Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
INTRODUCTION Overview Why are there so few women scientists? Why do we so rarely hear of their work? What happens to the over two thousand women who annually earn doctorates in science? What jobs do they get? Do they get jobs? Dead end jobs, or those with a future? Do they get equal pay for equal work? Have equal opportunity mandates changed their situation? Do they benefit as men do from public awareness of their work? Do they have similar opportunities to serve in policy advisory bodies? What is the outlook for their future? We chose to focus this first report on academic employment, primarily in faculty positions, both because existing data are more extensive than for other sectors and because educational institutions are the prime employers of doctoral scientists. In addition, the hierarchy of ranks and institutions is well-defined and makes it possible to compare how men and women fare in professional terms to a degree that is not readily matched in industry, for example. Beyond these pragmatic considerations, however, faculty status represents the quintessential scientific career. Ideally it provides total freedom of inquiry, insured by a degree of personal security unmatched in any other walk of life except the Civil Service. In practice, freedom of inquiry may be somewhat curbed by the availability of money and more recently by certain external regulations. Still, to many young scientists a tenured faculty post in a research university remains the most desirable career goal. How many women reach it? Scientific manpower has been the subject of many analyses since World War II; scientific womanpowerâabout one-tenth of the scientific doctoral labor forceâhas received little attention until recently, when equal opportunity legislation required employers to perform utilization analyses of their labor force. Data on doctorate production by sex and field have been published since 1920r but detailed employment information on a national sample of all science and engineering Ph.D's has been available only since 1973. While some analyses of women scientists' employment for individual disciplines have
appeared in the last few years, no systematic studies encompassing all science fields have been done. This Committee owes its existence to the pressures arising from the women's movement, specifically in academe, during the late sixties. A small conference was convened by the National Academy of Sciences in 1972 to begin exploration of women's status in science, followed by a research conference in 1974. Subsequently, this Committee was appointed. Concurrently, demands from employers, particularly academic institutions, for better information regarding women scientists arose in relation to their affirmative action obligations. Scope of the Study To assess whether and to what extent earlier patterns of faculty appointments have changed since the advent of affirmative action regulations, we will be examining extensive trend data on the production and employment of men and women doctorates. We hoped that it would be possible also to derive some insights that transcend statistical comparisons. The flow of scientists through graduate and postdoctoral training and into jobs, in academe or elsewhere, is subject to various influences not usually considered in affirmative action discussions. Training opportunities at both pre- and postdoctoral levels are highly dependent on research funding, which has been changing in the last decade, declining in real dollars and fluctuating widely among fields. The effects of the Vietnam War and the draft on science doctorates are almost impossible to assess. To what extent, if at all, did they reduce the numbers of new male doctorates or affect their quality? We have no data on this and can draw no conclusions. The decreasing enrollments in higher education are reducing the number of available appointments; some departments are contracting, and almost all are postponing tenure decisions as long as possible. Against such a background, what kind of hiring and promotion rates for women scientists can reasonably be expected? How do we interpret the changes we find? If no expansion is possible, what might the "good will efforts" which the law is willing to accept in lieu of actual numerical improvement encompass? What of the problems of obsolescence which are specific to science and not to other academic disciplines? A first- rate woman scientist trained a decade ago and unlikely to be considered at that time for a faculty post in a research university may have spent the intervening years teaching in a small college. Her ten-year-old qualifications do not fit her now for the position she should have had then. Can anything be done for her? Should something be done? Could her excellent capabilities, maturity, and experience in a
different sector of academic science be used to advantage in advisory functions? How many others like her are there? Is this a national problem? What are the restrictions faced by women who decide to interrupt or slow down their careers in order to have children? Are there employment options available that would utilize their talent on a rigorous but less than full-time basis? There is a continuing search by today's young men and women for ways to reconcile conflicting demands of their parental and career roles. While academic institutions cannot be charged with responsibility for either the problems or solutions that women face in this connection, they should avoid compounding the problem and should share responsiblity for exploring the development of solutions within the academic framework that would help meet the conflicts. For example, a quarter of a century ago New York Medical College took the pioneering step of offering part- time psychiatric residency for women physicians with young families. We have also been mindful of the compounded difficulties faced by minority women scientists; their problems are discussed fully in a conference report issued by the American Association for the Advancement of Science (1977). It is interesting to note here that minority women report more discrimination based on sex than on race. For the purposes of this report, we found minority women scientists to be too widely scattered through fields and departments to enable us to draw any general conclusions other than to deplore their absence. At the end of this report we consider the possible modifications or policy initiatives which could correct or ameliorate existing inequities. Among these are the creation of additional research opportunities, more innovative institutional contributions to solving dual- career problems, and expanded opportunites for service by women scientists in advisory functions. It is notoriously difficult, however, to devise remedial policies which do not in turn create some measure of disadvantage for innocent bystanders. Implicit in these questions is our assumption that men and women scientists are of comparable quality. Some scientists do not believe that assumption is justified, Lester (1974) among them. We explore that problem in Chapter 2, insofar as the usual proxy measures of ability can be applied. None of them really tells us much about research potential, or how we foretell the excellent from the merely very good. It is often assumed that women's careers must necessarily take a different path from men's because of
their different family responsibilities and constraints on mobility. If that is true, and given similar ratios of qualified candidates, unmarried women's career opportunities should be just like men's, and there should be no systematic differences in relative employment of women among fields unless some fields somehow impose greater demands on their practitioners than do others. As we begin this exploration, it must be stressed again that our concern is with the status of women scientists, rather than the situation of all women doctorates. Excellence in science, at least at advanced levels, can be fostered only in certain circumstances. It is far more dependent than other fields on concentrations of facilities and equipment and the presence of other workers in the same or related disciplines. The place where a potentially outstanding scientist finds employment, and the conditions of such employment, will therefore influence the eventual realization of that potential in significant ways. On the other hand, science demands aptitudes and preparation possibly more specialized or exacting than other disciplines so that we must examine the capacities of men and women scientists as they enter their professional careers. For this purpose, a comparison of all women doctorates with all men doctorates is inappropriate. Numerous comparisons of this type have yielded negative results for women with respect to measures of educational quality. In fact, these distinctions reflect the different field distributions of the two sexes. The fact that women doctorates as a group take longer than men to complete their degrees, for example, simply indexes the greater concentration of women in nonscience fields where both sexes customarily obtain their degrees after longer time periods. Our analysis, therefore, is limited to a comparison of men and women Ph.D.'s in the natural sciences, the social sciences, and engineering. It excludes those with doctorates in other fields as well as those with professional degrees in fields such as medicine but does include Ph.D.'s employed in medical or other professional schools. EQUAL OPPORTUNITY IN HIGHER EDUCATION In 1968, Executive Order 11246 extended the clause of the 1964 Civil Rights Act prohibiting discrimination in employment (Title VII) to cover institutions of higher education, which were previously excluded, and four years later the Office of Civil Rights issued its Guidelines for Affirmative Action in Higher Education to implement Title VII. Also in 1972, Title IX of the Higher Education Amendments specifically addressed the provision for
educational equity at all levels. Equal opportunity for study and employment in higher education regardless of sex is therefore a clearly stated goal of national policy; the explicit controversies surrounding the issue have dealt not with the desirability of achieving that equality but with the means for doing so. Since affirmative action guidelines require utilization analyses based on appropriate statistical information, there has been a natural tendency, for purposes of both general discussion and the establishment of legal evidence, to argue for or against the existence of discrimination on statistical grounds. Any discrepancy between the percentage of women or minorities qualified by training and experience to hold a given type of position and the percentage actually employed has been taken as legal evidence of discrimination, purposeful or not. (Relevant legislation and Executive Orders are summarized in Appendix A.) Several assumptions and issues are buried in the foregoing paragraph. The fact that equal opportunity laws came into being strongly suggests that equal opportunity had not existed previously. Was this a valid assumption? We think so. Until a decade ago, women were not admitted as undergraduates, and in some cases not as graduate students, to several highly selective universities which set the pace for academic science. Women who were admitted to graduate and professional schools sometimes had to meet higher standards. Numerous instances of more stringent criteria for the admission of women to selective graduate departments were cited by Harris (Harris, 1970). An illustrative case is that of the School of Veterinary Science, University of Pennsylvania, which until recently required a 3.6 GPA of women applicants and 2.6 of men (Davies, 1978). Major universities rarely appointed women to their science faculties (although the pool of pre-1950 Ph.D.'s included about ten percent women). Women were more likely to be employed by colleges and non-research universities and to be cancentrated in the lower ranks or as research staff. Moreover, they were not paid as much as similarly trained men at the same rank. For example, in 1973, a woman full professor was typically paid 15 to 20 percent less than a man in three major fieldsâchemistry, biology, and social sciences.* Was all this a result of discrimination? Some spokesmen for the universities have argued otherwise, suggesting that women preferred less demanding occupations in order to fulfill their family obligations, that restricted mobility made a normal academic career almost impossible for them, and that they were paid all they were worth. (See Lester, 1974, for an extended exposition of this point of view.)
