NATIONAL SCIENCE FOUNDATION

    The National Science Foundation (NSF) is responsible for advancing science and engineering in the United States. NSF carries out its mission primarily by making merit-based grants to individual researchers and groups at more than 2,000 U.S. colleges, universities and other institutions. Although NSF represents about four percent of the total federal budget for research and development, it accounts for approximately one-fourth of all federal support for basic research at academic institutions. NSF evaluates research and education proposals using two criteria: the scientific merit of the proposed activity and the prospective impact on society. NSF categorizes its programs to align with its three strategic goals: 1) Ideas (research); 2) People (education); and 3) Tools (facilities and instrumentation).

Ideas

    To foster discoveries in science and engineering, NSF primarily invests in researchers and educators at colleges and universities. The majority of grant recipients’ work is in basic research (Ideas), which can yield important scientific discoveries that may lead to many applications. These applications have driven economic growth and have enhanced the quality of life through advances such as better weather forecasting, earlier detection of cancerous tumors, and the creation of the Internet.

    Although private industry has expanded its support for basic research over the past several years, its research focuses mostly on the short-term in order to bring new products to market. Federal investments in basic research provide a long-term foundation for breakthrough applications in areas not usually supported by private industry.

NSF-supported scientists are using a video camera on the back of a horseshoe crab to decipher the neural code for vision.

    Overall Performance. NSF is the leading performer among federal agencies funding basic research. For example, of the nearly 10,000 awards NSF makes annually, 94 percent of the research awards are made through competition, based on merit review. A competitive merit review process ensures that high-quality research is funded. The accompanying table displays the percent of research competed at selected federal agencies.

    NSF’s competitive approach pays rich dividends. Its grants often lay the early foundation for future breakthroughs. For example, of the 11 Nobel Prize winners in the sciences in 2001, eight received NSF funding for the research that won them the award.

    To further ensure high quality in its programs, external panels assess approximately one-third of NSF’s programs each year, so that all programs are reviewed in a three-year period. During the past two years, these panels have judged the majority of the programs assessed to be of high quality and efficiently managed. NSF’s reputation for running an efficient and effective competitive merit-review process has enabled it to provide leadership to other agencies, such as the Environmental Protection Agency and the Department of Education, in improving their research programs.

    One of NSF’s many strengths is its flexibility to redirect resources to emerging science and engineering opportunities. Unlike other agencies that own and operate numerous laboratories, NSF owns facilities related to only a few programs, such as the U.S. Antarctic Program. NSF is largely free from ongoing institutional obligations. In addition, NSF awards do not last indefinitely. The average NSF grant is typically for three years. This minimizes research stagnation or funding research that ceases to be important or cutting-edge. NSF also maintains programmatic flexibility by funding over one-half of new grants entirely in one year, rather than through installments. Instead of carrying financial commitments into future years, NSF can quickly redirect resources to new areas of emerging opportunity.

    All these features contrast starkly with the increasing amount of federal research dollars directed by congressional earmarks to projects without due regard to competition or merit.

    The Administration’s overall aim is to position NSF to invest in priority research areas, such as information technology and nanotechnology, which connect discovery to learning, innovation, and benefit to society. Nanotechnology, which involves controlling the building of small and large structures atom by atom, holds promise for the development of technologies that could range from higher-performance materials to biomedical instruments as small as human cells.

     Priorities like these tend to arise from NSF’s core research efforts--disciplinary and multidisciplinary programs that support ideas generated by the academic community. NSF allocates approximately 25 percent of its research budget in priority areas that will deliver scientific breakthroughs, and 75 percent in core programs to build the capacity needed for the emergence of new technologies.

    Improving Performance. The President’s Budget proposes to improve the quality and efficiency of federal funding of basic research at NSF by:

• Emphasizing research in highly promising, multidisciplinary areas. In addition to nanotechnology, the 2003 Budget provides significant NSF funding for fundamental research related to bioterrorism, information technology, mathematics, and climate change. Each of these areas has the potential for significant breakthroughs.

• Improving the quality of a number of science and engineering programs by transferring them to NSF. Based on NSF’s noted expertise and success in funding competitive research, the budget transfers the National Oceanographic and Atmospheric Administration’s (NOAA’s) Sea Grant program and the U.S. Geological Survey’s toxic substances hydrology research program to NSF to conduct merit-based competition and improve program effectiveness. The Sea Grant program will be administered in partnership with NOAA to ensure that the agency's research and outreach objectives are reflected in the program's ongoing work. The Administration may also transfer non-competitive funding from the Smithsonian Institution’s astrophysics and environment programs to NSF, following a program review by an independent panel.

