972, funding of industrial R&D accounted for 68 percent of all R&D conducted in the United States, including 83 percent of the development, 55 percent of the applied research, and 16 percent of the basic research. Industry-Funded R&D R&D in this category is performed largely in the seven industries indicated in figure 20. They accounted for about 85 percent of all company R&D expenditures during the period reported. Each of these industries, except for “aircraft and missiles” and “motor vehicles,” had increasing R&D expenditures (in current dollars) through 4 Comparable data are not available for years prior to 1962. 27 Figure 19 Scientists and Engineers” Engaged in Industrial R&D, by Source of Funds, 1962-71 (Thousands) 240 — ~ ~ 230 ar] Industry 220 WE federal Government 210 F 200 190 180 170 0 1962 ‘63 ‘64 ‘65 ‘66 ‘67 ‘68 ‘69 ‘70 ‘71 January (a) Includes all scientists and engineers (full-time-equivalent basis) SOURCE: National Science Foundation 1970 (the latest year for which such data are available). In terms: of constant dollars, how- ever, only the petroleum, machinery, and professional and scientific instruments indus- tries indicated increases in 1970. Industry-funded R&D in 1972 was con- centrated in development activities which re- ceived 73 percent of the total funds as compared with 22 percent for applied research and 4 per- cent for basic research. This composition changed continuously over the decade toward more development (from 68 to 73 percent), less applied research (from 25 to 22 percent), and less basic research (from 7 to 4 percent). These long- term shifts do not appear to have been accel- erated by recent funding changes. 5 ~ Figure 20 Industry’s Own Funds for R&D, by Selected Industry, 1961-70 Current dollars (Millions of Dollars) 2,200 2,000 Electrical equipment 1,800 1,600 Chemicals ,e® 1,400 ° eee . 1,200 Machinery 1,000 200 Aircraft and missiles 600 Petroleum ke se = a le — --" oem Professional and scientific instruments Constant 1958 dollars (Millions of Dollars) 1,600 SENS 1,400 }— Electrical equipment Chemicals 1,200 }— gcecccee®™ Machinery 1,000 ° Motor vehicles ee 800 beeccce®” Aircraft and missiles 600 }— 400 Professional and scientific instruments | 1961 ‘62 63 ‘64 ‘65 66 ‘67 wea | 68 69.70 (a) GNP price deflator was used to convert current to constant dollars. SOURCE: National Science Foundation. Company-funded R&D is projected to in- crease by about 25 percent between 1972 and 1975, rising to some $14 billion, and the number of scientists and engineers employed in such R&D is anticipated to increase to 260,000 in 1975.5 Individual industries show some varia- tions from these projected trends. The drug industry (a part of the larger chemical industry) anticipates increases in R&D spending which are larger than the all-industry average; electronic firms expect increases in line with the rest of industry; companies in industrial chemicals and aerospace foresee future R&D growth at a pace somewhat below the rest of industry; and petro- leum firms expect only a slight increase in total R&D spending over the next few years. Federally Funded R&D The Federal Government funds R&D prin- cipally for defense and space purposes. In 1970, for example, all but 14 percent of the Federal funds for industrial R&D came from the Depart- ment of Defense and NASA. The funds from Federal agencies are directed to a small number of industries, with more than 90 percent of the funds going to five industries: aircraft and mis- siles; electrical equipment and communication; machinery; motor vehicles and other transporta- tion equipment; and professional and scientific instruments. Some 80 percent of all Federal expenditures for R&D in industry go to the first two industries.° Federal funds are concentrated in develop- ment activities, more so than are the funds sup- plied by industry. On the average throughout the 1961-72 period, development received about 85 percent of the total funds, applied research 12-14 percent, and basic research 2 percent. The funding reductions noted above had the greatest absolute effects on development activities, expenditures for which peaked in 1966 and subsequently declined 25 percent in constant dollars between then and 1972. (Estimated cur- rent dollar expenditures indicate a small in- crease in 1972.) Similarly, the 1972 funding level for applied research declined by 18 percent after its peak funding year of 1962. But propor- 5 Projections are based on a National Science Foundation survey in 1972 of 50 of the largest corporations in the United States. © National Science Foundation, Research and Development in Industry 1970, NSF 72-309. tionally, basic research was even more ad- versely affected; the 1972 funding level was down by 40 percent since 1967, the year of its maximum funding. R&D FUNDING BY UNIVERSITIES AND COLLEGES These institutions together with other non- profit organizations provide the remaining 5 percent of total national expenditures for R&D, with the universities accounting for 4 percent. University expenditures are concentrated in research, with basic research accounting for 78 percent of expenditures and applied research 20 percent in 1972. This pattern of funding distribution persisted with only minor changes throughout the 1961-72 period.” Research expenditures reported by universi- ties and colleges come from various non-Federal sources, including State governments, indus- tries, and foundations as well as from university funds. In 1970 the separately budgeted research expenditures from non-Federal sources were distributed among major fields of science as follows: Percent Mifes Scien cesu eee saree eine eee ee 47 Social/Sciemees’ ...ses024-046aseuse waoee 16 Empine@@nring: oo. ens ecie sented: seers 1 12 PhysicallSciences: yaas.2 44 sec acess ee 9 Mathematics sane: aera teeters 5 Environmental Sciences: .....4..<<..62008-0% 4 Bsy.cholopiye aesus.tsyoscasieiare ona ao Sea 22ree eee 0 By Source of Funds Current dollars (Millions of Dollars) 4,500 4,000 — 3,500 — Total 3,000 — 2,500 — Federal Government 2,000 — 1,500 — f Universities & colleges oma mt Industry saan {se S2"2=5 Other nonprofit institutions eooet 1960 "62 ‘64 66 '68 ‘70 72 (a) Federally Funded Research and Development Centers. (b) GNP price deflator was used to convert current to constant dollars. SOURCE: National Science Foundation. Constant 1958 dollars ‘) (Millions of Dollars) 3,000 — Total 2,500 — Federal Government 2,000 — 1,500 — yA Universities & colleges Industry — so es ee ee oe et Other nonprofit institutions 1960 ‘62 ‘64 ‘66 ‘68 70 72 (est.) 35 clined by some 10 percent, in constant dollars, for the period 1968-72, while Federal support of its own basic research decreased 6 percent in the same period. Also affected in this period were industry and nonprofit institutions where the reductions in Federal funds were 16 percent and 7 percent, respectively. Six agencies of the Government supply almost 95 percent of all Federal funds for basic research (figure 23). About 50 percent is provided by two of them—the National Aeronautics and Space Administration (NASA) (31 percent) and the Figure 22 Federal Expenditures for Basic Research, by Performer, 1960-72 (Millions of Dollars) 1,400 Current dollars 1,200 -— Universities & colleges & = x 1,000 4 ~ = @ Constant 1958 dollars @ 800 600 Federal intramural 400 FFROC's 200 Other nonprofit institutions 0 | | | | | [ree Ea || 1960 62 ‘64 66 68 70 72 (est.) (a) GNP price deflator was used to convert current to constant dollars. (b) Administered by universities SOURCE: National Science Foundation 36 Department of Health, Education, and Welfare (HEW) (20 percent). The proportion of total Federal funds for basic research provided by each of the several agencies during the 1960-72 period changed significantly. Chief among these were (a) the growth of NASA’s share from 16 percent in 1960 to 31 percent in 1972; (b) the de- cline of the share of the Defense Department (DOD) from 28 to 11 percent—a shift which occurred concurrently with the growth of NASA’s share; and (c) the decline in the Atomic Energy Commission’s (AEC) share from 17 to 11 percent. The 1967 decline in basic research obligations for DOD and AEC appear to ac- count largely for the reduced rate of growth in overall Federal expenditures for basic research which occurred in 1968. As noted earlier, NASA provides more funds for basic research than any other Federal agency. The entire activities of that agency, however, are considered as either R&D or support of R&D (outlays for construction of facilities). The latter now comprises less than 2 percent of total out- lays, and has never exceeded 14 percent. NASA’s obligations for basic research (as well as for applied research and development) include the related costs of spacecraft, launch vehicles, tracking and data acquisition, and the pro rata costs of ground operations and administration. The estimated FY 1973 Federal obligations (in current dollars) for basic research indicate an in- crease of almost 9 percent over the obligation level of FY 1972, which in turn represented a 12- percent increase over FY 1971.3 The increase in basic research expenditures in FY 1973 is ex- pected to be less than obligations. BASIC RESEARCH IN UNIVERSITIES AND COLLEGES Estimated expenditures for basic research in universities and colleges are shown in figure 24 in both current and 1961 dollars, for selected scientific fields.4 (Expenditure data for years 3 National Science Foundation, Federal Funds for Research, Development and other Scientific Activities, Vol. XXI NSF 72-317. In press. 4 The recently developed Academic R&D Price Index (A Price Index for Deflation of Academic R&D Expenditures, National Science Foundation, NSF 72-310) was used to convert current to constant dollars. The conversion, it should be noted, may not fully reflect increases in indirect costs which reduce the actual level of research; these costs appear to have increased at an even faster rate than direct expenses, as shown elsewhere in this report. ——EEE ~~ _—_EEEEEEEEEEEEEEEEEEEnEEEenetmmeemnmmmememE EERE EEEEEEERRaEEEEEeaeEREe Figure 23 Federal Obligations for Basic Research, by Supporting Agency, FY 1960-72 Current dollars (Millions of Dollars) 800 700 600 = =" All others ¢ see 1p Cee . egpereroeerecers Agriculture pass TTT eee 0 é | | | | 1960 62 ‘64 '66 ‘68 ‘70 72 Fiscal years (est.) (a) GNP price deflator was used to convert current to constant dollars SOURCE: National Science Foundation Constant 1958 dollars” Millions of Dollars) 800 7 700 ;-— 600 }— 500 400 300 200 All others is o* eee ecccceteecnes®” =? geneetseee eeeee > P goseeeseeree® Agriculture eoao® i | | | | | 1960 '62 64 66 68 70 72 Fiscal years prior to 1964 are not available.) The 10 fields represented in figure 24 form three groupings in respect to the relative growth of expenditures between 1964-72: fields recording the largest growth were the social sciences, environmental sciences, and psychology; fields with an inter- mediate level of growth were biological sciences, mathematical and computer sciences, engi- neering, and clinical medicine; and those with the smallest growth were chemistry, astronomy, and physics. Current dollar expenditures for all fields—ex- cept mathematical and computer sciences—in- creased from 1970 to 1972. In constant 1961 dollars, however, expenditures for 1972 de- clined or remained essentially unchanged from their 1970 level in physics, chemistry, astronomy, and engineering as well as in the mathematical and computer sciences. The de- clines were due principally to reductions in Federal expenditures for basic research, as shown in figure 25. Current dollar expendi- tures by the Federal Government increased between 1970-72 for all fields except physics and the mathematical and computer sciences. In terms of constant dollars, however, 1972 Federal expenditures were lower than 1970 expendi- tures in 6 of the 10 fields, with the largest reduc- 37 ed Figure 24 Estimated Expenditures for Basic Research in Universities and Colleges, by Field of Science, 1964-72 Current dollars Constant 1961 dollars (Millions of Dollars) 600 560 | ; | 520 + 480 440 Biological cls sciences *. in 400 360 | [ Clinical medicine 320 280 "3 Engineering 240 | & cs Social F sciences 200 -— 160 pee eeewneecaee Physics . Seneeeeseene,, Environmental , ‘ sciences 120 |, auscesensseses = Chemistry 80 kena Psychology ", “ts. Mathematical * & computer 40 sciences 1964 '66 '68 ‘70 ‘72 1964 ‘66 68 ‘70 ‘72 (a) Academic R&D price deflator SOURCE: National Science Foundation 38 Figure 25 Estimated Federal Expenditures for Basic Research in Universities and Colleges, by Field of Science, 1964-72 Current dollars (Millions of Dollars) 380 280 |- 0 1964 ‘66 ‘68 (a) Academic R&D price deflator SOURCE. National Science Foundation. ee eeee Clinical medicine Biological sciences Physics . Environmental | sciences Social sciences Chemistry Psychology Mathematical & computer sciences Astronomy Constant 1961 dollars” 1964 66 68 ‘70 340 320 300 240 220 200 180 120 100 80 60 40 20 39 tion (16 percent) occurring in physics. Federal funding for the latter field decreased by some 25 percent between 1968-72. In view of the declining research expendi- tures coupled with increased costs of per- forming research anda larger faculty body, some change in the involvement of academic staff in research is inevitable. One aspect of this change is indicated in figure 26, which presents the funds for applied and basic research per scien- tist and engineer (excluding graduate students) in Ph.D.-granting institutions. (These institu- tions accounted for 96 percent of all academic research in 1972.) The 1972 level of such funds was 15 percent lower than in 1968, in terms of 1961 dollars. This reduction is attributable to the Figure 26 Federal and Non-Federal Research Funds per Scientist and Engineer in Doctorate- Granting Institutions,’ 1964-72 Constant 1961 dollars “” (Thousands of Dollars) 14 Total Federal Non-Federal 1964 66 68 70 72 (a) Includes all scientist ie 'b) Academic R&D price deflator ind engineers (full-time-equivalent basis) employed in SOURCE: National Science Foundation 40 declining level of Federal funding between 1968- 72, coupled with the increasing number of scien- tists and engineers in these institutions; Federal research funds per scientist and engineer de- clined by 24 percent during the 1968-72 period, in contrast to the relatively unchanged level of support from non-Federal sources. The individual fields were affected somewhat differently by the combined changes in the levels of funding and scientific and engineering man- power, as shown in figure 27. (The fields of Figure 27 Research Expenditures per Scientist and Engineer in Doctorate-Granting Institutions, by Selected Field of Science, 1964-72 Constant 1961 dollars” (Thousands of Dollars) 26 Biological sciences Physics Engineering Chemistry Clinical medicine Psychology Social sciences Mathematical & computer sciences a p a | | _| 1964 66 ‘68 70 72 (a) Includes all scientists and engineers (full-time-equivalent basis) employed in universities (b) Academic R&D price deflator SOURCE: National Science Foundation astronomy and environmental sciences are not included because of the lack of acceptably relia- ble data.)5 Although the overall trend is a reduc- tion in the level of research support per scientist and engineer, the funds for some fields declined much more than others. Research funds per physicist, for example, declined by 35 percent between 1966-72, while funds per social scien- tist changed little even though the number of such scientists increased rapidly. The number of scientists and engineers en- gaged in research and development (on a full- time-equivalent basis) declined slightly between 1969 and 1971. This may represent a reduction in the average time devoted to R&D by the staff as a whole and/or a reduction in the number of staff engaged in any R&D at all. The available data are not sufficient to resolve this ambiguity. Moreover, it is generally difficult to accurately separate the time devoted to research from other academic activities. The proportion of the Ph.D. science staff in these institutions receiving Federal support and engaged in basic research is shown in figure 28 for several scientific fields.e The figure indicates that the proportion of Ph.D. academic staff who were wholly or in part supported by the Federal Government and devoted some portion of their time to basic research’? was 57 percent in 1970, down from 69 percent in 1964 and 1966. The largest decreases were in mathematics, chemis- try, psychology, and physics. Research support for young investigators® is of particular interest as an indicator, since the progress and quality of future research and innovation depend increasingly on individuals from this group. Federal support for young 5 It should be noted that the considerable variation in the level of funding among fields reflects, among other factors, differences in the cost of research associated with each field; some fields, for example, require extensive equipment for re- search while others require little. © Included are Ph.D.’s employed by academic institutions who indicated that basic research was their first or second work activity. In 1970, these persons accounted for about 75 percent of all academic Ph.D.’s. 7 These data are based on responses to the National Register of Scientific and Technical Personnel for the years 1964, 1966, 1968, and 1970. It is estimated that the responses account for approximately 80 percent of the Ph.D. scientists employed by universities and colleges. 