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)
RUSTUM ROY: Alternatives to Review by Peers: A Contribution to the Theory of Scientific Choice
147 ROY: ALTERNATIVES TO REVIEW BY PEERS interdisciplinary laboratory. The arrangement would work as follows: a governmental body concerned, for example, with chemistry, would maintain records submitted annually by every institution of higher education conducting research or graduate education or both . The data supplied would include information on the "productivity" of the institution with respect to the four functions, including the number of papers published , mainly in a set of journals agreed by the staff of the department or laboratory. Each member can only be counted as one, although his or her work may be split over two or more units, with no double counting of product permitted. At a later stage, citations might be used as a measure of scientific productivity, but for the first stage, there is little difference between using papers and using citations as a measure of productivity when one aggregates the publications of 20 to 40 persons over three to five years. * The university would provide the numbers of advanced degrees granted, in each unit, each year. A simple scheme of weighting different degrees would reflect the amount of effort used to produce them. The measure of the value of the particular scientific capacity of the unit for serving practical or missionoriented ends is the amount of financial support received from missionoriented government agencies. The effectiveness in the performance of the fourth is the total financial support for research received from private industry. These data would of course be available for each year; the measures used would be a rolling average of the preceding three or four years. The combined formula: The actual formula using the data provided would be as follows, where all numbers represent rolling three-year averages: Total sum to be granted to unit = A x number of publications + B x weighted number of advanced degrees + Cx sum received for research from mission-oriented agencies 4- D x sum received for research from private industry. The weighting factors—A, B, C, D —would be adjusted by each agency so that the total of money distributed to all institutions would equal the total budget. The relative values of these factors becomes a flexible device for making policy. The weighting scheme is easily understood by high administrators and legislators. For example, if it were desired to encourage collaboration between university and industry, it would only be necessary for the legislative body to increase the weighting of D; this signal would immediately be perceived and acted upon throughout the country. Likewise, if there were shortages of qualified persons in one area, and a surplus in another, the legislature would simply rule that, for example, factor B would be tripled in the field of shortage and halved in the field in which there was a surplus of trained persons. " Roy, Rustum, Roy, N. R.. and Johnson. G. G., "Approximating Total Citation Counts Irom First Author Counts and írom Total Papers". Scientometrics , V, 2 (1983), pp. 117-124.
