Braun Tibor, Schubert András (szerk.): Szakértői bírálat (peer review) a tudományos kutatásban : Válogatott tanulmányok a téma szakirodalmából (A MTAK Informatikai És Tudományelemzési Sorozata 7., 1993)
IAN I. MITROFF and DARYL E. CHUBIN: Peer Review at the NSF: A Dialectical Policy Analysis
123 MITROFF & CHUBIN: PEER REVIEW AT THE NSF and particularism is built into the peer review process. Why deny this fact, as Cole seems to do? And why present no data (for example, from interviews) which might contain clues about the tension — namely, the extent to which perceptions of quality are coloured by particularistic considerations?" Third, if we ponder — as we did in our reaction to the Hensler study — the kind of data needed to advance the peer review debate beyond its present impasse, our thoughts return to the characteristics of reviewer and/or applicant which might influence the final decision to fund or not to fund a proposal. Recent studies of the cognitive-styles of inquiry of scientists reveals that one of the key dimensions distinguishing various styles and scientists from one another is the ability to make, as well as to appreciate, fine differentiations between people, objects, or institutions. 5 4 Persons who excel at this ability are called 'high differentiators'. They, in short, have a high tolerance for ambiguity. As Gordon and Morse put it: High differentiators perceive their environment as a series of discrete parts while low differentiators see their environment as highly homogeneous . . . The ability to differentiate manifests itself in two related ways depending on the nature of the stimulus, human or inanimate. In interacting with people the high differentiator perceives and reacts to each as a unique individual possessing a combination of capabilities and inabilities. The low differentiator perceives people as being more or less alike and thus tends to suppress or ignore individual capabilities. 5 5 The point is that low differentiators would tend to see the personal characteristics of an investigator as irrelevant to a proposed investigation because they would see all investigators in a similar light. High differentiators, on the other hand, tend to see personal characteristics as very relevant. Specifically, then, does the sample studied by Cole contain an overabundance of low differentiators? Does a large sample tend to mask or damp out the effect of high differentiators? Does the institutional or social process of rating proposals induce even a high differentiator to act like a low differentiator? That is, does the social process of rating proposals foster a 'do-unto-others-what-might-be-done-unto-you' approach? In short, if we had a sample of clearly identified high differentiators and another of clearly identified low differentiators and we gave each the same set of proposals to rate, would their ratings be the same? Before one can reject the hypothesis that the characteristics of the rater/reviewer and those of the individual being rated are irrelevant to the ultimate disposition of a proposal, one must at least attempt to construct a kind of experiment to test the hypothesis.