8 Defined as those employed by colleges and universities who have held the Ph.D. less than seven years and who re- ported their primary or secondary work activity as basic research. 507-194 O- 73 - 4 Figure 28 Proportion of Ph.D. Academic Staff in Science Receiving Federal Support and Engaged in Basic Research, by Field, 1964-70 (Percent) 85 80 Physics Biology 75 Chemistry 70 All fields 65 Psychology 60 55 Earth & Marine 50 Mathematics 45 40 Social sciences 35 30 | | ——— 1964 ‘66 68 ‘70 SOURCE: National Science Foundation investigators engaged in basic research de- creased in recent years, falling from 64 percent in 1964 to 50 percent in 1970 (figure 29). While the proportion of senior investigators receiving such support also declined, the reduction was not so large as for the young investigators. Moreover, proportionally fewer young re- searchers obtained Federal support in certain fields (as indicated in figure 30), especially in mathematics, social sciences, and psychology, where the ratio of young to senior basic re- 41 Figure 29 Proportion of Young and Senior Ph.D. Academic Staff in Science Receiving Federal Support and Engaged in Basic Research, 1964-70 (Percent) 80 Senior researchers 70 Young researchers 60}— 50 [— 40 f- 0 Nelleisiies we 1964 '66 638 70 SOURCE: National Science Foundation. Figure 30 Ratio of Young to Senior Ph.D. Academic Staff Receiving Federal Support and Engaged in Basic Research, by Field, 1964-70 Mathematics Physics 10 95 65 a 60 85 55 50 s P Earth & marine sciences AS Social sciences 95 90 85 80 15 10 Biological sciences Psycholo 90 y ey 85 80 15 70 65 4 1964 66 68 ‘70 1964 ‘66 68 70 SOURCE: National Science Foundation. Chemistry 42. searchers declined between 28 and 18 percent. In other fields, the decline was less than 10 per- cent. Research at universities cannot be charac- terized completely by the parameters discussed so far, namely, manpower and funds. The state of basic research in universities must also be related to the health of the institutions them- selves. This is especially significant since the universities have traditionally shared the cost of research in their laboratories. The symbiosis between universities and basic research makes the overall financial situation of the universities a cause for concern in assessing the state of science, especially its future prospects. While re- search support will not by itself solve the finan- cial problems of universities, its decline has con- tributed to their difficulties, in that research grants often carry a number of continuing university costs of a long-term nature, such as building maintenance, administration, and a por- tion of long-term salary commitments to faculty. SS Figure 31 Federal Expenditures for Intramural Basic Science, 1960-72 (Millions of Dollars) 700 650 |— 600 ;— 550 iz Current dollars 500 -— 450 }— 400 }— Constant 1958 dollars) 350 F— 300 op IST esa SS et ee a q 1960 62 64 66 68 ‘70 72 (est.) (a) GNP price deflator was used to convert current to constant dollars SOURCE: National Science Foundation. BASIC RESEARCH IN FEDERAL LABORATORIES Total Federal expenditures for in-house basic research increased over the 1960-72 period in terms of both current (259 percent) and con- stant (154 percent) dollars (figure 31). How- ever, during the 1970-72 period, expenditures decreased by 13 percent in terms of current dollars and by 19 percent when expressed in con- stant 1958 dollars. The declines occurred prin- cipally in laboratories funded by the National Aeronautics and Space Administration (NASA) and the Department of Health, Education, and Welfare (HEW). In 1972, the in-house basic re- search effort represented 23 percent of all Federal expenditures for basic research. Simi- larly, it accounted for 14 percent of the total expenditures for basic research in 1972, com- pared with 12 percent in 1960. Data are not available for segregating the activities by field of science, but obligation data are presented in figure 32 for the Federal agencies which support the bulk of in-house basic research. In 1972, the total Federal obliga- tions for in-house R&D were divided among the agencies as follows: NASA (29 percent), Depart- ment of Defense (18 percent), Agriculture (17 percent), HEW (13 percent), Interior (8 per- cent), and Commerce (7 percent). TF SC ee ees Figure 32 Federal Expenditures for Intramural Basic Research, by Selected Agency, 1960-72 Current dollars (Millions of Dollars) 240 220 — 200 — 180 — 160 — 140 — 120 — 100 — 80 — Fiscal year (a) GNP price deflator was used to convert current to constant dollars. SOURCE: National Science Foundation. Constant 1958 dollars (@) (Millions of Dollars) 240 220 — 200 — 180 — 160!— 140 — 120 — 100 — 80 — 60 — rye Yas i woetceeeee” poe PL hbk Ot hd e AGUCULNG or ee aes HEW °° 40 — Interior gests 20 ote om me oe 1960 62 '64 66 68 ‘70 72 (est.) Fiscal year ee SSSFSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSsssesFeFesFeFessses 43 BASIC RESEARCH IN INDUSTRY Expenditures for basic research in industry are shown in figure 33. In current dollars, total basic research expenditures increased by 3 percent over the period 1968-72; in constant 1958 dollars, the decline in overall funding was 14 per- cent between 1968-72, which included a 16-per- cent reduction in Federal funding and a 13-per- cent decline in industrial support. Some 75 percent of all industrial basic research was performed by four industries in 1970: chemicals and allied products (36 percent); elec- trical equipment and communication (24 per- cent); aircraft and missiles (10 percent); and petroleum refining and extraction (5 percent).° There is considerable difference among these industries in the percentage of total R&D which is devoted to basic research. The chemical indus- try in 1970 devoted 12 percent of its R&D to basic research, as compared with 3 percent for the electrical equipment industry, 4 percent for the petroleum industry, and 1 percent for the aircraft and missiles industry. The principal fields of science in which basic research is performed are shown in figure 34 for the 1967-70 period, the only years for which such data are available. As shown, the fields of chemistry and engineering receive almost 65 percent of all basic research expenditures. A major change over the 4-year period was the reduction of basic research in the fields of phys- ics and astronomy; although the available data do not permit the separation of these two fields, expenditures for physics presumably accounted for most of the joint activities of these fields. In respect to the industries involved, basic re- search performed by the chemical industry accounted for almost 80 percent of the life science and 46 percent of physical science expenditures of all industries. The bulk of 44 Current dollars orn oo” Constant 1958 dollars! engineering research expenditures were by elec- trical equipment industries (41 percent) and air- craft and missiles industries (21 percent). ° National Science Foundation, Research and Development in Industry, 1970, NSF 72-309, Figure 34 Industrial Basic Research Expenditures, by Field of Science, 1967-70 Current dollars (Millions of Dollars) 250 Chemistry Physics & astronomy 50 — Mathematics 5 ee ees ee ee 0 | Environmental sciences | 1967 68 ‘69 70 (a) GNP price deflator was used to convert current to constant dollars SOURCE: National Science Foundation. Constant 1958 dollars ‘) (Millions of Dollars) 200 — Chemistry Engineering Mathematics 70 SD Ss es ye ee oe eae ee ee ee 0 Environmental sciences 1967 68 '69 45 Science and Engineering Personnel Science and Engineering Personnel Indicators in this section chart the growth of the national pool of scientists and engineers; present trends in the utilization of such personnel; provide data on undergraduate and graduate enrollments and financial support; depict growth patterns in the production of scientists and engineers; and measure changes in their employment level. Available data and present methodology did not permit the development of indicators of the quality and productivity of the Nation’s scientists and engineers. For the same reasons, it was not possible to devise reliable indices of the future demand for and supply of such personnel for inclusion in the present report. Improvements in both data (e.g., better information on the utilization of scientists and engineers in non-academic, non-R&D activities and the role of immigration and emigration of personnel) and methodology (e.g., better techniques for predicting the future state of the economy and for anticipating the emergence of national problems requiring the services of scientists and engineers) are required for more reliable forecasts of the supply-demand situation. As methodologies and data series used in projections are improved, indicators of supply-demand relationships will be included in future reports of this series. INDICATOR HIGHLIGHTS ao The total pool of active scientists and engi- o The distribution of scientists and engineers 48 neers in the United States grew by about 50 percent from 1960 to 1971, rising to some 1,750,000. The number with doctorates doubled during the period, reaching 10 per- cent of the total. Scientists and engineers comprised an in- creasingly larger proportion of total civilian employment over the last two decades, although the extent of the increase in the 1960's (167 to 210 per 10,000 workers) was less than that during the 1950's (93 to 167 per 10,000 workers). The proportion of natural scientists and engineers engaged in R&D increased to 37 percent between 1960-64, but declined steadily thereafter. This downward trend was more pronounced among academic than industrial scientists and engineers, and re- flects the growth in faculty needed for teach- ing, as well as the leveling off of R&D funds. Between 1968 and 1970 the number of natural scientists and engineers in industrial R&D declined to the 1967 level, the first such decline during the 1960’s. among major types of employers changed between 1960-70, with the proportion in industry declining from 74 to 66 percent and the fraction in universities and colleges rising from 11 to 18 percent. Total enrollments in high school courses of social sciences, natural sciences, and mathe- matics grew faster than total secondary enrollments between 1960-70, with the largest increases occurring in psychology and economics and the smallest in physics, chemistry, and mathematics. Physics was the only field in which the proportional growth was less than the increase in total enroll- ments. The number of undergraduate students at the junior-year level who declared majors in physics, engineering, and chemistry de- clined between 1970-71, whereas the number declaring majors in the applied social sciences and professional life sciences in- creased significantly. Graduate enrollments (full-time and part- time combined) in science doctorate depart- ments declined by almost 4 percent between 1969 and 1971, the first such absolute de- crease in the 1960’s. Such enrollments in science and engineering, as a percent of total graduate enrollments, declined steadily from 38 percent in 1963 to 31 percent in 1970. o The number of full-time graduate students in science and engineering receiving Federal support declined by 15 percent between 1969-71, while those depending on self- support increased by 19 percent froma lower base. a Annual awards of bachelor’s level degrees in science and engineering increased by a factor of 2.2 over the 1959-71 period, with the largest gains in the social sciences (4.1 times) and the smallest in the physical sciences (1.3) and engineering (1.2). First degrees in sci- ence and engineering, as a fraction of all bachelor’s level degrees, remained essen- tially constant at 30 percent, due in large part to the rapid growth of social science degrees. o Annual awards of master’s degrees in science and engineering rose by a factor of 2.5 over the 1959-71 period, with the largest gains in mathematical sciences (3.8) and social sci- ences (3.1) and the smallest in the physical sciences (1.9). Science and engineering mas- ter’s degrees, as a fraction of all master’s de- grees, declined from a high of 30 percent to @ Substantial changes in the demand for scien- tists and engineers, which may be produced by factors such as a redirection in Federal funding or the state of the national economy, may occur over periods of from two or three years. But be- cause scientists and engineers require training extending over several years, serious im- balances of supply and demand, inadequacies of training, maldistributions among areas of competence, and similar problems may be cor- rected only over a longer five- to ten-year span. Therefore, the extended time and high cost in- volved in producing scientists and engineers re- quire that careful, continuous attention be given to the nature, quality, and applicability of their professional training. CURRENT POOL OF SCIENTISTS AND ENGINEERS! Magnitude The total pool of active scientists and engi- neers in the United States grew by some 50 per- 22 percent in 1970-71, with the largest proportional declines occurring in engi- neering and the physical sciences. o Annual awards of Ph.D. degrees in science and engineering rose by a factor of 3.0 over the 1959-71 period, with the largest gains in engineering (4.6) and mathematical sciences (4.4) and the smallest in the physical sciences (2.4). Science and engineering Ph.D. de- grees, as a fraction of all Ph.D. degrees, de- clined from 62 percent in the mid-1960’s to 58 percent in 1970-71, with the largest proportional declines in the physical sciences. ao During the last decade, awards of science and engineering doctorates, in terms of location of high school graduation, became more evenly distributed among geographic regions of the United States. The proportion, how- ever, is almost 50 percent lower in the South Atlantic and East South Central regions than in other areas of the country. a Unemployment rates for scientists and engi- neers rose after 1969, reaching 2.6 and 2.9 percent, respectively, by early 1971. These rates—which were less than half those reported for all workers—declined to early 1970 levels by late 1972. cent between 1960 and 1970, rising from about 1,170,000 to more than 1,700,000 (figure 35). (It is estimated that by 1971 this number had fur- ther increased to about 1,750,000.) This rapid growth was due to an increase in science and engineering degree holders and to the “up- grading” of nondegree personnel, principally engineers. From 1960 to 1970 scientists and engineers with doctoral degrees increased from 90,000 to over 170,000. The number of engi- neering doctorates tripled during this period, while science doctorates increased by about 75 percent. As a result, the percentage of scientists with doctoral degrees remained almost con- stant during the past decade, while the percent- 1 Information on the current numbers, types of employer, employment activities, etc., of scientists and engineers must be assembled from numerous data sources. Since not all of these are updated annually, it is possible to develop the latest complete picture only for 1970, even though many pieces of this mosaic are already available for 1971. Unless otherwise specified, scientists and engineers include the natural scien- tists (including mathematicians), social scientists, and engi- neers. 