Miron L. Straf
National Academy of Sciences--National Research Council
As I read these excellent papers, two thoughts come to mind for a title for this session. Reading
Kirkendall's paper, I would name the session, "Statistician: Measure thyself;" reading the paper by
Collins, Rapoport, and Tupek, I would name it, "Let's win one for the GPRA."
I may have been asked to comment on these papers about performance measures for research and
statistical agencies, because of a relevant study of the Committee on National Statistics (CNSTAT).
The Committee has convened a panel to assess and recommend performance measures for local,
state, and federal officials to use in appraising progress in meeting the objectives of public health
Performance Partnership Grants in several areas, including mental health, substance abuse, HIV,
chronic diseases, and preventive health services. The panel, sponsored by the Department of Health
and Human Services, is chaired by CNSTAT member Edward B. Perrin; Jeffrey J. Koshel is study
director. My remarks are informed by the development of this study and discussions with the panel,
but are those of my own and not necessarily those of the panel, Committee, or the National Academy
of Sciences--National Research Council.
In fact, it's ironic that the Academy is involved at all in improving performance. The Academy is
a unique institution, quasi-government and quasi-academic, and, as a result, is the only institution
that is run with both government efficiency and academic effectiveness.
The quasi academic side approaches performance measurement and the Government Performance
and Results Act of 1993 (GPRA) with cynicism, epitomized by Sidney Winter:
The producers of management advice operate in an economic environment where the tides
of fad and fashion run strong. Frameworks, slogans, and buzzwords are brought forth in
great profusion with attendant fanfare and claims of novelty. Although large rewards often
accrue to successful fashion leaders, it is open to questio