Summary of Recommendations
- New ideas form the foundation of the research
enterprise. It is in our interests for the Nation's scientists to continue
pursuing fundamental, ground-breaking research. Our experience with 50 years of
government investment in basic research has demonstrated the economic benefits of
this investment. To maintain our Nation's economic strength and international
competitiveness, Congress should make stable and substantial federal funding for
fundamental scientific research a high priority.
- Notwithstanding the short-term projections of budget surpluses, the resources
of the federal government are limited. This reality requires setting priorities
for spending on science and engineering. Because the federal government has an
irreplaceable role in funding basic research, priority for federal funding should
be placed on fundamental research.
- The primary channel by which the government stimulates
knowledge-driven basic research is through research grants made to individual
scientists and engineers. Direct funding of the individual researcher must
continue to be a major component of the federal government's research investment.
The federal government should continue to administer research grants that
include funds for indirect costs and use a peer-reviewed selection process, to
individual investigators. However, if limited funding and intense competition
for grants causes researchers to seek funding only for "safe" research, the
R&D enterprise as a whole will suffer. Because innovation and creativity
are essential to basic research, the federal government should consider
allocating a certain fraction of these grant monies specifically for creative,
- The practice of science is becoming increasingly interdisciplinary, and
scientific progress in one discipline is often propelled by advances in other,
seemingly unrelated, fields. It is important that the federal government fund
basic research in a broad spectrum of scientific disciplines, mathematics, and
engineering, and resist concentrating funds in a particular area.
- Much of the research funded by the federal government is related to the
mission of the agency or department that sponsors it. Although this research is
typically basic in nature, it is nevertheless performed with overriding agency
goals in mind. In general, research and development in federal agencies,
departments, and the national laboratories should be highly relevant to, and
tightly focused on, agency or department missions.
- The national laboratories are a unique national resource within the research
enterprise, but there are concerns that they are neither effective nor efficient
in pursuing their missions. A new type of management structure for the federal
labs may provide one solution and deserves exploration. To that end, a national
laboratory not involved in defense missions should be selected to participate in
a corporatization demonstration program in which a private contractor takes over
day-to-day operations of the lab.
- We also have the obligation to ensure that the money spent on basic research
is invested well and that those who spend the taxpayers' money are accountable.
The Government Performance and Results Act was designed to provide such
accountability. Government agencies or laboratories pursuing mission-oriented
research should employ the Results Act as a tool for setting priorities and
getting the most out of their research programs. Moreover, in implementing the
Results Act, grant-awarding agencies should define success in the aggregate,
perhaps by using a research portfolio concept.
- Partnerships in the research enterprise can be a valuable means
of getting the most out of the federal government's investment. Cooperative
Research and Development Agreements (CRADAs) are an effective form of partnership
that leverages federal research funding and allows rapid commercialization of
federal research. When the research effort involved in a CRADA fulfills a
legitimate mission requirement or research need of the federal agency or national
lab, these partnerships should be encouraged and facilitated. Partnerships
between university researchers and industries also have become more prevalent as
a way for universities to leverage federal money and industries to capture
research results without building up in-house expertise. University-industry
partnerships should, therefore, be encouraged so long as the independence of the
institutions and their different missions are respected.
- International scientific collaborations form another important
aspect of the research enterprise. While most international collaborations occur
between individuals or laboratories, the U.S. participates in a number of
large-scale collaborations where the costs of large scale science projects can be
shared among the participants. In general, U.S. participation in international
science projects should be in the national interest. The U.S. should enter into
international projects when it reduces the cost of science projects we would
likely pursue unilaterally or would not pursue otherwise. Our experience with
international collaborations has not been uniformly successful, as our
participation in Mir and the International Space Station demonstrate.
Therefore, a clear set of criteria for U.S. entry into, participation in, and
exit from an international scientific project should be developed.
- Large-scale international projects often take place over many years,
requiring stable funding over long periods. The annual appropriations cycle in
Congress can lead to instability in the funding stream for these projects,
affecting our ability to participate. The importance of stability of funding
for large-scale, well-defined international science projects should be stressed
in the budget resolution and appropriations processes.
- It is also important that international science projects not appear to be
simply foreign aid in the guise of research. To that end, when the U.S. is a
major contributor of funds to projects with international participation, funding
priority must be placed on the U.S.-based components.
- America's pre-eminent position in the world suggests new roles for U.S.
science policy in the international arena. To take advantage of these
opportunities, the State Department must broaden its scientific staff expertise
to help formulate scientific agreements that are in America's interest. The
evidence suggests that the State Department is not fulfilling this role.
Mechanisms that promote coordination between various Executive branch Departments
for international scientific projects must be developed. The State Department
should strengthen its contingent of science advisors within its Bureau of Oceans
and International, Environmental, and Scientific Affairs and draw on expertise in
- A private sector capable of translating scientific discoveries into products,
advances and other developments must be an active participant in the overall
science enterprise. However, there is concern that companies are focusing their
research efforts on technologies that are closest to market instead of on
mid-level research requiring a more substantial investment. Capitalization of
new technology based companies, especially those that are focused on more
long-term, basic research, should be encouraged. In addition, the R&D tax
credit should be extended permanently, and needlessly onerous regulations that
inhibit corporate research should be eliminated.