49 Figure 35 Distribution of Scientists and Engineers, by Broad Field, 1960 and 1970 Total (Thousands) 1,200 1,096.0 1,000 — 800 — 795.0 ——— Engineers 600 — 400 — Mathematicians Social scientists 200 — — Life scientists — Physical scientists 1960 1970 SOURCE: Bureau of Labor Statistics and National Science Foundation. age of engineers with doctorates doubled—from a low base of less than one percent. In absolute terms more than four times as many science (90,000) as engineering (22,000) doctorates were awarded during this period. Thus, science doctorates as a percentage of all scientists con- tinued to be a much larger ratio than engi- neering doctorates as a percentage of all engi- neers. All told, more than 185,000 scientists and engineers with doctorates were working in the United States by 1971. Another measure of growth is the ratio of the number of natural scientists and engineers to total civilian employment (figure 36). This ratio increased from 93 per 10,000 working adults in 1950 to 167 in 1960 and then to 210 in 1970. Thus, the extent of the growth was greater in the 1950’s than in the 1960’s. R&D and Non-R&D Activities The relative changes during the last decade in the numbers of natural scientists and engineers engaged in R&D—as well as in the percentages of all natural scientists and engineers in R&D—have been substantial (figure 37). In 50 Ph.D.’s (Thousands) 160 146.1 140 — 120 — 100 — 82.0 80 — —=—— Mathematicians =— Social scientists 60 — A — a Life scientists Engineering jim oT 2a— | 75 Physical scientists 1960 1970 Figure 36 Natural Scientists and Engineers in Relation to Total Civilian Employment, 1950-70 (Number) 220 200 — Natural scientists and engineers _ per 10,000 in civilian employment ™ 180 — 160 — LOS 0 sete O48 7560 '-'58, 2160) a u'b2 a b4 am bbuem Oden 0 SOURCE: Department of Labor. or Figure 37 Natural Scientists and Engineers Engaged in R&D, 1960-70 (Percent) Percent of total in R&D 44 424 Total U.S. __ Industry 40 — a 38 = a 36 = 34 32 @ 30 Sa (Percent) 44 Academia = 42 — 40 Sa 38 Ee 36 34 o = Sax Cys a 1960 65 70 1960 70 = 1960 ‘65 (Index) Relative change (1960 = 100) (Index) 160 160 159 — Total US. _ Industry Academia = 150 140 = a 130 = 120 = 110 @ 100 - 1960 65 70 1960 SOURCE: U.S. Department of Labor and National Science Foundation. = 130 = 120 ‘70 ~—:1960 65 70 EE 1970, for the first time since 1960, the total number in R&D decreased. This trend con- tinued in 1971. Most of the decline took place in industrial R&D and was sufficient to reduce the total in this sector to the 1967 level. Not sur- prisingly, this downward turn followed closely the first reduction in national R&D expendi- tures (in constant dollars) in the 1960-70 period, a decrease felt most in industrial R&D. (See figure 14.) Another indicator of the downward trend is the decline in the percentage of all natural scien- tists and engineers engaged in R&D (figure 37). This trend began as early as 1964, when it had reached a high of 37 percent. The proportion then steadily declined to 34 percent in 1970. This trend was more pronounced among academic than industrial scientists and engineers. The relative decline of the proportion in R&D in industry suggests a greater growth of scien- tists and engineers in technological operations, management, and other non-R&D activities, and possibly a lower priority for R&D in periods of slow economic growth. The first factor was probably a prime cause for the decreases in the mid-1960’s, while the second was dominant at 51 the end of the decade. The decline in the proportion of academic scientists and engineers engaged in R&D, which started as early as 1961, was primarily due to the growth in faculty needed for teaching the rapid- ly increasing number of college students and, secondarily, the leveling off of funds (in con- stant dollars) available for R&D. The first factor was dominant in the early 1960's, while both factors were important in the late 1960's. Another noteworthy development of the 1960's was the increase (from 5 to 10 percent) of those with natural science and engineering doctorates in non-academic, non-R&D activi- ties.2 This change is especially significant as it took place during a “sellers’ market,” in which the individual scientist or engineer could usually choose his area of work. This employment trend is expected to continue, possibly at an accelerated rate, because of expected changes in the supply/demand relationship for doctorates. The percentage of doctorates involved in R&D is considerably higher than for nondoctorate scientists and engineers, although there is sub- stantial variation from field to field (figure 38). Especially notable are the relatively low percent- ages of Ph.D. mathematicians and social scien- tists involved in R&D. Distribution by Employment Sector The percentage distribution of scientists and engineers among types of employers changed somewhat during the 1960's, with a relative de- cline in industry and an increase in universities and colleges (figure 39). While the large size of industry as an employer tends to obscure rela- tive increases in other sectors, figure 39 shows a significant growth in the number of scientists and engineers employed in universities and colleges and other nonprofit institutions. ENROLLMENTS AND DEGREE PRODUCTION Enrollments in High School Science Courses An early indicator of changes in student interest in science is secondary school enroll- ments in science courses relative to total enroll- ment. Figure 40 relates enrollments for 1960-61 2 National Science Foundation, 1969 and 1980 Science and Engineering Doctorate Supply and Utilization, NSF 71-20. 52 Figure 38 Distribution of Scientists and Engineers, by Activity and Broad Field, 1970 Non-Ph.D.’s (Percent) 100 —— 75 Other 50K 25 tees 0 Total Physical Life Mathe- Social Engineers scientists scientists maticians scientists Ph.D.’s (Percent) 100 5 Other 50 a 0 Total Physical Life Mathe- Social Engineers scientists scientists maticians scientists SOURCE: U.S. Department of Labor and National Science Foundation and 1969-70 to a 1948-49 based index. Overall, enrollments in science and mathematics courses grew faster than total enrollment, with the largest growth occurring in psychology and economics and the smallest in physics, chemistry, and mathematics. The above aver- age increase in social science course enrollments may be due in part to the limited availability of such courses in earlier years. Only in the field of physics was the growth less than increases in total enrollment. College and University Enrollments and Degree Production Enrollments. The fraction of first-year college students who intend to work toward careers as — anne Figure 39 Employment of Natural Scientists and Engineers, by Sector, 1960 and 1970 Percent distribution 1960 Other nonprofit 1970 l nstitution ; Industry Percent increase, 1960 to 1970 | SOURCE: National Science Foundation Total Industry Federal Government Colleges & universities Other nonprofit institutions research scientists decreased steadily, from 3.5 percent in 1966 to 2.5 percent in 1971. How- ever, because of larger total enrollments, the absolute number of students intending to pursue this career remained at about the same level. Interest in engineering careers decreased from 9.0 to 5.3 percent over the same period. Although college freshmen frequently change their career interests, such changes have been generally away from science and engineering. This early indicator becomes more significant when related to fields selected by junior-year undergraduates for their major area of study. While total junior-year undergraduate enroll- ments increased by 7.6 percent between the fall of 1970 and the fall of 1971, fewer students chose majors in physics, chemistry, engi- neering, and mathematics, while basic social science, other physical science, and life science majors increased; applied social science students grew markedly (figure 41). 3 American Council on Education, Four Years After College Entry, ACE Research Reports, Vol. 8, no. 1, March 1973. Similar trends are evident in terms of enroll- ments for advanced degrees. Annual data from the Office of Education indicate that total enroll- ment for advanced degrees in science and engi- neering fields more than doubled between 1960- 70. However, such enrollment, as a percent of that in all fields, remained constant at about 38 percent until 1963, before declining steadily to 31 percent in the fall of 1970. Engineering and the physical sciences accounted for most of this decline. Related data, though not strictly comparable with those of the Office of Education, indicate some recent trends in graduate enrollment. Data collected by the National Science Foundation from 2,579 Ph.D.-granting departments showed a decline of 7.1 percent in first-year, full-time science and engineering graduate students from fall 1969 to fall 1971, with most of this change occurring in the last year. In the same period, the number of full- and part-time graduate students in these fields declined by 3.7 percent. The over- all change in first-year, full-time students includes greater-than-average declines in mathematics and the physical and social sciences, 53 Figure 40 Public Secondary School Enrollment in Selected Sciences and Mathematics Courses and Total Enrollment in Grades 9 through 12, 1960-61 and 1969-70 (Index, 1948-49 = 100) 100 200 300 400 =, 700 800 T T \ T l 1960-61 Biology . Chemistry Physics Economics Sociology Psychology Introductory algebra Introductory geometry Advanced mathematics Total enrollment grades 9-12 SOURCE: U.S. Office of Education and National Science Foundation. and smaller decreases in engineering, psychol- ogy, and the life sciences, as shown below. Change in First-year, Full-time Graduate Enrollment in Science and Engineering in Doctorate Institutions, 1969 to 1971 Percent change 1969 to 1971 All areas: 2 43: Sheen cc eee or — 7.1 Engineering’ S20 -tnacn-acenemeenans — 0.3 Physicalisciences: (a. 4 se eee —15.2 Mathematics s-s--ee eee eee —12.0 hifelsciences: passerine eee — 53 Socialisciences! +e senor — 93 Psychology s.ctoe «cae neers — 3.3 54 Figure 41 Percent Change in Majors Declared by Junior-Year Students, 1970 to 1971 (Percent) 10 0 10 20 30 40 50 i : == = = = Physics aaa Engineering Be «| Chemistry | Mathematical sciences | Basic social sciences ee ‘All other” life sciences BRE “Other” physical sciences Lene? Preprofessional life sciences ae: Se Applied social sciences Re SaaS SOURCE: Ace Higher Education Panel Financial Support. The availability of financial support may influence the number of graduate students entering the sciences and engineering, although the measurement of such direct effects may be confounded by university efforts to provide support for students in all fields of grad- uate study. Moreover, there are certainly other motivational factors affecting the choice of field for graduate education. The sources of graduate support for major fields of science and for engineering are shown in figures 42 and 43. The largest percentage of students supported by fellowships, trainee- ships, and assistantships is in the physical sciences, and the smallest in the social sciences. The number of science and engineering students supported by the Federal Government declined by 15 percent between 1969-71 whereas those depending on self-support increased by 19 per- cent (from a lower base). Graduate Production. Annual awards of bachelor’s and first professional degrees in the sciences and engineering are shown in figure 44 for the 1959-71 period. The annual recipients of Figure 42 Distribution of Full-time Graduate Students in Doctorate Departments, by Area of Science and Type of Support, 1967-71 (Percent) 0 20 40 60 80 100 uf y y y Outside support (fellowships, traineeships, research Self and and teaching assistantships). other support. RS SIN eS All areas Physical sciences Life sciences Mathematical sciences Psychology Engineering Social sciences SOURCE: National Science Foundation. Figure 43 Distribution of Full-time Graduate Students in Science and Engineering, by Source of Support, 1969-71 (Thousands) 0 40 80 120 160 | all sources U.S. Government support Institutional support Support from all other sources Self-support SOURCE: National Science Foundation. social science degrees increased by a factor of 4.1 over the period, well above the growth (2.2 times) in the total science and engineering de- grees awarded at that level. Social science de- grees—as a proportion of all first degrees in sci-. ence and engineering—rose from about one- fourth in 1959-60 to almost one-half in 1970-71. The annual production of graduates in the life and mathematical sciences increased by factors of 2.2 and 2.4, respectively, over the period, whereas those receiving degrees in the physical sciences and engineering rose by factors of only 1.3 and 1.2, respectively. First degrees in science and engineering, as a fraction of first degrees in all fields, remained essentially constant at some 30 percent between 1959-71. The large increase in the annual recipients of social science degrees was responsi- ble for maintaining the fraction at a constant level; engineering degrees, as a proportion of degrees in all fields, declined continuously from 9.6 percent to 5.1 percent during the period, and the physical sciences fell from 4.1 to 2.4 percent. 55 All science and engineering fields Life sciences Engineering - == msg oe Of eee es oe 1 fa Mathematical sciences, |, seees*s eee = = on ee ee es =a ie 722 a Physical sciences All science and engineering fields Life sciences © on on Ce ) Mathematical sciences 9S mwane did fag 28 8 oo ooo co ce coc e eee ees Se ae ee oe os oe Physical sciences al] Annual awards of master’s degrees in science and engineering are shown in figure 45. The number of these degrees awarded annually in- creased by a factor of 2.5 during the period, with the largest increases occurring in the mathe- matical (3.8) and social (3.1) sciences and the smallest in the physical sciences (1.9). As a frac- tion of master’s degrees in all fields, sciences and engineering degrees declined from a high of 30 percent in 1964-65 to 22 percent in 1970-71. The largest proportional declines were in engi- neering and the physical and life sciences. This may indicate that relatively fewer persons were seeking advanced degrees in these fields, or that there is a trend toward working directly for the Phe: Annual awards of Ph.D. degrees are presented in figure 46. The greatest growth occurred in engineering, which inreased by a factor of 4.6, and in the mathematical sciences (4.4), both of which exceeded the 3.0 increase for total Ph.D. degrees in science and engineering. The life and social sciences increased by factors of 2.9, and the physical sciences by 2.4. As a percentage of Ph.D. degrees in all fields, the annual recipients of doctorate degrees in the sciences and engi- neering declined from a high of 62 percent in the mid-1960’s to 58 percent in 1970-71. The largest proportional declines were in the physical sciences. The rapid growth of recipients of science and engineering degrees is not a development specific to science. Even in the case of the Ph.D. degree, where the growth rate was greatest, the ratio of science to nonscience Ph.D.’s has re- mained almost constant since the early 1920’s. Furthermore, the rapid growth rate is not solely a matter of advanced education. A large part of our modern 20th-century society exhibits the same rapid growth; this appears in such areas as the annual production of books, telephones in use, production of electronic systems, consump- tion of electricity, and use of raw materials. While growth rates in science and engi- neering degrees granted during the 1960's were substantial, they at most kept pace with degrees granted in other fields. Actually, first degrees in natural science and engineering—and nearly all advanced science and engineering degrees—grew more slowly than degrees in all other fields combined. The relative decline has been most pronounced for first and master’s degrees in engineering and for doctorates in the physical sciences. Taken as a whole, these indi- cators point to a relative decline in students Figure 45 Master’s Degrees in Science and Engineering, 1959-60 to 1970-71 (Thousands) Number of degrees awarded 50 45 — Total science and engineering 40 3) 30 - 25 20 os— Engineering __pmue oo 10 - a” = Social sciences 5 ai F Physical sciences 0 As a percent of all master’s degrees (Percent) Total science and engineering Engineering euasseeenns 10% A > Social sciences me Pk hee wore eeccseetcoen, ° ° Life sciences | on cceetee* a Ah aces ecceeees mn oe oe oe et 5 esMathematical sciences Be ° Physical sciences Se eS ee ee ee 1959- 60- 61 62, 63- 64 65 66 67- 68 69 /70- BOR Oar 625 a nbd bd ameponie Goi 67: sl BBenogioan7 0h gt Academic year SOURCE: U.S. Office of Education. 