These arguments in turn rest on assumptions about the nature of academic careers and the relative abilities of women and men, and the likely responses of the two groups to potential conflicts between professional demands and family obligations. The commitment in time, energy, and dedication required of a tenured faculty member is large, and may in fact conflict with many other desired activities for men as well as women. The degree of flexibility a tenured appointment permits is also very high, however, and this might be more important in accommodating other obligations than the high total level of effort. In any case, these considerations are speculative; very little is known about how scientists make career choices at this level, or how they assess personal costs against potential professional benefits. Instances of either male or female scientists refusing academic appointments solely because they are too demanding appear to be rare at best. The inference that women commonly do so while men do not is unsupported. Restrictions on geographic mobility for career development pose a different sort of problem. Under present social circumstances, most women with families are probably in fact less mobile than men, although "commuter couples" are increasingly common in academe. A parallel flexibility on the part of universities in creating joint employment for such couples exists in a few cases but is not widespread. In any case we know little about the professional benefitsâ or costsâof high mobility, or indeed about its incidence among scientists. A year or two devoted to rebuilding a research group and reorganizing facilities following a professional move may represent a long-term loss of research productivity which actually overshadows the gains in professional opportunity or other benefits. Whether women faculty members are less likely than men to move for better opportunity, or whether they are less likely to have the opportunity to move, remain unresolved questions. Legal Definitions Regardless of the basis for limitations on the status of women in academe, equal opportunity laws do not distinguish intent from historical accident; they deal only with end results. If women or other "affected groups" appear in a given employee category in proportions lower than their representation in the appropriate availability pool they are assumed to be victims of discrimination in the first instance. At this writing, the burden of proving otherwise legally rests with the employer. Affirmative action policies, their execution, and the controversy surrounding them deserve futher comment here. As the regulations apply to faculty employment, they require equal opportunity to be considered for a job and selection
on merit criteria only, with the choice between two equally matched candidates to favor a woman or minority candidate. To ascertain whether their choices are indeed bias free, institutions are required to perform periodic utilization analyses and set goals and timetables for rectifying imbalances. However, relatively few universities have affirmative action plans which actually contain numerical goals, and they have enjoyed considerable latitude in setting those goals on the basis of their own internal staffing projections. More important, the penalties provided by lawâthe withholding of federal monies until an institution is in complianceâhave only been imposed on a token basis, i.e., for a period of a week or two until the institution agreed to come into compliance at some future time. The most important sanction which the law provides is a pre-award compliance review conducted on site for grants and contracts exceeding $1 million; after six years, it was applied for the first time in the last few months. At this writing, most of these reviews are incomplete and it is too early to judge their general effect. Any conclusions regarding the effectiveness of affirmative action policies unfortunately will be clouded by the widely acknowledged capriciousness of enforcement efforts, the sometimes highly localized interpretations of regulations by enforcement officers, and the frequent problems which have been generated by these actions. Unresolved Issues The questions we have posed, and addressed throughout this report, deal with the opportunities afforded to individual women scientists (although they must be framed in terms of groups of individuals). Another set of issues concerns the universities and the fabric of science itself. Can it be argued that the major universities have impoverished themselves by virtually excluding women from their faculties? Are science departments of lower quality than they would be had they hired more women? Will they be better if and when they do? Would there be more women science students if there were more women science faculty? Would that provide a welcome source of additional talent, or merely flood already overpopulated fields? Answers to such questions would remain speculative at best, and none are suggested, but readers should bear these issues in mind. Data Sources The tools at our disposal for examining these issues are the extensive and detailed data collected by the National
