How to Measure Performance
Neal Lane, a computational physicist who had been provost of Rice University, succeeded Acting Director Frederick Bernthal in 1993. Popular and effective, he would remain until moving to the White House in 1998 as President Clinton's Science Advisor. While NSF management was in sure hands, the Board had plenty of oversight challenges. The Government Performance and Results Act (GPRA) of 1993 requires federal agencies to account for program results through a system of performance measurements. Progress is noted by OMB and Congress at budget time. But the open-ended nature of discovery-driven research is notoriously resistant to short-term assessment. What's more, grantmakers pressured for close accountability can become too cautious, neglecting emerging fields or risky investigations for those already likely to bear fruit.
The Foundation won approval to use more qualitative measures of performance. Its first GPRA plan was issued in 1995. A report, NSF in a Changing World, summarized the NSF strategic plan in which the Board was to watch closely whether Foundation awards and actions matched the new GPRA outcomes. Such outcomes included making "connections between discoveries and their use in service to society," and whether, for example, the Foundation's $700 million education programs were producing a "diverse, globally oriented science and engineering workforce."