507-194 O- 73 - 6 Figure 46 Doctor's Degrees in Science and Engineering, 1959-60 to 1970-71 Number of degrees awarded (Thousands) 20 18 - All sciences and engineering fields Life sciences Z Physical sciences on ~~ e ===: o awane nme iia Feeesrerrrn ee kat social sciences eooeeee eooeee Fhaineering AR ° Mathematical sciences eooeeoee®? eecee eccaseenee® As a percent of all doctor's degrees (Percent) 64 All sciences and engineering fields 62 - 60 — 20;— - 18 =e Physical sciences ic; Life sciences a Social sciences °*e Seescse voeeee ; e i — ASS oe” =, == = @# Engineering oe” pee 6 — Mathematical sciences = : 1959- 60- 61- 62 63- 64 65- 66 67- 68 69- 70- GO free G lies 76250 G3 have Odie ic 65) 6b abyei oe 69% -720te 7L Academic year SOURCE: U.S. Office of Education. CO _ 37 majoring in science and engineering fields, al- though the absolute numbers of those who so choose are still increasing. Doctorate Awards by Geographic Region Figure 47 shows science and engineering Ph.D.’s awarded in terms of the geographic region of high school graduation in relation to the population of those regions. This ratio varies considerably with New England, the Middle Atlantic, West North Central, and Mountain regions showing larger proportions, while the South Atlantic and East South Central regions are lower by almost 50 percent. This indicator shows an uneven pattern of pursuit of advanced education (to the doctorate) among high school graduates, although some progress was made in reducing these regional differences during the last decade. Figure 47 Geographic Origins, by High School Graduation, of Ph.D.’s in Science and Technology, 1970 1970 WEST NORTH CENTRAL wv 15cm! & WEST SOUTH CENTRAL 1970 compared with 1960 WEST MOUNTAIN NORTH CENTRAL F 4 mOmTH oaKoya p WEST SOUTH CENTRAL (a) Includes Alaska and Hawaii SOURCE: National Research Council and National Science Foundation. CENTRAL CENTRAL NEW ENGLAND w EAST Degrees award/10,000 population Te 0.9 or more 0.7 to 0.89 DO 05 t0069 Less than 0.5 ATLANTIC NEW ENGLAND Ww. EAST Ratio of degrees awarded/ 10,000 population 1970 to 1960 ee 3.00 or more 2.50 to 2.99 PI) 2.00 to 2.49 Less than 2.00 wg ¢ MIDDLE ; bes) ATLANTIC =) 4 v SOUTH ATLANTIC 58 SUPPLY AND UTILIZATION Unemployment For most of the 1960's the production of new scientists and engineers could not match the de- mand for their services. In recent years, how- ever, the demand for scientists and engineers de- clined as a result of several converging factors: R&D funding (in constant dollars) declined on the average by 1 percent per year between 1967 and 1972, due primarily to an average annual de- cline of 3.3 percent in Federal R&D funding; con- currently, the Nation underwent a fairly steady period of inflation, reduced economic growth, and less emphasis on space and defense. However, because of the long timelags in the response of the educational system, the produc- tion of scientists and engineers continued in spite of the fall in demand, creating a supply/de- mand mismatch. Unemployment of scientists and engineers accelerated from 1969, reaching about 2.6 percent for scientists and 2.9 percent for engineers by early 1971 (figure 48). National Science Foundation surveys show that unemployment in 1971 was more severe for engineers than for scientists, as indicated in figure 49; that, generally, those with higher de- grees were less likely to be unemployed; and that younger scientists and engineers were most adversely affected. Unemployment rates were more severe in the defense and aerospace areas and in specific disciplines such as physics. Among unemployed scientists and engineers, defense (11 percent) and space-related activities (4 per- cent) were most frequently cited as the last areas of employment. Although there was a relative increase in the unemployment of scientists and engineers, the base level for such a comparison was low. Even with the large relative increases up to 1971 the overall science and engineering unemployment rate was still only about half that for all workers. The unemployment situation has improved somewhat since then. The unemployment rate for scientists and engineers declined in 1972, as has that for all professional workers, and employment prospects for new graduates were reported as better in 1972, although still not as good as those in the mid-1960’s. Underemployment Although unemployment has been relatively small, the change from a “sellers’ market” to a Figure 48 Unemployment Rates, 1963-72 (Percent) 7 Percent 6 All worker =m O25 ony Engineers id Quarters All workers Professional and technical workers orn x’ 1b x ae | Engineers @& Ree yf 1963) 164 ee GSvetGb aoe B]is ees O8i SOS m0 71 72 SOURCE: U.S. Department of Labor. “buyers’ market” has tended to produce under- employment—employment that fails to fully utilize the training of scientists and engineers. Although a real problem, underemployment is difficult to assess since “underutilization of training” is a subjective judgment. Some indica- tion of underemployment of new Ph.D.’s can be inferred from a survey of university depart- ment chairmen conducted by the National Re- search Council.4 The survey found that in January 1971, 1.2 percent of new (1968-69) Ph.D.’s were listed as employed in positions that did not make appropriate use of their graduate training, and that this percentage was double that reported the year before. 4 National Research Council, Employment of New Ph.D.’s and Postdoctorals in 1971, Washington, D.C., 1971. 59 Figure 49 Unemployment Rates for Scientists and Engineers, by Age Group and Highest Degree, 1971 By age group (Rate) 5.0 40 3.0 2.0 — 10 - Be 0 BS pe gs ic aS oa ES | 24 and 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65 years under Age group and over By highest degree Scientists Engineers 5 4 3 2 1 Q (Percent) 0 1 2 Bane: 5 | | | Total Doctorate Master's Bachelor's Less than bachelor's SOURCE: National Science Foundation Holding Actions Several other factors should be considered in a review of employment. For most of the 1960's, the percentage of those who planned to con- tinue their training immediately upon receipt of their science and engineering Ph.D.’s remained essentially constant, except for those in the life and physical sciences, who showed a steadily in- creasing tendency toward postdoctoral study (figure 50). Then, in the late years of the decade, the fraction of all Ph.D.’s taking postdoctoral study increased somewhat.> This may, in part, have been an early indicator of the shrinking employment market for scientists and engi- S The sharp increases shown in figure 50 between 1968 and 1969 are misleading because of changes in definitions in postdoctorate study. However, analysis of the data indicates increases in the proportion over earlier years. 60 neers. The number of science and engineering Ph.D.’s in temporary postdoctoral study was still increasing in 1971. The availability of post- doctoral study thus provides a number of new Ph.D.’s with an alternative to employment competition and at least temporarily helps re- lieve pressure on the labor market. A related matter is the length of time between receipt of the bachelor’s degree and the doctor- ate. The median time had been decreasing in almost all fields when, in 1968-69, it started to rise again in the physical sciences, mathematics, engineering, and the life sciences (figure 51). This may be due in part to graduate students attempting to prolong their study because of the scarcity of jobs. Age Distribution Without a flow of young people into science and engineering, creativity in these fields would tend to decline overall, as would be the case in any profession. In the last decade or so, the number of scientists or engineers, particularly the doctorate population, has grown substan- tially. Therefore, with this large flow of new en- trants the average age has not risen; for doctor- ates it has actually declined. Based on current and projected degree produc- tion and projected growth in employment, it appears that the average age of the science and engineering population will not rise sub- stantially in the coming decade. However, even though the overall average age may not change markedly, the proportion in the younger age groups will decrease significantly unless recruit- ment of new young talent continues. Other Changes As mentioned previously, the deployment of doctorates into nonacademic, non-R&D activi- ties increased during the 1960’s. This trend in- creased during the most recent years, as supply expanded relative to demand. Similarly, in- creased employment of doctorates was evident in junior and community colleges, as well as in 4- year institutions. The number of doctorates em- ployed full time in the former institutions in- creased nearly twice as fast as all science and engineering staff between 1969 and 1971, with the result that the proportion of full-time staff holding doctorates increased from 8.8 to 10.6 percent in 2-year institutions.° © National Science Foundation, Resources for Scientific Activities at Universities and Colleges, 1971, NSF 72-315. Figure 50 Percent of New Science and Engineering Ph.D.’s Planning to Engage in “further education or training” or ‘postdoctoral study,” by Field, FY 1960-71 (Percent) 40 4 / o 30 — Life sciences ee s == Physical sciences = ysical sciences All science and engineering fields . Mathematics e “=” Engineering fame ee ee | AQGON 61), 62%, -163') 164: S65 p66 s aby GSE 69 70s a7 Fiscal year of doctorate (a) Due to a change in definition, 1969 through 1971 data are not strictly comparable with earlier years SOURCE: National Research Council a _ Figure 51 Median Number of Years from Baccalaureate to Doctorate of Doctorate Recipients in Science and Engineering, by Field, FY 1960-71 (Year) 10 e PS e. Social sciences ° om *. “~ Physical sciences & s ooos, e a= 6 — . Ngee cast eke eooee ° Mathematics i; = 0 [exe leew AGGO SGI e625 163%5'64s 65 ce 60. “Osa G85 209) Oar Fiscal year of doctorate SOURCE: National Research Council 61 Institutional Capabilities Institutional Capabilities This chapter presents indicators of the state of the institutional system of science and technology. The indices include aspects of the infrastructure involved in training scientists and engineers; types and numbers of organizations engaged in R&D; composition and patterns of concentrations among these organizations; and expenditures for research equipment and facilities. Indicators of these several aspects are presented in the context of the system of institutions—colleges and universities, Federal installations, and industry—within which the bulk of training and R&D is accomplished. The present indicators in this area are incomplete in several respects, primarily because of the lack of current and/or detailed information. No indicators are presented for nonprofit institutions or for local government; only fragmentary and dated information was available on the number, size, and activities of Federal installations; indicators of industrial R&D are limited to relatively aggregated aspects of expenditure and manpower; and indices of the state of research equipment and facilities do not include information on the quantity, type, and utilization of scientific instruments and specialized facilities. INDICATOR HIGHLIGHTS The number of academic institutions award- ing degrees in science and engineering in- creased from some 1,100 in 1960-61 to almost 1,300 in 1969-70, with the largest in- creases occurring in institutions which awarded master’s and Ph.D. degrees. Doctoral-granting institutions employed almost 75 percent of all academic scientists and engineers in recent years, and awarded more than 80 percent of all master’s degrees in science and engineering and more than 50 percent of the bachelor’s degrees. The 20 institutions awarding the most Ph.D. degrees in science and engineering accounted for a decreasing fraction of all such degrees awarded, down from one-half of the total awards in 1963 to two-fifths in 1971. Science and engineering graduate enrollments in these institutions declined proportionally over the period. program per 26 departments during 1970-72; plans for 1972-74 indicate a reduction of the ratio of new additions to 1:66. The largest net increases were in the areas of computer sci- ences and psychology. Expenditures for laboratory equipment, pro- vided through research grants from the National Science Foundation and major Na- tional Institutes of Health, declined between 1966-71. These expenditures as a fraction of total grant funds, fell from 12 percent to 6 percent during the period. Federal obligations to universities and colleges for R&D plant and major equip- ment declined 75 percent between 1965 and 1971. As a proportion of all Federal obliga- tions for academic science, funds for R&D plant dropped from 8 percent to 1 percent during the period. A radio astronomy facility (known as the a Private doctoral institutions awarded a de- VLA) authorized and funded in FY 1973 was creasing proportion of all Ph.D. degrees in the first new major research facility started science and engineering, falling from 41 per- since 1968, although some 30 facilities, in cent of the total awards in 1963 to 34 per- various areas of science, were proposed in cent in 1971. Science and engineering grad- recent years and evaluated as technically de- uate enrollments in these institutions peaked Eoialolandiconesble: in 1969 and declined thereafter, in contrast to , en oi Mee the continued growth of such enrollments in Federal intramura expenditures 1 blic instituti creased throughout the 1961-72 period, with public institutions. é the Department of Defense accounting for a New doctoral programs in existing doctoral the largest share of such funds, followed by 64 departments increased at the net rate of 1 NASA, the Department of Agriculture, and the Department of Health, Education, and Welfare. Together, these four agencies accounted for 86 percent of total such ex- penditures in 1972. a Federal intramural R&D expenditures, as a percent of total U.S. R&D expenditures, in- creased from 13 percent to 16 percent between 1960-72, while the number of scien- tists and engineers (full-time-equivalent) en- gaged in such R&D rose from 12 to 13 per- cent of the U.S. total between 1961-71. o Large industrial companies (5,000 or more employees) employed an increasing propor- tion of the total industrial R&D personnel between 1963 and 1971 (up from 79 percent to 85 percent), while the share for small companies (less than 1,000 employees) de- clined from 10 percent to 6 percent. o The R&D intensiveness of U.S. industry, as measured by the ratios of R&D expendi- tures to net sales and R&D scientists and engineers to total employment, increased @® Scientific and technological activities are per- formed through a mutually complementary system of institutions and associated human re- sources. The system of institutions consists principally of colleges and universities, which both train scientists and engineers and perform research; Federal laboratories, which focus pri- marily on research and development directly re- lated to their respective missions; and private industry, which conducts research and develop- ment leading to new and improved technology, processes, and products. The characteristics and capabilities of this system can be described in terms of the types of institutions involved, the activities they perform, and the effectiveness with which the education and/or R&D func- tions are carried out. SCIENCE AND ENGINEERING EDUCATION General Institutional Capabilities Higher education in the sciences and engi- neering grew rapidly throughout the past dec- ade. The number of colleges and universities awarding science and engineering degrees in each year between 1961-71 is shown in figure between 1960-64 but declined thereafter toa level in 1970 which was lower than in 1960. The largest declines occurred in the most R&D intensive industries. o The R&D intensiveness of manufacturing industries declined by some 25 percent between 1964-70 in the five most R&D- intensive industries, and remained essen- tially unchanged in other manufacturing industries during the 1960-70 period. o The R&D intensiveness of nonmanu- facturing industries increased by 50 percent between 1960 and 1966, but remained con- stant thereafter through 1970. a Industrial R&D is concentrated in relatively few manufacturing industries and in a small number of companies within these industries. In 1970, five industries had 81 percent of the total industrial R&D expendi- tures, while accounting for only 48 percent of total manufacturing sales, and 100 companies had some 80 percent of the total expenditures. 