Research Council in its Surveys of Earned Doctorates and Comprehensive Surveys of Doctorate Recipients.2 These are the only available longitudinal data that encompass Ph.D.'s in all science and engineering fields. The Survey of Earned Doctorates, an annual survey containing responses of virtually all new Ph.D.'s in the United States, provides data on background characteristics, educational patterns, and post-degree plans at the time the degree is obtained. The present report relies heavily on this source for data on the 1977 doctorates and, for information on earlier cohorts, on the accumulated data from these surveys, referred to as the Doctorate Records File. In some instances, tabulations of these data by field and sex reveal very few cases in specific cells, e.g., women in physical sciences, but it should be stressed that these numbers reflect the entire population, rather than a sample, of the Ph.D's in a given category. Sampling error is not a consideration with respect to these tables (Tables 2.1, 2.4- 2.7, 2.10-3.2, and 3.5). In contrast, the biennial Survey of Doctorate Recipients is administered to a sample. The sample of 65,000 doctoral scientists and enginers is drawn primarily from the Doctorate Records File but also includes some individuals who earned their doctorates at foreign institutions. Data in this report on employment status, sector, activity, rank, and salary are from this source. The numbers in the tables from the Survey of Doctorate Recipients represent the sample weighted to yield an estimate of the doctoral population in the national labor force. In addition to these two major sources of statistical data, numerous individual studies and reports have been reviewed, and reference is made to these throughout the report. Limitations of the Data In tabulations from the Survey of Doctorate Recipients, small estimates may reflect even smaller numbers of sample cases. When the number of sample individuals in a cell is fewer than three, no figures or percentages are presented. For other cells containing small numbers for the estimated population, the reader is urged to exercise great caution in the interpretation of percentages. This reminder is repeated in a footnote on each table from the Survey of Doctorate Recipients (Tables 2.8, 2.9, 3.3, 3.4, and 3.6- 4.20). A. discussion of the survey is included in Appendix D. On other grounds, statistical findings must be applied with caution to determinations of sex discrimination; group 8
differences in quality or mobility would produce entirely legitimate statistical biases, for example. in addition, each academic hiring decision is in some sense unique and will involve personal assessments which, no matter how sincerely performed, may be swayed by one prejudice or another. In the absence of systematic sex discrimination in academic appointments, however, the sum of such individual decisions as reflected in aggregate statistics should not show bias but reflect the sex distribution of the available scientists of comparable quality. A further limitation in examining statistical data, however disaggregated, is that they cannot tell us much about the flow of individuals through the various professional levels. For example, we cannot tell whether the increase seen in numbers of women at senior ranks in some fields represents an upgrading of in-house candidates or recruitment from other institutions. Nor do the data allow us to distinguish between those junior research faculty members who move up and those who are forced to move out. Organization of the Report The first chapter of the report examines- some of the obstacles that women must overcome to become professional scientists. The following chapter assesses the characteristics, educational patterns, and supply of women doctorates in the sciences. Chapter 3 examines sex differences in postdoctoral training patterns. The fourth chapter presents recent developments in the academic employment of men and women scientists. Chapter 5 reviews the participation of women in three major groups within the national science advisory apparatus. The sixth chapter provides an overview of the current prospects of women scientists in academe as well as recommendations for improving these prospects. NOTES 1 See Table 4.19A on page 90. 2 These are described in Appendix C and D, respectively, and copies of the questionnaires are included.