- Partnerships meant to bring about technology development also are important.
Well-structured university-industry partnerships can create symbiotic
relationships rewarding to both parties. These interactions and collaborations,
which may or may not involve formal partnerships, are a critical element in the
technology transfer process and should be encouraged.
- Partnerships that tie together the efforts of state governments, industries,
and academia also show great promise in stimulating research and economic
development. Indeed, states appear far better suited than the federal government
to foster economic development through technology-based industry. As the
principal beneficiaries, the states should be encouraged to play a greater role
in promoting the development of high-tech industries, both through their support
of colleges and research universities and through interactions between these
institutions and the private sector.
- The university community, too, has a role in improving research capabilities
throughout its ranks, especially in states or regions trying to attract more
federal R&D funding and high-tech industries. Major research universities
should cultivate working relationships with less well-established research
universities and technical colleges in research areas where there is mutual
interest and expertise and consider submitting, where appropriate, joint grant
proposals. Less research-intensive colleges and universities should consider
developing scientific or technological expertise in niche areas that complement
local expertise and contribute to local economic development strategies.
- To exploit the advances made in government laboratories and universities,
companies must keep abreast of these developments. The RAND Corporation's RaDiUS
database and the National Library of Medicine's PubMed database serve useful
purposes in disseminating information. Consider expanding RaDiUS and PubMed
databases to make them both comprehensive and as widely available as possible.
- Intellectual property protections are critical to stimulating the private
sector to develop scientific and engineering discoveries for the market. The
Bayh-Dole Act of 1980, which granted the licensing rights of new technologies to
the researchers who discover them, has served both the university and commercial
sectors reasonably well. A review of intellectual property issues may be
necessary to ensure that an acceptable balance is struck between stimulating the
development of scientific research into marketable technologies and maintaining
effective dissemination of research results.
- While the federal government may, in certain circumstances, fund applied
research, there is a risk that using federal funds to bridge the mid-level
research gap could lead to unwarranted market interventions and less funding for
basic research. It is important, therefore, for companies to realize the
contribution investments in mid-level research can make to their competitiveness.
The private sector must recognize and take responsibility for the performance
of research. The federal government may consider supplementary funding for
private-sector research projects when the research is in the national interest.
Congress should develop clear criteria, including peer review, to be used in
determining which projects warrant federal funding.
- Science and engineering also provide the basis for making decisions as a
society, as corporations and as individuals. Science can inform policy issues,
but it cannot decide them. In many cases science simply does not have the answer,
or provides answers with varying degrees of uncertainty. If science is to inform
policy, we must commit sufficient resources to get the answers regulators need to
make good decisions. At the earliest possible stages of the regulatory process,
Congress and the Executive branch must work together to identify future issues
that will require scientific analysis. Sufficient funding to carry out these
research agendas must be provided and should not be overly concentrated in
- For science to play any real role in legal and policy decisions,
the scientists performing the research need to be seen as honest brokers. One
simple but important step in facilitating an atmosphere of trust between the
scientific and the legal and regulatory communities is for scientists and
engineers to engage in open disclosure regarding their professional background,
affiliations and their means of support. Scientists and engineers should be
required to divulge their credentials, provide a resume, and indicate their
funding sources and affiliations when formally offering expert advice to
decision-makers. The scientific opinions these experts offer also should
stand up to challenges from the scientific community. To ensure that
decision-makers are getting sound analysis, all federal government agencies
pursuing scientific research, particularly regulatory agencies, should develop
and use standardized peer review procedures.
- Peer review constitutes the beginning, not the end, of the scientific
process, as disagreement over peer-reviewed conclusions and data stimulate
debates that are an integral part of the process of science. Eventually,
scientists generate enough new data to bring light to previously uncertain
findings. Decision-makers must recognize that uncertainty is a fundamental
aspect of the scientific process. Regulatory decisions made in the context of
rapidly changing areas of inquiry should be re-evaluated at appropriate times.
- Aside from being based on a sound scientific foundation, regulatory decisions
must also make practical sense. The importance of risk assessment has too often
been overlooked in making policy. We must accept that we cannot reduce every risk
in our lives to zero and must learn to deploy limited resources to the greatest
effect. Comprehensive risk analysis should be standard practice in regulatory
agencies. Moreover, a greater effort should be made to communicate various risks
to the public in understandable terms, perhaps by using comparisons that place
risks in the context of other, more recognizable ones.
- The judicial branch of government increasingly requires access to sound
scientific advice. Scientific discourse in a trial is usually highly contentious,
but federal judges have recently been given the authority to act as gatekeepers
to exclude unreliable science from the courtroom. More and more judges will seek
out qualified scientists to assist them in addressing complex scientific
questions. How these experts are selected promises to be an important step in the
judicial process. Efforts designed to identify highly qualified, impartial
experts to provide advice to the courts for scientific and technical decisions
must be encouraged.