52. The major growth was in institutions which granted the master’s and Ph.D. degrees; their numbers increased by 57 and 45 percent, respec- tively. Institutions at the bachelor’s degree level failed to show the same systematic growth, in part because of the widespread evolution of these colleges into higher level institutions. Scientists and engineers employed by universities and colleges are concentrated in doctoral-granting institutions. In 1971 these institutions employed almost 75 percent of all academic scientists and engineers, a fraction which remained essentially unchanged during recent years! (figure 53). The institutions awarded more than 80 percent of all master’s de- grees in science and engineering and more than 50 percent of the bachelor’s degrees during the 1964-70 period (figure 54). Patterns of Growth in Doctorate Institutions Institutional capabilities for graduate education in science and engineering grew in 1 Two-year institutions and other institutions which do not grant science or engineering degrees are not included. 65 Figure 52 Number of Institutions of Higher Education by Highest Degree Awarded in Science and Engineering, 1960-61 to 1970-71 (Number) 800 = aes S23 eee eee Bachelor's-granting institutions 700 — 600 — 500 — 400 — 300 — Master’s-granting institutions (a 200 Doctorate-granting institutions 100 — ) SS SSS ee 1960- 61- 62. 63 64 65 66 67 68 69 70- 61 62 63 64 65 66 67 68 69 ZO ir (a) Not available SOURCE: U.S. Office of Education and National Science Foundation two ways: the number of graduate-level institu- tions increased and existing graduate institu- tions expanded their graduate programs. Some aspects of the growth pattern of doctoral institu- tions are shown in the following tables. The first table shows the growth in the number of Ph.D.-level institutions and the divi- sion of doctorate awards among them. The increasing number of institutions are divided into three groups, in terms of the number of Ph.D. degrees awarded. Number of institutions granting Ph.D.'s 1962- 1965- 1969- 1970- 63 66 70 71 Institutions granting First one-third of Ph.D.’s ..... 11 13 15 16 Next one-third of Ph.D.’s .... 24 28 35 36 Last one-third of Ph.D.’s ..... TD pe A Ge 17.2 77, Total number of INStItUtIONS acme esos gyrA alts pad, = G2eK2) 66 Figure 53 Scientists and Engineers Employed by Universities and Colleges, 1965-71 Number of scientists and engineers (Thousands) 60 50 = 40 — Doctorate-granting institutions 100 = 90 = 80 Percent of total 1965 1967 1969 1971 70 Bachelor's 14 12 12 11 Master's 13 13 16 15 60 — _ Doctorates 73 74 73) 74 50 = 40 — Master’s-granting institutions 30 204 en ae ss SSD rs Bachelor’s-granting institutions 10= 9 ee eee 1965 67 69 ‘71 (a) Includes all scientists and engineers (full-time-equivalent basis) employed in universities. SOURCE: National Science Foundation The table shows that the number of institutions in each group increased, with the largest increase occurring in institutions which awarded the smallest number of Ph.D.’s. Growth in the last group of institutions, how- ever, was matched by the expansion of graduate programs in the larger Ph.D.-granting institu- tions. Thus, in 1970-71 as in 1962-63, 7 percent of the institutions produced one-third of all sci- ence and engineering Ph.D.’s, 15 percent pro- duced the second third, and 78 percent pro- duced the remaining one-third of the Ph.D.’s. The proportion of full-time graduate students enrolled in the three groups of institutions over the 1962-71 period are shown in the following table. Figure 54 Science and Engineering Degrees Awarded at Baccalaureate, Masters, and Doctorate-Granting Institutions, 1963-64 and 1969-70 Doctorate-granting institutions Thousand 100 — e 60 — 40 — Master's degrees 20 — Ph.D. degrees Percent of total full-time enrollment for advanced degrees 1962- 1965- 1969- 1970- 63 66 70 71 Institutions granting First one-third of Ph.D.’s ..... 29 29 28 28 Next one-third of Ph.D.’s .... 29 30 31 31 Last one-third of Ph.D.’s ..... 42 41 41 41 This table shows the close correspondence between the proportion of graduate students in the three groups of institutions and the propor- tion of Ph.D.’s produced by them.? Similarly, there was little change over the period in the proportional distribution of students among the three groups of institutions. First-year, full-time graduate students shows a similar division and constancy among the groups of institutions. Percent of first-year full-time enrollment for advanced degrees 1962- 1965- 1969- 1970- 63 66 70 71 Institutions granting First one-third of Ph.D.’s ..... 25 25 25 26 Next one-third of Ph.D.’s .... 27 29 29 28 Last one-third of Ph.D.’s ..... 49 46 46 46 Baccalaureate and master's institutions Thousand: 100 — 80 — oo 60 — Bachelor's degrees Dog (b) es ~oeeo= - a= 0 — ee Other patterns of growth are indicated by using a different method of grouping institu- tions; namely, the first 20 institutions ranked in terms of numbers of Ph.D.’s awarded, the next 20 institutions, and all other doctorate-granting institutions. The individual institutions included in the first two groups changed in rank considerably between 1962-63 and 1970-71 as shown in the table on page 68. Institutions in these groups generally have the following characteristics: they were usually among the top institutions in terms of Federal funds received, were the choice of the largest numbers of Federal fellowship awardees, and were generally included among the highly ranked graduate departments as cate- gorized by the American Council on Education.3 The increase in Ph.D. degree awards and grad- uate enrollments among these groups of institu- tions are shown in figures 55 and 56. Although the number of science and engineering 2 The deviation may be due to a greater tendency for students in the last group of institutions to conclude their education with the master’s degree, to transfer to larger institutions for the Ph.D., or to discontinue graduate education. 3K. D. Roose and C. J. Andersen, A Rating of Graduate Programs, American Council on Education, 1970. 67 68 Changes in Ranking of 40 Institutions Conferring the Largest Number of Ph.D.’s in the Sciences and Engineering, in terms of Number of Ph.D.’s Conferred and Amount of Federal Obligations Received Rank in order of Ph.D. degrees conferred 1970-71 1962-63 1 1 University of Illinois 2 2 University of California-Berkeley 3 3 University of Wisconsin 4 7. University of Michigan 5 11 Stanford University 5 15 Cornell University 7 5 Purdue University 8 4 Massachusetts Institute of Technology 9 8 University of Minnesota 10 20 Michigan State University 11 14 University of California-Los Angeles 12 10 Ohio State University 13 6 Harvard University 14 22 University of Washington 15 9 Columbia University 16 18 Pennsylvania State University 17 13 Iowa State University 18 17 University of Texas 19 23 Northwestern University 20 12 University of Chicago 20 16 New York University 22) 25 University of Pennsylvania 23 49 University of Missouri 23 2 Case-Western Reserve University 25 27 University of Maryland 26 43 University of Tennessee 27 54 Texas A & M University 28 40 University of Florida 28 24 Indiana University 30 64 University of Arizona 31 21 Yale University 32 19 Princeton University 82 44 North Carolina State University-Raleigh 34 39 University of California-Davis 35 34 University of Colorado 36 77 University of Massachusetts 37 42 Oregon State University 38 45 University of Southern California 38 38 University of Kansas 40 33 Duke University 40 28 University of lowa 42 30 Louisiana State University 43 37 University of North Carolina-Chapel Hill 44 29 Rutgers-The State University 47 32 University of Pittsburgh 49 36 University of Utah 54 31 Johns Hopkins University 54 26 California Institute of Technology 57 354 Carnegie-Mellon University 1 Includes funds for medical schools, but not FFRDC’s 3 Ranked below 100 2 Separate institutions in 1962-63. Source: National Science Foundation Rank in order of total federal obligations! 1970-71 o 10 %) 4 7 LZ. 40 + Carnegie Institute of Technology in 1962-63. 1962-63 774 Figure 55 Number of Ph.D. Degrees Awarded in Science and Engineering by Selected Groups of Doctorate-Granting Institutions ® (Ph.D. degrees) 8,000 7,000 — First 20 6,000 — 5,000 — $ 100 Remaining institutions Second 20 3,000 — 2,000 — 1962- 1965- 1969- 1970. 63 66 70 71 (a) Ranked by number of Ph.D. degrees awarded in each year SOURCE: Office of Education and National Science Foundation doctorates increased in each group of institu- tions, the smallest growth occurred in the first and second 20 institutions. As a result of this pattern, the proportion of all such doctorates awarded by the first 20 institutions declined from 51 to 39 percent between 1962-63 and 1970-71, as compared with a nearly constant proportion of 20 percent for the next 20 institu- tions and an increase from 29 to 42 percent for the remaining institutions. Growth and distribution patterns similar to these occurred for graduate student enroll- ments, as shown in figure 56. Total full-time enrollments increased overall, although a small decline was recorded among the first 20 institu- tions in 1970-71. By the end of the period, the share of total enrollments for the first group of institutions had declined to 33 percent, down from 43 percent in 1962-63. The share for the Figure 56 Graduate Students in Science and Engineering Enrolled in Selected Groups of Doctorate-Granting Institutions ® (Thousands) Total full-time students 70 60 — Remaining Institutions 50 is First 20 40 — Second 20 0 | ahaa First year, full-time students a5 30 — Remaining Institutions 25 ae 20 = First 20 10 Second 20 0 | | 1962- 1965- 1969- 63 66 70 (a) Ranked by number of Ph.D. degrees awarded in each year SOURCE: National Science Foundation 1970- 71 69 next 20 institutions fell from 20 to 18 percent, whereas the proportion in other institutions rose from 37 to 49 percent. Shifts in the same direction occurred for first- year, full-time graduate enrollment. By 1970-71, the proportion of such students in the first 20 institutions declined to 29 percent, froma high of 37 percent in 1962-63. The share of such enrollments in the second 20 institutions re- mained almost constant at about 17 percent, in contrast to the proportion in all other institu- tions which climbed to 54 percent, up from 45 percent in 1962-63. This dispersion of graduate education during the last decade is of course a continuation of a long-run trend created by the increase in institu- tions involved in this level of education in sci- ence and engineering. Private and Public Institutions in Graduate Education The responsibility for graduate training of scientists and engineers is shared by private and public institutions. Despite their smaller number, private institutions have exerted much influence in shaping and enhancing graduate education as a whole. The number of public institutions granting the doctoral degree in the sciences and engi- neering increased somewhat more rapidly than that of private institutions over the 1963-71 period, as indicated in figure 57. In each of these years an average of five public and three private institutions awarded the doctoral degree for the first time. As a result, the number of public institutions awarding the doctoral degree ex- ceeded private institutions by 41 in 1971, as com- pared with 24 in 1963. While differences in the number of institu- tions of each type changed only moderately, doctoral graduates from public institutions in- creased more rapidly (figure 57). As a conse- quence, the proportion of graduates from public institutions rose from 59 to 66 percent during the period and those from private institutions declined correspondingly from 41 to 34 percent. When the level of recent graduate enroll- ments are considered, the diminishing role of private institutions is even more apparent (figure 58). Enrollments in private institutions peaked in 1969 and declined by 5 percent in 1970, while graduate enrollments in public institu- 70 Figure 57 Institutions Granting Ph.D. Degrees in Science and Engineering, by Control, 1963-64 to 1970-71 Number of institutions (Number) 140 130'— Public institutions 120 — 110 — 100 — Ph.D. degrees awarded (Thousands) 14 Public institutions NX 0 1963- 1964- 1965- 1966- 1967- 1968- 1969- 1970- 64 65 66 67 68 69 70 71 Academic year SOURCE: Office of Education and National Science Foundation. ee Figure 58 Graduate Enrollment by Control of Institutions, Selected Years, 1965-70 First-year graduate enrollments Total graduate enrollments (Thousands) 200 180 — 160 — Public institutions 140:— 120 — 100 — Private institutions 80 — J 60 — 40 — 20;— 1965 1969 1970 1965 1969 1970 SOURCE: National Science Foundation. tions continued to increase. By 1970, only 28 percent of the total science and engineering graduate students were in private institutions. Most of the decline (some 75 percent) was due to cutbacks in first-year enrollments; these fell by 9 percent between 1969 and 1970, in contrast to such enrollments in public institutions which in- creased 8 percent. The largest private institutions, in terms of the number of Ph.D.’s produced, had the largest declines in first-year enrollments. Of the 20 largest Ph.D.-producing institutions, public and private, 6 are private universities. Of the six, only one had an increase in first-year enroll- ments; as a group, enrollments declined by 12 percent between 1969 and 1970. Such enroll- ments in the 14 largest public institutions, on the other hand, rose by 5 percent during the same time.# Coincident with these recent declines in private institution enrollments were reductions in Federal R&D expenditures. Such funds de- clined by 8 percent (in constant dollars) between 1968-70 for private institutions, but remained essentially unchanged for public institutions. The magnitude of this decline is of considerable significance since 85 percent of research in pri- vate institutions is federally supported, com- pared with 65 percent for public institutions. While there are many factors affecting the growth and capacity of public and private institutions, a fundamental one appears to be the pervasive and worsening financial condition of these institutions. Growth of Doctoral Programs: 1970-74 A survey was conducted by the American Council on Education of doctoral departments in science and engineering to determine the recent and probable future growth of doctoral programs.® The survey indicated that the ratio of net additions of such programs to existing doctoral departments was 1:26 during 1970-72. Plans for the 1972-74 period, however, indicate a reduction to 1:66, i.e., anet gain of one program per 66 existing departments. The largest net in- creases in 1970-72 were in the areas of computer sciences and psychology. Plans for 1972-74 indi- cate the greatest relative increase will again be in computer sciences. RESEARCH EQUIPMENT AND FACILITIES Research instrumentation and modern labora- tories are basic tools of science. They provide man with his quantitative and most precise window on the real world. They permit the study of phenomena otherwise inaccessible to investigation, provide the means for accurate measurement and observation, and facilitate 4 Special tabulation based on data collected for the National Science Foundation report series, Resources for Scientific Activities at Universities and Colleges 5 National Science Foundation, Science Resources Studies Highlights, “Changes in Graduate Programs in Sciences and Engineering 1970-72 and 1972-74,” July 21, 1972 (NSF 72- 311). 