• Improving efficiency of research by increasing grant size. One means of improving research efficiency is by providing adequately-funded grants to ensure the proposed work can be accomplished as planned. Inadequately funded grants can result in an inefficient research process, with an award only funding a portion of a research project. A researcher then has to write additional proposals to get funding to complete the project and realize research objectives. The 2003 Budget increases the average annual NSF award size to $120,000, an increase of approximately $30,000 since 1998. NSF believes reaching this award size will result in approximately 200 fewer awards (a two percent reduction), from 2002. NSF also believes that the increased size will help ensure that its grants are more effective in achieving research project objectives.

People

    NSF invests in People—students, researchers, and educators—to strengthen math, science, environmental, and engineering education, thus equipping the American workforce for the challenges of the 21^st Century.

NSF-supported graduate fellows are helping teach math and science concepts to students in kindergarten through twelfth grades.

    Overall Performance. Longtime concern persists over the state of grades K–12 science and mathematics education in the United States. The Third International Math and Science Study compared American and other countries’ students in math and science and found that U.S. fourth graders did relatively well in both subjects. But by the time they reached their senior year in high school, U.S. students ranked among the worst in the world. In 2000, the National Assessment of Educational Progress showed no improvement in U.S. student performance in science and limited improvement in mathematics since 1996.

    Achievement in mathematics and science is most directly dependent on state and local educational systems. NSF’s role is in supporting new models of math and science education. In the past decade, NSF has supported new models that, if successful, could be adopted by state and local districts, which have the resources to implement those models. Initial indications are that some of these NSF-supported models are proving successful in improving student achievement. For example, over the first six years of the NSF-funded Chicago Urban Systemic Initiative, the percentage of fourth grade students meeting Illinois state standards in science increased from 46 to 66. For the NSF-funded San Antonio Urban Systemic Initiative, the average scores of African-Americans in grade 4 on the Texas Assessment of Academic Skills increased by 32 percent over four years, and those of Hispanic students by 39 percentage points, compared to a 16 percentage point increase for Texas fourth-graders overall.

    In the area of graduate education there is concern that fewer U.S. students are enrolling in U.S. graduate science and engineering programs. Since 1993, enrollment of U.S. students in graduate level science and engineering programs dropped by nine percent. During the same period, enrollment of foreign students on temporary visas increased by three percent. If fewer scientists and engineers are entering the workforce, U.S. high technology firms may have to increasingly rely on foreign high technology workers who are in the U.S. on temporary non-immigrant worker visas.

    A recent survey of the Department of Education found that 57 percent of surveyed U.S. baccalaureate recipients did not apply to science and engineering graduate programs for financial reasons. Using their bachelor in science or engineering degree to get a job that may pay more than twice the level of a graduate student stipend (salary) is often more enticing to a person carrying debt from undergraduate school. One strategy of enabling U.S. students to go on to graduate school is to provide competitive stipends to ease the financial burden. NSF performance in the 1990s in providing competitive stipends was not good. From 1993 to 1999, NSF stipend levels dropped as a percentage of starting salaries for bachelor students in the sciences and engineering from 65 percent to 52 percent and the difference may be growing wider.

    Improving Performance. The President’s Budget proposes to strengthen math and science education in the United States by:

• Improving the quality of math and science education in Grades K–12 through the President’s Math and Science Partnerships Initiative. Support for the President’s Math and Science Partnerships initiative is increased in the 2003 Budget. The Partnerships Initiative builds on the fact that while states and local governments deliver education, NSF has a proven record in supporting successful models to enhance math and science curriculum and student test scores as a result. The Initiative provides funds for states and local school districts to join with institutions of higher learning, particularly with their departments of mathematics, science, and engineering, to beef up math and science education.

• Attracting the most promising U.S. students into graduate level science and engineering by providing more competitive stipends. The 2003 Budget increases the annual stipends for NSF’s fellowship and traineeship programs from $21,500 to $25,000 to further attract U.S. students to graduate level programs in science and engineering. NSF also will conduct a study on graduate stipends in 2002 to recommend what the ultimate target for graduate stipends should be as well as develop measures to assess its impact on the larger national effort to increase and improve graduate students in science and engineering.

• Improving quality of environmental education programs. Based on NSF's noted expertise and success in funding competitive programs, the budget transfers the Environmental Protection Agency's environmental education program to NSF to improve program effectiveness and merit-based selection.

