There is also the problem of the time frame for measuring the outcomes of basic research. There is generally a long gestation period for such research. That means that measurement must be a continuous process. It should not be done once; it should be done on a regular, long-term basis.
The final issue area is that this assessment process should look to the future as well as to the past. Some means should be developed for determining whether the right basis for budgeting and program planning is being established. The NSF needs to think in terms of the requirements for solving problems that could be identified by constituent groups, as it were, or that are important to constituent groups. This may lead to the identification of certain fields that are important; if so, it may be possible to develop some way of determining whether the nation's educational and research allocations are appropriate for building the basis needed to resolve those problems.