71 data collection and analysis. Sophisticated equip- ment is now a prerequisite for significant re- search advances in most fields of science. The excellent instrumentation available in the United States is generally regarded as a prime factor contributing to the leading position of American science. Since the requirements for instrumentation are constantly changing with progress in science, a continuing investment is necessary to maintain the quality of this basic tool. The Federal Government is a prime source of support for research equipment and facilities. This includes basic laboratory equipment, as well as major equipment such as wind tunnels, accelerators, reactors, radio telescopes, etc., which are used for more than a single project, and R&D plant capital grants that fund the construction and maintenance of major R&D facilities. Research Equipment A major source of laboratory apparatus for universities and colleges has been the Federal system of research grants, which have often in- cluded funds for laboratory equipment as part of the grants. Funds for research apparatus from this source declined in recent years, even though the overall grant funds increased. The extent of the decline is suggested by figure 59 which depicts the proportion of total project grant funds allocated for permanent research equip- ment. The data presented here are fragmentary in that only the National Science Foundation (NSF) and a part of the National Institutes of Health are included;* they are, however, major sources of research equipment funds and prob- ably typify the situation in general. This figure shows that as a proportion of total project grant funds, support for the purchase of permanent laboratory equipment declined by one-half—from nearly 12 percent to 6 per- cent—between 1966 and 1971. Funds for re- search grants, on the other hand, increased by about 15 percent over the same period. In the case of the NSF, the reduced support for research equipment appears to have been absorbed largely by an upward shift in indirect ° National Institute of General Medical Sciences and National Heart and Lung Institute. LD, Figure 59 Proportion of NSF and NIH® Research Project Grant Funds Allocated for Permanent Laboratory Equipment, FY 1966-71 (Percent) 12 0 SE ee eis 1966 ‘67 68 69 ‘70 71 Fiscal year (a) National Institute of General Medical Sciences and National Heart and Lung Diseases. SOURCE: National Science Foundation. costs, as shown in figure 60.7 Salaries and wages, as a fraction of total grant funds, changed little during the period. R&D Plant and Equipment The Federal Government has been a prime source of support for major research equipment and R&D plant capital. However, Federal obliga- tions to universities and colleges for new and im- proved R&D plant declined after the mid-1960’s, 7 The higher proportion of NSF grant funds for research equipment in figure 60, over that shown in figure 59, is due to the inclusion of funds for both permanent and expendable equipment in figure 60. Figure 60 Distribution of NSF Research Project Funds, by Type of Expenditure, 1964-72 (Percent) 60 Total salaries and wages 50 — 40 — 30 — Research associates/assistants Wi salaries and wages Indirect costs ©3060 ene : Peeccceces, Joe" —_ Other costs mone EEE SE 1964 66 68 70 72 (a) Permanent and expendable equipment. SOURCE: National Science Foundation. falling by more than 75 percent between 1965 and 1971 (figure 61). Federal funds for R&D plant, asa fraction of total Federal obligations for academic science, declined from 8 to just over 1 percent during the same period (figure 61). Since the effectiveness and, in many in- stances, the significance of research depends directly upon the availability of appropriate tools and plant, expenditures for them are as essen- tial as expenditures for the performance of re- search itself. However, support for these critical items has fallen well behind that for research. Major Research Facilities Many fields of science now require large, specialized facilities to achieve significant ad- vances and to initiate research in promising new 507-194 O- 73-5 Figure 61 Federal Obligations for Academic R&D Plant, FY 1963-71 Total (Millions) 135 120 — Current dollars 60 — so — s0— 0 ae sO (ee) (cree arse (CS As a percent of Federal obligations for academic science (Percent) 1— 0 | Ee SSS 1963 ‘65 67 69 71 Fiscal year (a) Based on academic R&D price deflator. SOURCE: National Science Foundation. areas. Until the authorization of the radio astronomy facility, known as the Very Large Array, in fiscal year 1973, no major research facilities —those requiring $5 million or more for construction—have been started since the Batavia accelerator in 1968. Yet nearly 30 new facilities have been proposed and evaluated as technically desirable and feasible in recent years. These are listed in the table below along with their estimated costs. There is an evident need for such facilities in many scientific fields—physics and astronomy, biology, and environmental sciences as well as engineering. FEDERAL INTRAMURAL RESEARCH AND DEVELOPMENT The Federal Government’s R&D installations are engaged in a broad array of functions and activities, representing a significant portion of the national R&D effort. Organizationally and programmatically, each individual installation’s orientation is primarily to the needs and mis- sion of its parent or sponsoring agency. Both in the Congress and in the general public, there is a perceptible view that the Federal laboratories should be more widely utilized in helping to solve current national problems. Al- though Federal agency missions, in one way or another, are oriented to meeting national needs and responsibilities, there are no national R&D installations which are independent of a partic- ular executive branch agency for their programs, funding, personnel authorizations, and related essentials. Number of Installations Although an inventory of the security un- classified Federal R&D installations in 1969 listed a relatively large aggregate number of installations, there are only a few Federal Technically Desirable and Feasible Basic Research Facilities Costing $5.0 Million or More Environmental Biological Sciences Sciences Engineering Physical Sciences Upper Atmosphere National Resource Earthquake Engineering NRAO Upgrade 184” Synchrotron Observatory Centers ($50M Facility (630M) VLA ($76M) Berkeley ($7M) ($14M) over 10 years) Automation Technology Neroc 440’ dia. Heavy Ion Lab for NCAR 4th Brookhaven Center Institute ($5M) Steerable ($30M) Nuclear Physics (620M) generation for Biological Instrumentation ($5M) NOAA: Geophysical and Fluid Dynamics Laboratory computer ($18M) computer ($15M) Center ($8M) Cell Production and Fractionation Centers ($25M over 5 years) Construction and Industrialized Building ($6M) National Institute of Ecology ($10M) Productivity and Machine Design ($6M) Source: National Science Foundation 74 Engineering Software Technology Transfer Resource Center for Resource Center for Array of Gravitational Detectors ($5M) Owens Valley Inter- ferometer ($8M) Storage Ring Brookhaven Isabelle (650M) 100” Optical telescope ($5M) 200” Southern Hemisphere telescope (420M) Storage Ring Berkeley-Stanford ($75M) Project PEP NRAO Homology telescope Recirculating Linear Accelerator Upgrade SLAC ($17M) Stanford Mountain Top Cosmic Ray Observatory ($5M) Calculation Center for Chemistry ($10M) Upgrade Cornell Electron Synchrotron ($5M) agencies responsible for these installations.’ A total of 17 departments, independent agencies, and commissions reported nearly 700 installa- tions, over 85 percent of which are within the Departments of Agriculture, Defense, Interior, and Health, Education, and Welfare (HEW), and in the National Aeronautics and Space Administration (NASA). Another feature of the Federal in-house R&D installations is the large number of small units; approximately one-half of all the installations have nine or fewer professional staff members. The majority of these installations are under the jurisdiction of the Department of Agriculture’s Agricultural Research Service and Forest Serv- ice. Many of these small laboratories are actually in close juxtaposition to larger installations or are located on the campuses of land-grant institutions. At the other end of the size spec- trum, the largest installations belong to NASA, and are staffed by almost 5,000 professionals. Federal Funding of Intramural R&D Cost data on Federal laboratories can be approximated from funding data on Federal intramural R&D. Such data have been collected since 1955, and include overall costs of Federal R&D performance as well as costs of administer- ing R&D grants and contracts. Insome agencies, such as the National Science Foundation which maintains no Federal laboratories, and the Atomic Energy Commission, which has only three relatively small laboratories, the costs of grant and contract administration represent the great bulk of the intramural effort. In others, such as the Department of Defense, the costs of administering grants and contracts are a minor portion of total intramural costs. This situation would also hold for NASA, and the Depart- ments of Health, Education, and Welfare, Agriculture, Commerce, and Interior. Thus, funding data in this sector must be interpreted with considerable caution. Between 1961 and 1972 Federal R&D obliga- tions for intramural performance rose by a factor of two and one-third—from $1.9 billion to $4.5 billion—as Federal agency programs grew. In constant (1958) dollars the rise was not as steep—from $1.8 billion to $3.1 billion in 1972. Each year in the entire period marked a new high 8 National Science Foundation, Directory of Federal R&D Installations, NSF 70-23. in current dollars, although between 1966 and 1969 the rise was very moderate and, in real terms, actually showed a slight decline (figure 62). The rate of growth of Federal intramural R&D funding over the 1960-72 period exceeded that of the national R&D effort. During 1960, the Federal Government obligated $1.7 billion for intramural R&D which represented 23 percent of total Federal R&D obligations and 13 percent of total national R&D expenditures (figure 62). Estimates for 1972 indicate that Federal intra- mural R&D accounted for 27 percent of the Federal R&D budget and 16 percent of national R&D expenditures. Increases in the Federal intramural shares of R&D totals have resulted not only from higher intramural funding, but also from the fact that Federal funds for R&D in industry dropped in recent years, as the result of reductions in development programs of NASA and the Atomic Energy Commission. Funding by Agencies During this time, the Department of Defense (DOD) intramural R&D obligations were the largest of all agencies, although its share of the total of such Federal obligations decreased from 71 percent in 1961 to 54 percent in 1972 (figure 63). While the absolute level of DOD’s intra- mural funding showed an almost continuous in- crease, the expansion of intramural work on the part of other agencies—notably NASA and HEW—was responsible for the decline in the DOD share. NASA made up 10 percent of the Federal total in 1961 and rose toa high of 28 percent in 1965, reflecting the buildup of the Apollo program. Thereafter its share declined in most years, falling to 20 percent in 1972. The Departments of Health, Education, and Welfare (especially the National Institutes of Health), Agriculture, Interior, and Commerce experienced little change over the 1961-73 period in their respective shares of Federal intra- mural R&D funding. Two of these agencies, Agriculture and Commerce, rely on intramural resources more fully than extramural per- formers, while the Interior Department does approximately one-half of its R&D _ intra- murally. The Department of Transportation did not come into existence until 1966, but its attention to problems of ground and air transportation 75 Figure 62 Federal Obligations for Intramural R&D Performance, FY 1961-72 Total (Billions of Dollars) 5 Current dollars Constant 1958 dollars ‘) 5 ae a ee es ee ee es (est.) As a share of Federal and national R&D totals (Percent) 30 24 — Share of Federal total Se a ey va qe Share of National total | SE ee ee eee ee 1961 ‘62 '63 ‘64 ‘65 ‘66 ‘67 '68 '69 "70 ‘71 "72 Fiscal year (a) GNP price deflator was used to convert current to constant dollars. SOURCE: National Science Foundation. 76 Figure 63 Federal Obligations for Intramural R&D Performances, by Agency, FY 1961-72 (Billions of Dollars) 3 DoD NASA fee ee eS ES ele (Millions of Dollars) 400 . HEW ° A) 300 — ° e : : ° jb oor? : ” +——_19/0—_> R&D scientists and engineers per 1,000 employees (Number) 30 oe 20 ae 10 = Companies with— Less than 1,0u0to 5,000to 10,000 or tt to 10,000 or 1,000 4,999 9,999 more 999 more Employees SOURCE: National Science Foundation. —_— to a lesser extent, to the increasing R&D in industries other than the five noted above. Industrial R&D in these manufacturing indus- tries is also heavily concentrated in a relatively small number of companies. The four com- panies having the largest R&D investment spent 18 percent of all industrial R&D funds in 1970; the largest 20 spent 55 percent; and the largest 100 spent 79 percent. Some 300 companies spent 91 percent of all funds for industrial R&D. This pattern changed little over the past decades. This concentration, however, partially reflects the fact that certain industries, particularly some of those which are most R&D intensive (e.g., air- craft and missiles), are largely comprised of a relatively small number of large companies. R&D in such industries tends, perforce, to be concentrated in these large companies. 