petition among institutions that would lead to enhanced graduate programs (mentoring, course offerings, research opportunities, and facilities) and processes (time to degree, career guidance, placement assistance). To be sure, institutions can and should undertake many of those improvements in graduate programs even without this stimulus, and many have already implemented reforms to make graduate school more enticing. Institutional efforts to prepare graduate students for the jobs they will obtain in industry or academe and to improve the benefits and work conditions for postdoctoral scholars also could make career prospects more attractive.
The new program proposed here and led by NSF should draw advice from representatives of federal research agencies to determine its areas of focus. On the basis of that advice, NSF would make competitive awards either as part of its existing Graduate Research Fellowship program or through a separate program established specifically to administer the fellowships. The focus on areas of national need is important to ensure an adequate supply of suitably trained doctoral scientists, engineers, and mathematicians and appropriate employment opportunities for these students upon receipt of their degrees.
As discussed in Box 7-1, one question is whether these programs will simply produce science and engineering students who are unable to find jobs. There are also questions that the goal of increasing the number of domestic students is contrary to the committee’s other concern about the potential for declining numbers of outstanding international students. As past National Academies reports have indicated, projecting supply and demand in science and engineering employment is prone to methodological difficulties. For example, the report Forecasting Demand and Supply of Doctoral Scientists and Engineers: Report of a Workshop on Methodology (2000) observed:
The NSF should not produce or sponsor “official” forecasts of supply and demand of scientists and engineers, but should support scholarship to improve the quality of underlying data and methodology.
Those who have tried to forecast demand in the past have often failed abysmally. The same would probably be true today.
Other factors also influence the decisions of US students. As the recent COSEPUP study, Policy Implications of International Graduate Students and Postdoctoral Scholars in the United States, says:
Recruiting domestic science and engineering (S&E) talent depends heavily on students’ perception of the S&E careers that await them. Those perceptions can be solidified early in the educational process, before students graduate from high school. The desirability of a career in S&E is determined largely by the