ably include investigators who do not pursue health-related research and exclude others from related fields who may be actively involved in such studies. Economists, for example, were not included in our analysis, even though studies of the economics of health and medicine make major contributions to health-related research. Because less than 2 percent of the nation's economists focus on health economics, 4 including this field in our analysis would overestimate the size of the workforce. Nonetheless, the definition of the behavioral and social science workforce described above can provide a general estimate of the number of investigators in this field and an indication of the major trends affecting the workforce, such as changes in size, age, and composition.
As in the basic biomedical sciences, the behavioral and social science workforce has more than doubled since 1975, from just over 25,800 Ph.D.s to 57,800 in 1997 (see Figure 3-1 ). In contrast to the steady growth of the basic biomedical workforce during that period, most of the increase in the behavioral and social sciences occurred between 1975 and 1989, when the size of the workforce climbed from 25,802 to 48,844 Ph.D.s. Growth slowed markedly thereafter, and by 1995 the workforce totaled 52,324 Ph.D.s. By 1997 the number of behavioral and social scientists had jumped again, to 57,843.
This recent growth, however, should be interpreted with caution. A change in the survey methodology in 1993 may have affected subsequent estimates of workforce size by classifying individuals by occupation (e.g., scientist, professor, manager), rather than by scientific field. In addition, part of this reported workforce growth stems from a change in the analysis of survey responses by the National Opinion Research Center when it took over the management of the Survey of Doctorate Recipients in 1997.
Since 1975 the representation of women and minorities in the behavioral and social science workforce has grown significantly (see Figure 3-2 ). In the mid-1970s, 20.5 percent of the behavioral science workforce were women and 2.5 percent were minorities. Since then, as increasing numbers of women and minorities have earned Ph.D.s in these fields, their representation in the workforce has steadily risen. By 1997, 41.9 percent of behavioral and social scientists were women and 6.9 percent were minorities.
In the absence of a rapid increase of new Ph.D.s, as seen in the basic biomedical sciences, the median age of the behavioral and social science workforce has grown over the last decade, rising from 44.4 in 1987 to
Sikes, Violet. American Economic Association. Personal communication, October 1999.