Sir

A recent News story by Colin Macilwain implies that the US Government Performance and Results Act (GPRA) will prompt National Science Foundation (NSF) staff to look over the shoulders of grant recipients in an unprecedented way (Nature 393, 100; 1998). The foundation has always required grant recipients to report on annual progress and to provide final project reports. Rather than monitoring grants more intensely, we will simply be capturing this information in a way more useful for reporting on performance of our programmes.

We are also asking for the help of grant recipients in identifying less immediately obvious effects of their activities, and to communicate those in terms that lay readers can understand.

For GPRA, NSF will examine and report on the performance of programmes as a whole, rather than the performance of individual grants. Processes are already in place in some directorates to obtain summary information on programme activities. GPRA requires that we obtain that information in a way that can be aggregated to the level of NSF as a whole and prepared for assessment by panels of external experts. We are attempting to structure this aspect of GPRA so that it has minimal impact on NSF staff and the community.