Portfolio Management for Research Project Support


As the only federal agency charged with strengthening the nation's research performance in essentially all fields of science and engineering, NSF aims to support a balanced portfolio of excellent research projects across all fields of science and engineering. Balance is sought through program planning, recognition of scientific opportunities, project selection, resource management, partnerships to reinforce and help fund our priorities, and attention to enabling present and future researchers to produce excellent research.

Candidates for Enhanced Investment

One of the first steps in managing the portfolio is to identify areas where enhanced investments will have a significant impact on the progress of science and engineering. Building intellectual capital that takes advantage of emerging scientific opportunities is key to keeping the U.S. at the cutting edge of research. NSF has identified three factors that are driving current change in the conduct of science and engineering research:

Attention to these factors helps us develop mid-term objectives for progress toward the overall goals of the NSF strategic plan.

Ensuring Sufficient Resources for Each Research Project Funded

NSF realizes that adequate resources are vital to each project's success, and is committed to providing them to every research project and center it supports. An agency goal is to continue to raise the median annual award size, as we have done over the last several years, to sustain the "buying power" of funded research projects. NSF has determined that, if necessary, it will decrease the number of awards made before decreasing award sizes.

Median Annualized Award Size


NSF is also taking steps to increase award durations as a mechanism of improving performance. Increasing award durations helps the research enterprise in two ways: it allows more time for the successful completion of projects, and it streamlines the application process by requiring less frequent re-submission of proposals. Longer award duration is balanced with the need for recurring project evaluation and merit review. Although the average duration of research project support has remained stable over the past ten years, an agency goal is to modestly increase award durations during FY 1998.

Average Duration
(years)


Merit Review System

NSF processes about 30,000 competitive proposals each year, including almost 25,000 for research project support. Over 95 percent of these proposals undergo external merit review. (Proposals not subject to formal external review include such activities as small workshops and the Small Grants for Exploratory Research Program, which are reviewed internally). NSF receives over 170,000 reviews each year, including over 70,000 reviews by mail, to help evaluate these proposals.

The funding rate is the number of competitive awards made during a year as a percentage of total proposals competitively reviewed. Funding rates for Research Project Support have fluctuated over the last several years.

Funding Rates for Research Project Support

NSF is currently examining all aspects of the proposal review process, including ways to streamline it without diminishing the quality of review decisions. A task force has proposed new generic merit-review criteria to be used in NSF project selection; these criteria are now being reviewed by a public-comment process. Another team, largely external to NSF, is developing options for addressing current strains on NSF's merit review system.

NSF recognizes the importance of having a diverse group of reviewers as part of the merit review process; an ongoing agency goal is to continue to increase the number of female, minority, and non-academic reviewers in the coming years. The number of women invited to review proposals has more than doubled in the last eight years, and women as a percentage of all reviewers has risen from 10 percent to 14 percent. Likewise, NSF recruits reviewers from under-represented ethnic groups from geographically diverse areas, and from non-academic scientists and engineers to provide a balanced review perspective.

The merit review process enables NSF to tap expertise in the entire science, engineering, and education community in order to assess and increase the overall value of the public's investments in research projects. This allows NSF to advance the state-of-the-art, shifting funds into promising new areas of inquiry, while maintaining balance across all the disciplines. As current awards expire, funds are reallocated to new opportunities. Supporting areas of emerging opportunity using the merit review process maintains the traditional high standards of excellence of NSF-supported research.

Monitoring Participation of Special Groups

NSF encourages special groups of investigators to increase their participation in the science research enterprise: new principal investigators (PIs) (those who did not have an NSF award in the 5-year period preceding the proposal), female principal investigators, and minority principal investigators. Each of these groups is crucial to making the national research infrastructure healthy and whole.

New principal investigators are important to the nation because they include future leaders of U.S. research. Bringing top notch young scientists and engineers into the system ensures continued excellence in research projects. While new principal investigators naturally have lower than average funding rates due to their inexperience, they received over 28 percent of competitive research project awards in both FY 1995 and FY 1996. NSF monitors these rates to ensure that a cadre of outstanding new investigators is indeed always arriving on the national R&D scene and receiving funding.

An ongoing goal of NSF is to ensure that the research enterprise is open to participation by all scientists and engineers.

Funding Rates for New and Prior Investigators in Research Projects


All principal investigators receive written copies of all reviews. Most of the new PIs that have been declined use these reviews to improve their proposals and resubmit them to NSF.

Just as new principal investigators bring a different perspective and ensure the future health of science in the nation, so female and minority principal investigators broaden perspective and ensure the health of science in the nation now. NSF's ongoing goal is to maintain these funding rates, while also increasing the number of females and minorities submitting proposals.

Funding Rates for Female and Minority Investigators in Research Projects


Partnerships

NSF derives many benefits from joining with other research fund providers in shared projects, and encouraging partnerships among various research performers. Effective partnerships broaden the views of participants from diverse settings, and increase their mobility and understanding of the goals of different parts of the research enterprise. They also uncover common objectives and efforts which can benefit from comparative advantages and economies of scale. For example, through interagency agreements, NSF-funded researchers may gain access to resources available through other federal agencies and other sources.

Promoting partnerships is one of NSF's core strategies. An indication of NSF's ability to use this strategy is the recent high level of support for NSF-funded projects provided by non-NSF sources, estimated to have been at least $1.4 billion in FY 1995 (for all of NSF's investments combined), and $1.7 billion in FY 1996. Besides strengthening linkages among the parts of the scientific enterprise, such funds magnify the effects of NSF efforts at priority setting.

Program Evaluation

Performance highlights for the Research Project Support function-results from the research projects we invest in-are set out below. In addition to such results, NSF uses evaluation studies to determine the merits and impacts of programs, to find ways in which they can be improved, and to inform planning for the future. These studies are of various sorts.

Two recent examples of program evaluation come from the Engineering Activity. Preliminary results from an ongoing retrospective study of NSF's role in the emergence of significant engineering-based innovations found that NSF's involvement can best be described as enablement, that it extended over many years and included numerous modes of support, and often occurred at crucial points in an innovation's development. Depending on the innovation studied, NSF's main role was either direct research support, or contributions to the supporting technologies and the relevantly trained people who helped develop the innovation, or organizational and coordinating leadership to move the innovation (the Internet, in this case) into the private sector. Another study, of interaction between Engineering Research Centers and participating companies, found that nearly 90 percent of firms received benefits from their participation in the ERCs, and two-thirds also experienced some positive impact on competitiveness. Companies experienced a wide variety of benefits, with the most valued being hiring students and graduates of the ERCs.

Other examples are the two recent evaluation studies of the NSF Science and Technology Centers program. One of these studies was performed by the Committee on Science, Engineering, and Public Policy of the National Academy of Sciences; the other by Abt Associates. In addition, an ad hoc STC Advisory Committee was convened to provide NSF with recommendations concerning the future of the STC program within the context of NSF's Strategic Plan. The committee used input from the two earlier studies as well as from NSF staff and STC directors. All these assessments concluded that the Science and Technology Centers program is making a major contribution to the nation's scientific and engineering research for a modest investment. In addition to recommending the continuation of the program, these reviews made many other recommendations, including suggestions about relative emphasis among goals and about management of the program, that contained valuable information and judgments.


Back to the Key Program Functions page

Next Section: Research Project Support Performance Highlights