81 A Delphi Experiment A Delphi Experiment @ Various aspects of the — scientific-tech- nological enterprise and external conditions that influence its capabilities and performance are not amenable to purely quantitative treatment. Some of these were explored through a public opinion survey, the results of which are summarized in the following section of this report. Others involve considerations of a pre- dominately scientific or technical nature. Several of the latter aspects and conditions were investi- gated on an experimental basis, using a Delphi technique to solicit and synthesize the judg- ments and opinions of a cross section of the scientific and technological community. The study was carried out over the period of July- August 1972. The topics explored in this experimental effort were: Panel 1—The future role of science and tech- nology in areas of high public concern; Panel 2—Impacts of recent R&D _ funding changes on science and technology; Panel 3—Technologial innovation — including current impediments and measures for enhancement; Panel 4—Basic research including criteria for support and means for improving its effectiveness; Panel 5—Allocation of financial resources among fields of scientific research; and Panel 6—Future directions for graduate educa- tion in science and engineering. Participants in the Delphi exercise, who are listed in Appendix B, were selected for their extensive experience and knowledge in science and technology and the interaction of the two with society. Panels, ranging in size from 10 to 42, were composed of participants encompassing the disciplines, experience, and institutions rele- vant to the specific topics. Panelists represented a broad spectrum of disciplines (physical, life, and social sciences and engineering); experience (management, research, teaching); and institu- tions (colleges and universities, foundations, government, and industry). The Delphi technique used in this experiment solicited the judgments of the participants 84 through a relatively structured set of questions organized into two rounds. In responding to the first round of questions each participant was invited to suggest additional aspects and ques- tions either to attain greater detail or to expand the scope of the topic. The second round incor- porated these suggestions, provided feedback to each participant of both his first-round responses and those of the panel as a whole, and extended the questions. Participants in the second round responded again to the first-round questions—altering their initial responses if de- sired, in light of the group responses and sugges- tions—as well as to the questions added between rounds one and two. The second-round responses to panels reported hereafter are aggregates of the individual responses, with each participant contributing equally to the collective judgment. The Delphi methodology used in_ this experiment had both merits and shortcomings for the purposes of this effort. On the positive side, it proved to be a relatively efficient means for obtaining the collective judgment of a large number of respondents, under conditions which encouraged the expression of individual view- points free from the pressures of face-to-face encounters. As used here, however, the tech- nique had some serious weaknesses: the size of some of the panels may have been too small to represent the variety of viewpoints associated with some topics; and the controlled and limited nature of the inquiry may have resulted in misinterpretation of certain questions, as well as difficulties in responding to them. Posing questions about inherently complex and subtle issues in the most appropriate way was often the most difficult, and least successful, aspect of the experiment. Finally, it should be noted that panelists responded to three different kinds of questions: questions soliciting cause or effect interpretations; questions soliciting predictions; and questions soliciting recommendations about possible future policies. Although expert judgment is involved in each case, interpretations and predictions differ from recommendations in that the latter involve normative considerations to a greater extent. Furthermore, in the absence of com- parable previous studies the results should be interpreted with caution. They are presented in this report in the same spirit as that in which they were obtained: as experimental findings. The judgments of the panelists are sum- marized in the following pages. Only those judg- ments expressed by the panels as a whole are presented. This omits the many interesting com- ments and suggestions provided by individual participants but which were not reviewed by the entire panel. DELPHI TOPICS Panel 1—42 Panelists National Problems Warranting Greater R&D Panelists judged the extent to which science and technology could help ameliorate several problems of high public concern and identified the areas in which expanded R&D was warranted. These judgments are presented in table A. Although R&D was viewed as essential for alleviating many of the Nation’s problems, it was rarely regarded as sufficient in itself. The full effectiveness of science and technology was seen as dependent upon appropriate social, economic, and political policies. A number of problem areas identified by the Delphi panelists as warranting expanded R&D were the same areas as those chosen by the general public for applications of science and technology.! The areas in common which were most favored by the public were health care, pollution control, drug abuse, and crime. 1 See the following section, “Public Attitudes Toward Science and Technology.” Table A—National Problems Warranting Greater R&D Effort Problem area! Pollution (including water pollution, solid waste disposal, land pollution Areas which could benefit Areas warranting from science and major increases in R&D (percent of panelists) technology (percent of panelists) Power/energy resources (including greater conservation and more efficient use of power resources) Industrial productivity ......-.+--200see eee eee nese Adequacy of natural resources .......+++0+ee0eeeeees High cost and ineffectiveness of health services Deterioration of international economic of the United States Population growth .........-0-+-seee nsec eens Inappropriateness and expense of education Magnitude, quality, and delivery of information DY De TL Sercaomeane Jud Gn oe hos ep aR Tene DE Duos Inadequate urban planning Poverty Breakdown in efficiency and innovativeness of public sector services Inadequate employment opportunities .......... Nuclear war Urban crime Disrepect for established institutions Group conflict and alienation Irresolution of international conflict Changing values (sex mores, work ethics, etc.) .......--+ Racial discrimination Sacda oe OF, 92 cc dentcesns 97 86 Beaoe 81 70 SareS 80 82 WEES OBS 79 86 position enagevarsre 66 68 soreria 56 53 A Ae eb 50 62 Sahel 47 40 OSHS 46 70 Speen 45 63 ayeaessns 43 56 Pees 36 48 eBo cay 28 47 eos So eD 27 34 Becca 23 46 AS GmGe 8 17 Soeaic 8 33 Sak opti 8 32 orn he 3 12 3 31 1 Items in italics were suggested to panelists as examples; others were added by panelists and presented in the second round 85 Panel 2—16 Panelists Impacts of R&D Funding Changes Delphi panelists identified and assessed certain R&D funding changes which had occurred since 1965 in terms of their beneficial or detrimental impacts, and identified some of the major consequences—both positive and negative—of these changes. Funding changes, which a majority of panelists believed to be either beneficial or detrimental, are presented in table B. Table B—Funding Changes Assessed as Beneficial or Detrimental to R&D Percent of panelists Beneficial changes! assessing change as beneficial Increased spending for health related RGD GN oz. re ate, cect fe foie inhale ersy cvatetere 69 Increased support for social sciences ....... 67 Increasing percentage of total R&D funded by industry rather than Government .... 65 Percent of panelists Detrimental changes! assessing change as detrimental Decreased R&D funding (in constant-dollars)).. 212 ease ste yestiea 100 Frequent rapid changes of programs and directions of funding ........ 100 Decrease (in constant dollars) in basic research funds ........... 95 Allocation of funds by — student COUNT feiece rere rate oteua los ration srretreti ays 82 Closing down or significantly diminishing support for some large industrial RED laboratories: tactte nic cs eect ae 81 Decrease in R&D activities of small firms . 67 Reduction in rate of initiation of construction of “Big Science” facilities ............. 66 Increased allocation to R&D areas with short-term application ............... 52 1ltems in italics were suggested to panelists as examples; others were added by panelists and presented in the second round. Three changes in R&D funding were viewed as beneficial. Increased spending for health- related research and development was regarded as having been a positive trend because of the likelihood of improved health services for the general community, rather than for basic science 86 advances in the life sciences per se. Similarly, in- creased support for the social sciences were deemed beneficial largely because of the possibilities for developing a more scientific ap- proach to social problems; increased R&D fund- ing by the private sector was seen as probably re- sulting in greater emphasis on efforts which contribute to the economy in significant and immediate ways. Some panelists, however, warned that the latter change produced a shift in R&D toward short-term and low-risk efforts aimed at insignificant technological advances. Most of the changes regarded as detrimental relate to decreases in R&D funding. In respect to reduced R&D and basic research funding (in con- stant dollars), the many consequences sug- gested by individual panelists included: a de- crease or delay in developing new knowledge; the loss of scientific manpower and the reduc- tion of the number of future scientists and engi- neers; increased reliance on Federal administra- tors for program selection; the demise of signifi- cant basic research programs formerly supported by the Department of Defense; and in the long run, a loss of the U.S. leadership posi- tion in basic sciences. As in the case of funding decreases, all panelists judged frequent, rapid changes of pro- grams and directions of funding to be detri- mental because of the discontinuities intro- duced into research programs and the conse- quent waste of financial and human resources. The panelists also registered concern about criteria employed in allocating support for re- search. The group thought that allocation on the basis of “student count” (a funding change identified by the panelists) was detrimental to the scientific and technical enterprise since it results in an ever greater concentration of re- search in larger institutions. A small majority (52 percent) of the panelists believed that increased allocation to R&D areas with short-term applica- tion was harmful because of the resulting re- duced support and training in the basic sciences; a slower rate of development of new knowl- edge; the sacrifice of long-term needs; and in- creased support of second-rate projects masked as topical and relevant research. In contrast, 25 percent of the respondents thought that in- creased allocation to areas with short-term application was beneficial because of possibly earlier benefits to society in important need areas. The R&D efforts of industry were also a prime concern of the panelists, who thought that de- creased support for some large industrial R&D laboratories were detrimental to science and technology. The panelists related this funding change to decreases in productivity and in beneficial technological advance, and weaken- ing of the international trade position of the United States. The group (76 percent) also rated the effects of a decrease in R&D activities of small firms as detrimental. They warned against the loss of important innovative groups and the technical obsolescence of small firms. However, 40 percent of the panelists believed the un- favorable consequences would be minimal be- cause significant advances require greater capabilities than are usually available to small firms. Finally, the panel registered mixed reactions to the reduced rate of initiation of construction of “Big Science” facilities. The majority of panel- ists viewed the change as detrimental, and related it to a decrease in the rate of discovery in the long run, decline in the rate of research activity, and loss of international scientific leadership. Panel 3—14 Panelists Changes Needed to Improve Technological Innovation and Diffusion This examination attempts to identify major impediments to technological innovation and diffusion in this country, and actions which might be taken to improve the situation. The impediments of both types suggested by the panel are presented in table C. The major impediments to technological innovation which the group identified can be grouped into three areas related to (a) incen- tives and government policies, (b) industrial management, and (c) research activities and manpower. Some of the suggested factors, how- ever, were regarded as beneficial by several of the panelists, including patent policies which inhibit innovation (25 percent); lack of a govern- ment agency to support the introduction of tech- nological innovation (18 percent); and antitrust laws which prohibit cooperative ventures (17 percent). Major impediments to the diffusion of innova- tions can be grouped into (a) lack of financial incentives and characteristics of the market for innovative products and services, (b) policies and practices of industry itself, and (c) human re- sources. Several panelists evaluated the sug- gested factors as actually beneficial including: oversized, overorganized industry (22 percent); lack of a government agency to support the introduction of technological innovation (18 per- cent); and antitrust laws which prohibit coopera- tive ventures (17 percent). Panel 4—12 Panelists Adequacy of Current Basic Research Efforts This inquiry assessed the appropriateness of the total funding level for basic research in terms of: (a) Criteria considered to be important in determining an appropriate funding level for basic research, and _ the appropriateness of the current level on the basis of these criteria, (b) Propensity for “risk-taking” in poten- tially “high-payoff” areas of research, an (c) Factors perceived to be impeding basic research in general. Criteria for Support of Basic Research The panelists emphasized the significance of both external factors and the opportunities and needs of the scientific enterprise itself, in determining an appropriate funding level for basic research (table D). Using these criteria, they suggested moderate increases, ranging from 2 to 12 percent, over the 1972 funding levels. The average of the proposed increases would raise the total funding in current dollars to $4.4 billion in 1973. High Risk-High Payoff Research The panelists also evaluated the adequacy of basic research efforts by assessing the propen- sity to undertake “high risk-high payoff” basic research. “High risk-high payoff” basic research refers to projects which may have a low probability of producing results and yet promise results, if achieved, of such significance that the projects are deemed worth the risk. An example 87 88 Table C—Factors Impeding Technological Innovation and Diffusion Impeding Factors! Percentage of panelists rating factor as impediment to: Innovation Incentives and Government Policies Inadequate profit from innovative work to help social sector PVODLENIS: -% staa; conieecene Spee IR ican Eee eee tei sees eke 100 Disaggregated and diffused market in several sectors needing innovation (local government, housing, heatlh services, etc.) .... 100 Inability to obtain venture capital funds for high-nisksinmovations? ereterc oieleisersteterel clots iey-"- c 385 46:6) 157 OV G:b meee eats S Department of Agriculture... 225 272 303 347 391 514 548 623 550 598 632 641.1 66.7 Department of Health, Educa- tion, and Welfare ....... 17/5 24 29:2" 3618) 4101) “42)3 54S Se56s/aee 56.9) 368: 9m can laa 4727, 52.6 Department of the Inferiok=sc 207) | 2075 22:6) 2 24 2752916 Sorenson 33:02 0 eed 32.9 Department of Commerce .. 8.5 10.4 14.3 17.7 18:9. MO SSeS oem OS 20/6 en 26m 2419 28.7 ‘GNP price deflator was used to convert current to constant dollars. Source: National Science Foundation, Federal Funds for Research, Development, and Other Scientific Activities, Vol. XX| (NSF 72-317). Table 33. Industrial basic research expenditures, by source, 1960-72 [Dollars in millions] Federal Total Industry Government Constant Constant Constant Current 1958 Current 1958 Current 1958 Year dollars dollars’ dollars dollars' dollars dollars’ NOGOMR Mevcvcieetare $376 $364 $297 $287 $ 79 $ 76 OGTR ene seta 395 377 314 300 81 77 TOG ZS ren ee, 488 461 345 326 143 135 TOGO yee aes eee 522 487 375 350 147 137 UC Tey ROR aon eackee 549 504 384 353 165 152 (SGS i osacnsee aes 592 534 406 366 186 168 MOGGERS 624 548 451 396 173 152 NOS TRe ee 629 535 427 363 202 172 1OGB Aes 642 525 462 378 180 147 Ghose be RRS Sc 618 482 458 357 160 125 1970S ee 629 465 470 347 159 117 1971 (prelim.) ... 625 441 455 321 170 120 T9O72(EStA) eee 660 453 480 329 180 123 ' GNP price deflator was used to convert current to constant dollars. Source: National 122 Science Foundation, National Patterns of R&D Resources, 1953-73 (NSF 73-303). Table 34. Industrial basic research expenditures, by field of science, 1967-70 (Dollars in millions} Table 36. Natural scientists and engineers in relation to total civilian employment,' 1950-70 Field of science Engineering) ......:. Ghemistryseavewcescucs = Physics and astronomy ........ Life sciences ....... Mathematics ........ Environmental SCIENCES) Ss .tk.cis!ors a Engineering .......- Chemistry .......... Physics and astronomy ........ Life sciences ....... Mathematics ........ Environmental SCIENCES= = sone 1967 1968 1969 1970 Current dollars 172 181 171 174 162 191 213 200 146 126 110 97 69 76 74 80 12 14 13 14 14 11 12 8 Scientists and Scientists and Total civilian engineers per Constant 1958 dollars' 146 148 133 129 138 156 166 148 124 103 86 72 59 62 58 59 10 11 10 10 12 9 9 6 ‘GNP price deflator was used to convert current to constant dollars Source: National Science Foundation, Research and Development in Industry, 1970 (NSF 72-309). engineers employment 10,000 civilian Year (in thousands) (in thousands) employees 1950 556.7 58,920 94 1951 611.8 59,962 102 1952 685.9 60,254 114 1953 748.7 61,181 122 1954 783.7 60,110 130 1955 812.6 62,171 131 1956 873.7 63,802 137 1957 958.9 64,071 150 1958 1,001.2 63,036 159 1959 1,057.9 64,630 164 1960 1,104.0 65,778 168 1961 1,151.5 65,746 175 1962 1,210.3 66,702 181 1963 1,280.8 67,762 189 1964 1,327.0 69,305 191 1965 1,366.9 71,088 192 1966 1,417.5 72,895 194 1967 1,476.7 74,372 199 1968 1,525.0 75,920 201 1969 1,567.7 77,902 201 1970 1,594.7 78,627 203 116 years of age or over Source Estimates prepared by the National Science Foundation based on data collected by the Foundation and the Bureau of Labor Statistics, Department of Labor Table 35. Total scientists and engineers, by broad field, 1960 and 1970 Total Ph.D.'s Field 1960 1970 1960 1970 Total Aer ore ain 1,167,000 1,731,000 89,200 171,800 Engineers: 2esentspreteccrecnceace 797,000 1,100,000 7,500 25,700 Physical/scientistseenaees -- se 168,000 253,000 30,400 55,000 Life:scientistsia eee 98,000 180,000 24,700 42,000 Socialiscientistss -sseeeerrenrae 70,000 126,000 22,600 38,400 Mathematicianss=eseeeeereee ee” 34,000 76,000 4,000 10,700 Source: Estimates prepared by the National Science Foundation based on data collected by the Foundation and the Bureau of Labor Statistics, Department of Labor 123 Table 37. Natural scientists and engineers in R&D, 1960-70 R&D scientists Relative change as percent of total (1960=100) Universities Universities Year Total U.S. Industry andcolleges Total U.S. Industry and colleges 1S6O0) ser 35.0 34.7 42.3 100.0 100.0 100.0 NOGA), cateesrsrc 35.6 35.2 42.8 106.1 105.2 107.8 19627 aemict 36.5 36.3 42.2 114.5 113.0 113.6 1963) fo. = 37.1 37.1 41.9 123.2 121.7 122.0 W964 oe 37.5 37.1 39.5 129.0 124.5 128.6 1OCSmee eee 37.5 37.2 38.4 132.9 127.6 132.7 1966 ..... 37.1 36.8 36.8 136.3 131.1 135.0 IN\s¥/ ee aae 37.5 36.8 SI) 7/ 143.5 136.2 158.6 1968'~>.... 36.2 36.1 36.6 143.2 137.0 154.9 1969 ener 35.0 35.2 SISR// 142.2 13725 150.4 NOVO) tec 33.6 33.5 33.2 138.7 131.9 155.8 Source: Estimates prepared by the National Science Foundation based on data collected by the Foundation and the Bureau of Labor Statistics, Department of Labor Table 38. Distribution of non-Ph.D. and Ph.D. scientists and engineers, by activity and broad field, 1970 Non-Ph.D.'s Ph.D.’s Field Total R&D Other Total R&D Other NOtal esccior. siococmeno ee 100.0 31.3 68.7 100.0 52.6 47.4 Physical scientists .......... 100.0 38.9 61.1 100.0 66.5 33.5 Liferscientistsiaacree neces sc 100.0 Shlar/ 68.2 100.0 56.7 43.3 Mathematicians ............. 100.0 33.5 66.5 100.0 27.1 72.9 SocialiscientistS #--- eee soe 100.0 31.7 68.3 100.0 30.7 69.3 Engimeersisicccricncrsctsrasrners 100.0 29.7 70.3 100.0 59.1 40.9 Source: Estimates prepared by the National Science Foundation based on data collected by the Foundation and the Bureau of Labor Statistics, Department of Labor Table 39. Employment of natural scientists and engineers, by sector, 1960 and 1970 Percent distribution Percent increase Sector 1960 1970 1960-70 A Role: | ere rctemane ter cen Sen aon ION 100.0 100.0 44.4 RrivatevindUsinyi cen. sms see mreece niet: 73.5 69.7 36.8 Federal Government .............-..-. 15:3) 15.0 41.8 Universities and colleges ............. 10.4 14.3 98.8 Nonprofit institutions... «ctesenie 8 1.0 88.5 Source: Estimates prepared by the National Science Foundation based on data collected by the Foundation and the Bureau of Labor Statistics, Department of Labor. 124 Table 40. Public secondary school enrollment in selected science and mathematics courses and total enrollment in grades 9 through 12, 1960-61 and 1969-70 Estimated enrollment (thousands) Index (1948-49=100) Course 1948-49 1960-61 1969-70 1948-49 1960-61 1969-70 BiQlOgVA Sete reese. 1,062 1,853 3,197 100 175 301 CHEMISTY eae as ce 412 745 1,160 100 181 282 PHYSICSIASe as es coe 291 402 482 100 138 166 ECONOMICS eset cre sees 255 293 844 100 115 331 SOCIOIOGV: eee See ser os 186 289 495 100 155 266 RSYChHOlOGWieees cs areas 47 140 344 100 298 732 Introductory algebra ........ 1,042 1,607 2,627 100 154 252 Introductory geometry ...... 599 960 1,530 100 160 255 Advanced mathematics ...... 609 1,174 1,756 100 193 288 Total enrollment, Gradesi9=12.ce esas 5,399 8,219 12,442 100 152 230 Source: National Science Foundation, Science Resources Studies Highlights, “Enrollment Increase in Science and Mathematics in Public Secondary Schools, 1948-49 to 1969-70" (NSF 71-30), Oct. 15, 1971. Table 41. Percent change in majors declared by junior-year students, 1970 to 1971 Percent Major field Fall 1970 Fall 1971 change PIYSICS 2 Sosa SOUR AS ER TASTAL/ 6,759 —8.4 ENGineCeninG strc cernmicc Sec cients 66,421 61,575 ied Chemisthy saree re cso Soo aces 13,949 13,646 —2.2 Mathematical sciences .................. 34,800 34,581 —.6 Basic*soclaliSGlenGeSavc-scscms ice oes 156,446 170,388 8.9 “All other” life sciences ................. 50,212 56,896 13.3 “Other” physical sciences ............... 8,302 9,556 15.1 Preprofessional life sciences ............. 11,303 14,326 26.7 Applied social sciences ................. 16,375 23,552 43.8 Source: ACE Higher Education Panel, Survey of Enrollment of Junior-Year Students (Fall 1970 and Fall 1971). 125 Table 42. Percent distribution of full-time graduate students in doctorate departments, by area of science and type of support, 1967-71' Mathe- All Engineer- Physical matical Life Psychol- Social Type of major support? areas ing sciences sciences sciences ogy sciences MLOtAl YS Ae. as.b a herent yer 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Fellowships and traineeships: MOGI ees seek mo ee eee 32.4 Silk 27.3 26.0 35.9 42.3 36.5 HOGS) Oc cciarcsaricer a ocketoaresoprercrsmuctet rays 32.1 29.7 26.4 26.6 36.7 41.9 36.2 AQGO react ete ck hie aaron 29.7 26.2 23.4 23.7 35.7 41.6 33.4 NOPOI ce kettians partes hice ena 27.6 23.6 20.8 20.6 33.0 40.6 32.6 VOM Se sa cas as i seeerecetrs ayes pereiee 25.4 22.0 18.5 18.8 30.2 35.0 29.3 Research assistantships LIC sVGeies COP Ean aGoACnardS 22.9 29.0 31.4 9.3 25.4 16.6 11.0 hc lots leer eso Rees ceeea OO etc 22.1 29.7 30.3 8.5 24.0 15.1 10.5 NOES nsec aiaschemssine os ee ceva 21.6 29.3 31.6 9.1 22.2 14.5 9.5 MOTO cic pee tert ohvers pateisren les 21.7 30.0 31.3 9.6 22.2 14.4 9:5 NOTA ces cosnevera ant cteyrense cues aeenehces 20.9 29.4 30.1 9.1 22.5 13.6 9.4 Teaching assistantships I ley freee en ena om toner Gra 23.0 12.7 Siiv, 40.5 21.4 18.7 18.6 TO GBs ee Seyret sete renter ter ete 23.5 13.2 32.9 41.8 21.6 19.1 18.4 NQGOS as Scccceuis lair anse wns incnaiare tases chatter 24.0 13.8 33.8 42.3 22.8 19.3 18.9 MOOS See Nera pe ke ate trary lease 25.2 14.0 36.5 46.2 23.5 20.0 19.6 TCV URS ISA eterna Se rete OEM MC iehe 6 24.7 14.1 37.4 44.1 23.0 20.4 19.5 Other types of support? NOG Tie rsir ssa tepe ses ageastencheneceenreteraiee 21.7 27.1 9.6 24.2 Wes 22.5 33.9 WOGBE Ase Socios As Sines omerce One 22.4 27.4 10.5 23.0 liveth 24.0 34.9 NOGQ Pree Sache seas eevee ieee oer 24.7 30.7 11.2 24.9 19.3 24.6 38.2 NOT OWE ee, scart omatiele reat rees 25.5 32.4 11.3 23.5 2lkS 25.1 38.3 ETAL SSS oreo mires bmn ara Re 29.0 34.5 14.0 28.1 24.2 31.0 41.8 ‘Based on data submitted by 2,236 doctorate departments submitting traineeship applications in each of the years 1967-70 and 2,990 doctorate departments reporting for 1971 2 Major support is defined as a total stipend of $1,200 or more, exclusive of tuition, during an academic year. 3 Includes principally family or self-support Source: Special tabulations from National Science Foundation surveys of graduate student support. 126 Table 43. Distribution of full-time graduate students in science and engineering, by source of support, 1969-71 Type of major support 1969 1970 1971 Motalafull-timeStuGentSiems cess meres serene 131,923 131,902 129,939 WES S GOVERMMEM bie secrauccpcciie cise Gin cus aowreaeare seeks 48,373 45,640 41,263 Institutionalisuppontercia asco. vacates 47,415 48,915 48,298 SElf=SUPPONteeeeweiscecicrsje cere soe cots eratsievens oye eee 24,123 25,155 28,801 Other sourceswemee ices Sekaan. sy actaaes naereicene 11,994 12,192 Tale Sie: Fellowships and traineeships .................. 38,972 36,453 32,988 WESHGOVERNIMEMNS ese ive ste aisle ste Seemann 26,671 24,070 20,959 Institution algSUpPOllass Je sktsc:s stents nei 6,777 6,740 6,628 SelfESUPPO Msc ois karan ere ee N.A. N.A. NLA. OthemnsoOurcesverseser eco Sch cu cee 5,524 5,643 5,401 Researchsassistantsmips: sccccc-ca-) eee 28,506 28,500 27,249 WISSGoVernmentacac cscs. ss ace eee 18,641 18,451 17,519 InstitutionalisupPOMme ss Se aS aes eee 7,700 8,034 7,588 Self-SUPPOMERERe Gascin ones. Sos ee —0— —0— == OMhERSOUNCES Beane OS Oe SR tae 2,165 2,015 2,142 sheachingiaSsistantShipS: csc nese sone eis oe 31,221 32,616 32,335 WISEGovermmentese ccs see ees Seen 295 336 384 InStiLUtIonNAalkSUPPONIa aan ccse sacs: Sele eee 30,790 32,015 31,729 SElfESUPPONTRE Sais Sar te as usa 0 0 0 Othersounceswacsees sores ea 136 265 222 Source: Special tabulations from National Science Foundation surveys of graduate student support Table 44a. Bachelor's and first-professional degrees in science and engineering, 1959-60 to 1970-71 1959- 1960- 1961- 1962- 1963- 1964- 1965- 1966- 1967- 1968- 1969- 1970- Field of science 60 61 62 63 64 65 66 67 68 69 70 71 All sciences and engineering .. 120,937 121,660 127,469 135,964 153,361 164,936 173,471 187,849 212,174 244519 264,122 271,176 Physical sciences ... 16,057 15,500 15,894 16276 17,527 17,916 17,186 17,794 19442 21,591 21,551 21,549 Engineering ........ 37,808 35,866 34,735 33,458 35,226 36,795 35,815 36,188 37,614 41,553 44772 45,387 Mathematical SCIENCESIe ane 11,437 13127 14610 16128 18,677 19,668 20,182 21,530 24084 28263 29,109 27,306 Life sciences ....... 24,141 23,900 25,200 27,801 31,611 34842 36,964 39408 43260 48713 52129 52640 Social sciences ..... 31,494 33,267 37,030 42,308 50,320 55,715 63,424 72929 87,774 104399 116,561 124,284 Source: U.S. Office of Education, Earned Degrees Conferred, annual series Table 44b. Bachelor’s and first-professional degrees in science and engineering as a percent of all bachelor’s and first-professional degrees, 1959-60 to 1970-71 1959- 1960- 1961- 1962. 1963- 1964- 1965- 1966- 1967- 1968- 1969- 1970- Field of science 60 61 62 63 64 65 66 67 68 69 7U 71 All sciences and engineering .. 30.6 30.3 30.3 30.2 30.5 30.6 Sle2 31.6 31.6 31.8 317 30.7 Physical sciences ... 41 348) 3.8 3.6 3.5 3.3 3.1 3.0 29 2.8 2.6 24 Engineering ........ 9.6 89 8.3 74 70 6.8 6.4 6.1 5.6 5.4 5.4 5.1 Mathematical SCIENCES): 6... a4. 29 3.3 3:5 3.6 37 3.6 3.6 3.6 3.6 37 3.5 3.1 Life sciences ....... 6.1 5.9 6.0 6.2 6.3 6.5 6.6 6.6 6.4 63 6.3 6.0 Social sciences ..... 8.0 8.3 88 94 10.0 10.3 114 12.3 13.1 13.6 14.0 14.1 Source: U.S. Office of Education, Earned Degrees Conferred, annual series. 127 Table 45a. Master’s degrees in science and engineering, 1959-60 to 1970-71 1959- 1960- 1961- 1962- 1963- 1964- 1965- 1966- 1967- 1968- 1969- 1970- Field of science 60 61 62 63 64 65 66 67 68 69 70 71 All sciences and engineering 20,012 22,786 25146 27,367 30,271 33835 38083 41,800 45.425 48425 49318 50,624 Physical sciences 3,387 3,799 3,929 4132 4567 4918 4992 5412 5,508 5,911 5,948 6,386 Engineering . 7,159 8,178 8,909 9635 10,827 12056 13678 13,885 15188 15243 15597 16,347 Mathematical sciences ... 1,765 2,238 2,680 3,323 3,603 4,294 5,610 5,733 6,081 6,735 7,107 6,789 Life sciences ... 3,751 4,085 4,672 4718 5,357 5,978 6,666 7,465 8,315 8,809 8590 9,738 Social sciences 3,950 4486 4,956 5,559 5,917 6,589 7,137 9305 10,333 11,727 12,076 12,364 Source: U.S. Office of Education, Earned Degrees Conferred, annual series. Table 45b. Master’s degrees in science and engineering as a percent of all master’s degrees, 1959-60 to 1970-71 1959- 1960- 1961- 1962- 1963- 1964- 1965- 1966- 1967- 1968- 1969- 1970- Field of science 60 61 62 63 64 65 66 67 68 69 70 71 All sciences and engineering . 26.9 25u 29.6 29.9 29.9 30.2 27.1 26.5 25.6 24.9 23.6 21.9 Physical sciences 45 49 46 45 45 44 3.5 3.4- 3.1 3.0 28 2.8 Engineering 9.6 10.4 10.5 10.5 10.7 10.7 97 8.8 8.6 78 74 71 Mathematical SCIENCES se). s aes 2.4 29 3.2 3.6 3.6 3.8 3.6 3.6 3.4 3.5 3.4 72.) Life sciences ....... 5.0 By 5.5 Ly 5.3 5.3 47 47 47 45 41 3.8 Social sciences ..... 5.3 GH 5.8 6.1 59 5.9 5.5 59 5.8 6.0 5.8 5.3 Source: US. Office of Education, Earned Degrees Conferred, annual series Table 46a. Doctor's degrees in science and engineering, 1959-60 to 1970-71 1959- 1960- 1961- 1962- 1963- 1964- 1965- 1966- 1967- 1968- —1969- 1970- Field of science 60 61 62 63 64 65 66 67 68 69 70 71 All sciences and engineering .. 6,056 6,531 7,249 8,055 9,025 10,252 11,298 12,759 14128 15839 17,639 18,466 Physical sciences ... 1,838 1,991 2,122 2380) 2,495) 22829 310451 3/4628 931593 3,859 4313 4,391 Engineering ........ 786 943 1,207 1,378 1693 meer2124 2,304 2614 2.932 3,377 3,681 3,654 Mathematical SCIENCES asses 303 344 396 490 596 688 801 870 983 1,161 1,343 1,327 Life sciences ....... 1,647 1,646 1,804 1,908 2,181 2,474 2696 2,900 3,445 3,779 = 4,131 4,746 Social sciences ..... 1,482 1,607 1,720 1899 ne2 100 Mer 2 137 452) 2,913 3,175 3,663 4,171 4,348 Source: U.S. Office of Education, Earned Degrees Conferred, annual series Table 46b. Doctor's degrees in science and engineering as a percent of all doctor's degrees, 1959-60 to 1970-71 1959- 1960- 1961- 1962- 1963- = 1964- 1965- 1966- 1967- 1968- 1969- 1970- Field of science 60 61 62 63 64 65 66 67 68 69 70 71 All sciences and engineering .. 61.6 61.8 62.4 62.8 62.3 62.3 61.9 61.9 61.2 60.5 59.0 57.5 Physical sciences ... 18.7 18.8 18.3 18.6 16.9 17.2 16.7 16.8 15.6 147 144 13.7 Engineering ........ 8.0 8.9 10.4 10.7 117 12.9 12.6 12.7 12.7 12.9 12.3 11.4 Mathematical SCIENCES 23 <-aee 3.1 3.3 3.4 3.8 41 42 44 42 43 44 45 4.1 Life sciences ....... 16.8 15.6 15.5 14.9 15.1 15.0 14.8 14.1 14.9 144 13.8 14.8 Social sciences .... 15.1 15.2 14.8 14.8 145 13.0 13.4 14.1 13.7 14.0 14.0 13.5 Source: 128 U.S. Office of Education, Earned Degrees Conferred, annual series. Table 47. Geographic origins, by high school graduation, of Ph.D.’s in science and technology, 1970 Ph.D.’s in science and Population engineering by State of (thousands) high school graduation Region, division, and April 1, 1960 = April 1, 1970 State (census) (census) FY 1960 FY 1970 United States, Total ..... 179,323 203,166 5,128 14,272 NORTHEAST New England ........... 10,509 11,842 436 968 Maineviecmerctoctsecce.a: 969 992 29 39 New Hampshire ....... 607 738 33 67 Vernmontiieigace = se 390 444 10 23 Massachusetts ........ 5,149 5,689 240 543 Rhode Island ......... 859 947 35 73 Connecticut .......... 2,535 3,032 89 223 Middle Atlantic ......... 34,168 37,153 1,499 3,641 N@WiYOnKsacak elect ae 16,782 18,191 951 2,058 New Jersey ........... 6,067 7,168 231 586 Pennsylvania ......... 11,319 11,794 317 997 NORTH CENTRAL East North Central ...... 36,225 40,252 1,066 2,862 (Olney sos Arch iene 9,706 10,652 253 730 Indiana vere sisters see 4,662 5,194 102 320 [intGiseesssces: fetes Se 10,081 11,114 403 915 Michiganwsech aacss..2- 7,823 8,875 168 519 WiISCONSINGwss a: 26. coe 3,952 4,418 150 378 West North Central ..... 15,394 16,319 493 1,415 Minnesotai- 525.2 oon. 3,414 3,805 115 301 IOWA SerscisSaiece 2,758 2,824 98 293 MiSSOUNIi seeks 4,320 4,677 118 349 North Dakota ......... 632 618 16 49 South Dakota ......... 681 666 23 70 Nebraska ............. 1,411 1,483 50 123 KanSaSvee acceso cicnie cers 2,179 2,247 8) 230 SOUTH southyAtlanticgga-.....<- 25,972 30,671 402 1,399 Delaware sansa eae 446 548 10 36 Marylanditere ses. ofan 3,101 3,922 59 243 District of Columbia ... 764 757 36 Tie Mirgimiausayere sites 3,967 4,148 74 261 West Virginia ......... 1,860 1,744 49 111 North Carolina ........ 4,556 5,082 43 186 South Carolina ....... 2,383 2,591 27 79 Georgiainsascice nem 3,943 4,590 42 156 Floridac.nc-tesee ec 4,952 6,789 62 250 East South Central ...... 12,050 12,803 176 602 KentuckyGenssceeereeriee 3,038 3,219 50 143 Tennessee ............ 3,567 3,924 46 205 Alabamaiics cit eecancee 3,267 3,444 46 145 Mississippi ........... 2,178 2,217 34 109 West South Central ..... 16,951 19,321 304 1,149 ArkansaS (sg-c%.. cache 1,786 1,923 42 81 Eouisianal at a : : x if a” +9 by i - hy se We " : a Fd 7 4g ; i f ¥ ae } SN f j. * SS - 5 4 BS) : ¢ i “eset i q : if Ly 4 : 3 4 a me =f at x { ¥ 3 indy m % ? 2 Z : t ~ ers - ae id ’ Be gs ea : : 2 é st i x 4 * ‘ < eS bey .y phy 2 Se eer S: a 4 + ; rt ‘ : ‘ 4 AO ‘ G . * j E Sg er * Ue ~ > : ~ \ 3 ‘4 \ : oy v x ; 8 . : ee oe | ena ss 4 t . = ‘ \ j \ e * i ¥ ‘ x Te x S ‘: " \ ad x § a P x > 2 5 x 4 a