- In Congress, science policy and funding remain scattered piecemeal over a
broad range of committees and subcommittees. Similarly, in the Executive branch,
science is spread out over numerous agencies and departments. These diffusive
arrangements make effective oversight and timely decision making extremely
difficult. Wherever possible, Congressional committees considering scientific
issues should consider holding joint hearings and perhaps even writing joint
- No factor is more important in maintaining a sound R&D enterprise than
education. Yet, student performance on the recent Third International Math and
Science Study highlights the shortcomings of current K-12 science and math
education in the U.S. We must expect more from our Nation's educators and
students if we are to build on the accomplishments of previous generations. New
modes of teaching math and science are required. Curricula for all elementary
and secondary years that are rigorous in content, emphasize the mastery of
fundamental scientific and mathematical concepts as well as the modes of
scientific inquiry, and encourage the natural curiosity of children must be
- Perhaps as important, it is necessary that a sufficient quantity of teachers
well-versed in math and science be available. Programs that encourage
recruitment of qualified math and science teachers, such as flexible credential
programs, must be encouraged. In general, future math and science teachers should
be expected to have had at least one college course in the type of science or
math they teach, and, preferably, a minor. Ongoing professional development for
existing teachers also is important. Another
disincentive to entry into the teaching profession for those with a technical
degree is the relatively low salaries K-12 teaching jobs offer compared to
alternative opportunities. To attract qualified science and math teachers,
salaries that make the profession competitive may need to be offered. School
districts should consider merit pay or other incentives as a way to reward and
retain good K-12 science and math teachers.
- The revolution in information technology has brought with it exciting
opportunities for innovative advances in education and learning. As promising as
these new technologies are, however, their haphazard application has the
potential to adversely affect learning. A greater fraction of the federal
government's spending on education should be spent on research programs aimed at
improving curricula and increasing the effectiveness of science and math
- Graduate education in the sciences and engineering must strike a careful
balance between continuing to produce the world's premier scientists and
engineers and offering enough flexibility so that students with other ambitions
are not discouraged from embarking on further education in math, science, or
engineering. While continuing to train scientists and engineers of unsurpassed
quality, higher education should also prepare students who plan to seek careers
outside of academia by increasing flexibility in graduate training programs.
Specifically, Ph.D. programs should allow students to pursue coursework and gain
relevant experience outside their specific area of research.
- The training of scientists and engineers in the U.S. occurs largely through
an apprenticeship model in which a student learns how to perform research through
hands-on experience under the guidance of the student's thesis advisor. A result
of this link between education and research is that students and post-doctoral
researchers are responsible for actually performing much of the federally-funded
research done in universities. Mechanisms for direct federal funding of
post-docs are already relatively common. Expansion of these programs to include
greater numbers of graduate students in math, science and engineering should be
- Increased support for Masters programs would allow students to pursue an
interest in science without making the long commitment to obtaining a Ph.D., and
thus attract greater numbers of students to careers in science and technology.
More university science programs should institute specially-designed Masters of
Science degree programs as an option for allowing graduate study that does not
entail a commitment to the Ph.D.
- The length of time involved and the commensurate forfeiture of income and
benefits in graduate training in the sciences and engineering is a clear
disincentive to students deciding between graduate training in the sciences and
other options. Universities should be
encouraged to put controls on the length of time spent in graduate school and
post-doctoral study, and to recognize that they cannot attract talented young
people without providing adequate compensation and benefits.
- Educating the general public about the benefits and grandeur of science is
also needed to promote an informed citizenry and maintain support for science.
Both journalists and scientists have responsibilities in communicating the
achievements of science. However, the evidence suggests that the gap between
scientists and journalists is wide and may be getting wider. Closing it will
require that scientists and journalists gain a greater appreciation for how the
other operates. Universities should
consider offering scientists, as part of their graduate training, the opportunity
to take at least one course in journalism or communication. Journalism schools
should also encourage journalists to take at least one course in scientific
- As important as bridging the gap between scientists and the media is, there
is no substitute for scientists speaking directly to people about their work. In
part because science must compete for discretionary funding with disparate
interests, engaging the public's interest in science through direct interaction
is crucial. All too often, however, scientists or engineers who decide to spend
time talking to the media or the public pay a high price professionally, as such
activities take precious time away from their work, and may thus imperil their
ability to compete for grants or tenure. Scientists and engineers should be
encouraged to take time away from their research to educate the public about the
nature and importance of their work. Those who do so, including tenure-track
university researchers, should not be penalized by their employers or
- The results of research sponsored by the Federal government also
needs to be more readily available to the general public, both to inform them and
to demonstrate that they are getting value for the money the government spends on
research. Government agencies have a
responsibility to make the results of federally-funded research widely available.
Plain English summaries of research describing its results and implications
should be prepared and widely distributed, including posting on the