Tools

    NSF invests in widely accessible, state-of-the-art science and engineering Tools—sophisticated instruments, equipment, facilities, databases, and large surveys. NSF’s funding of facilities has grown and diversified and now includes shared-use research facilities that are often connected by high-speed networks.

    Except for U.S. Antarctic Program facilities, NSF does not directly operate the large-facilities that it supports, such as the Gemini North telescope in Hawaii or the Terascale Supercomputer in Pittsburgh. NSF primarily makes awards to universities and non-profit organizations to construct, manage, and operate large projects.

The Gemini North Telescope on Mauna Kea in Hawaii provides some of the sharpest images of any telescope on Earth.

    Overall Performance. Research agencies must strive to keep the development and upgrade of research facilities on schedule and within budget. In running the facilities, agencies should keep the operating time lost due to unscheduled downtime to a minimum. NSF does relatively well in meeting these goals. In 2000, all 11 construction projects that NSF supports were within 10 percent of their estimated annual cost, and seven of the 11 projects were within 10 percent of meeting their annual schedule milestones. Also in 2000, 22 of 26 operating facilities kept time lost due to unscheduled downtime to less than 10 percent of the total scheduled operating time. For major capital projects completed since 1996, cost growth on five science projects was eight percent (or $36 million), generally less than the average cost increase for major projects at most science agencies.

    Although NSF has done relatively well in managing construction of its large facilities, project complexity, cost, and risk are increasing. Future projects will challenge traditional NSF approaches. To address this concern, the Administration directed NSF to develop a plan to enhance its management of large facility projects. In response, NSF is now implementing a Large Facilities Projects Management and Oversight Plan that improves the process for reviewing and approving large projects and increases oversight of its projects. All current and future large projects will be subject to these new guidelines and oversight.

     Improving performance. The President’s Budget proposes to improve NSF investments in Tools by:

• Enhancing infrastructure capabilities in astronomy, earthquake research, and the environment. The budget proposes initiating construction of the international Atacama Large Millimeter Array telescope in Chile and the Earthscope projects across the United States. The Atacama Large Millimeter Array will be the world’s most sensitive, highest resolution, millimeter-wavelength telescope. This telescope will serve as a testing platform for theories of star birth and stellar evolution, galaxy formation and evolution, and the evolution of the universe itself. Earthscope will provide several instruments, some portable, to investigate the structure and evolution of the North American continent and the physical processes controlling earthquake and volcanic eruptions. This will provide significant data to assess and mitigate national risks associated with earthquakes, volcanic eruptions, and landslides. The 2003 Budget also provides funding to test at least two sites of the National Ecological Observatory Network, which will provide an integrated network of regional environmental research observatories.

• Improving priority setting and the visibility of the selection process for large facility projects. For the first time, the 2003 Budget identifies funding for early-stage planning and development of potential new, large facility projects. This will increase the visibility of NSF’s facility selection process. The Office of Science and Technology Policy also will request that the National Academy of Sciences review the scientific merit of IceCube and other proposed U.S. neutrino collectors in the context of current and planned neutrino research capabilities throughout the world. Neutrinos are one of the fundamental particles that make up the universe and are also one of the least understood. Understanding neutrinos better will mean greater understanding of the universe.

Status Report on Select Programs

    The Administration is reviewing programs throughout the federal government to identify strong and weak performers. The budget seeks to redirect funds from lesser performing programs to higher priority or more effective ones.

Strengthening Management

    NSF is a relatively well-run agency. Funding for the agency has grown significantly in the past decade, while the agency’s staffing level has remained flat. The agency has accommodated the increase in funding and responsibilities through the use of information technology and continued reliance on outsourcing support of NSF’s review process to the academic community. Nevertheless, there are major hurdles on the horizon.

    Of the total federal funds NSF receives, 95 percent go to researchers and educators; the agency’s overhead is only five percent. Many in Congress and the NSF Inspector General have questioned whether the agency has enough resources to adequately manage its growing portfolio and conduct adequate oversight of its awards. The 2003 Budget addresses these concerns by providing a significant funding increase to expand award oversight by providing for more travel to review large award recipients, providing additional personnel through temporary and permanent appointments, and enhancing information technology (IT) systems to improve worker productivity and efficiency of the award process.

    NSF has been better managed and has a better baseline evaluation than most other agencies. For example, NSF is the only agency to receive the top rating for financial management. NSF is also a federal government leader for e-government and information technology. The growing demands on NSF, however, will require it to further improve its management. In particular, NSF needs to improve results of its human capital management, competitive sourcing, and integration of performance and the budget efforts. A scorecard of NSF’s activities for the President’s management initiatives follows. The agency is performing well, but there are areas of concern.