COMMENT ON "THE REWARD SYSTEM IN SCIENCE".

The article presents comments on "Chemists in British Universities: A Study of the Reward System in Science," by sociologists S.S. Blume and Ruth Sinclair. The author finds many flaws in their study. Blume and Sinclair do not report tests for the significance of gamma. Such tests might have indicated a need to re-interpret the their conclusion that "when productivity is held constant the association between recognition and industrial involvement falls substantially. They might have tried an even more appropriate strategy to determine the effect of the third variable (Productivity). James A. Davis (1967) has suggested a procedure for computing a partial gamma, a statistic that has a "proportionate reduction in error" interpretation. When computed for the data this partial gamma has a value of .33. Again, although there has been a decrease in the size of the coefficient, the original relationship has by no means been destroyed. Statistical refinement mentioned would yield plausible alternate interpretations of the data.