Tuesday, August 16, 2005

British Journal of Medical Psychology (1984), 57, 385.


TYPE: A rejoinder to Jenkins & Zyzanski

John J. Ray

Ray & Bozek (1980) suggested that the Jenkins Activity Survey or 'JAS' (for measuring A-B personality) contained items measuring three different qualities: dominance, achievement motivation and 'freneticism'. Although freneticism was found to be only a minor component of the JAS, it seems to correspond most closely to what the JAS is said to be measuring.

Jenkins & Zyzanski's (1982) reply to that paper suggests, however, vacillation over whether there is any such thing as a single 'A-type' at all. Although patients are assigned a single score representing their degree of 'A-ness', Jenkins & Zyzanski (1982, p. 219) say: 'We conceive of the type A behaviour pattern as a configuration of psychological traits involving independent components' (italics added). This is confusing. Is there one trait (A-ness) that predicts CHD (coronary heart disease) or are there several independent traits? If there are several independent traits should they not be described and studied separately? If they are independent, what might be true of the one might not be true of all. Only if they were related traits would it make sense to group them together. Is there, in fact, a 'Type A' personality?

I originally thought that the items of the JAS were designed to measure aspects of a single personality type (or 'behaviour pattern') called 'A-B' and therefore Studies II, III and IV of Ray & Bozek (1980) used only items which could be shown to be central to what the inventory as a whole was measuring. This practice was continued in Ray & Simons (1982), where only the 18 most discriminating items of the JAS were used. Jenkins & Zyzanski (1982), however, suggest that this shortening makes my results non-comparable with theirs. However, in their own work they have used many different forms of their inventory and usually score only 21 items to get the A-B score (see the JAS manual). Their practice gives the impression that satisfactory results are obtained with a wide variety of forms of the inventory. They mention that I did not use their weighting system to score each item but they fail to mention their own earlier finding (see Ray & Bozek, 1980) that unweighted additions of item scores give results that are indistinguishable from those derived by weighting formulas.

Using unit weights, therefore, I reprocessed the data from Study I of Ray & Bozek (1980) to score both the 21 items prescribed by the Jenkins group for measuring 'A-B' and the 18 items used by Ray & Simons (1982) to score 'A-B'. After attenuation correction, the two forms of the scale correlated + 1.00. Operationally, therefore, the concepts of 'A-B' used by me and by the Jenkins group are identical.


Jenkins, C. D. & Zyzanski, S. J. (1982). The type A behaviour pattern is alive and well - when not dissected: A reply. British Journal of Medical Psychology, 55, 219.

Ray, J.J. & Bozek, R.S. (1980) Dissecting the A-B personality type. British Journal of Medical Psychology 53, 181-186.

Ray, J.J. & Simons, L. (1982) Is authoritarianism the main element of the coronary-prone personality? British J. Medical Psychology
55, 215-218.


No comments: