Thursday, August 18, 2005

British Journal of Medical Psychology, (1982), 55, 215-218.


By J. J. Ray and L. A. Simons

Clients at the Sydney Coronary Heart Disease Prevention Programme were screened for actual CHD and sufferers were compared with non-sufferers an four personality scales to measure respectively: A--B. dominance, achievement motivation and 'Freneticism'. There were 112 sufferers and 201 controls. Sufferers were found to have significantly higher scores an dominance -- the Ray (1976) Directiveness scale -- but also to have significantly lower scores on the A--B measure. This latter reversal of the usual relationship was an artifact of the fact that older people are both more CHD prone and get lower A-B scores. When age was controlled for there was no relationship between A-B type and CHD. This left the authoritarian style of dominance measured by the Directiveness scale as the sole predictor of CHD. This was held to be a belated vindication of claims made in the pioneering work of Dunbar (1943)

In their recent comprehensive review article, Matteson & Ivancevich (1980) report a strong preponderance of studies confirming that the A-B personality type is implicated in coronary heart disease. They also find several lacunae in the research to date. One of their concerns is that the nature and identity of the A-B construct have been insufficiently identified. In particular, how is it related to conventional psychological personality tests? Since their paper went to press a series of four studies addressed to just this question has been reported (Ray & Bozek, 1980). In this paper the Jenkins Activity Survey (one of the main measures of the A-B construct) was subjected to the sort of psychometric refinement usual in psychological test construction. The correlations of each item in the scale with the total score on the scale were calculated and only those items showing the highest correlation were retained -- thus leading to an improvement in scale reliability. It was also found that the items of this shortened scale clustered very clearly with items from two conventional psychological personality scales -- measuring respectively dominance and motivation to achievement. There was also a third unique component in the JAS tentatively labelled 'Freneticism' (from the adjective 'frenetic'). The question as to which of these three components might be providing the prediction of CHD (coronary heart disease) was left open. Are coronary-prone people dominant, ambitious, frenetic, or all three together? The present study was devised in the hope of throwing some light on this question.

One of the difficulties in ascertaining whether CHD victims are high on some attribute is to specify in relation to whom they may be considered high. In theory, of course, one wishes to compare them with the population norm, but this is a difficult parameter to derive. We cannot examine the whole population so we must use a sample of it. Yet most conventional sample types contain serious biases. Other hospital patients have some type of illness. People who respond to public calls to come forward, people who return mail questionnaires and people who cooperate with doorstep interviewers, are all minorities who might be quite unrepresentative of their less cooperative brethren. Particularly when it comes to questions measuring things like freneticism, dominance and ambition, volunteers must almost be expected to differ from non-volunteers.

In these difficult circumstances, the most common strategy adopted in past published research in this area has been to take same large available group that is hopefully not too unlike the general population and dissect it into those who have had CHD and those who have not. The non-victims can then act as a control group with which to compare the victims. It is then this pattern into which the present research falls. We examine not whether the CHD victims are high on some variable in relation to the general population but, rather whether they are high on the variable in relation to their non-victim controls. Such a study need not of course be retrospective but, as Matteson & lvancevich (1950) point out, most such studies have in fact been retrospective.


The present work took advantage of the fact that there exists in Sydney, Australia, a voluntary public programme of screening for CHD risk factors which annually attracts around 10000 Australians from all walks of life. Details of the programme are available elsewhere (Simons & Jones, 1978). A small minority of screenees already have CHD. There is thus avaiIable a useful experimental and control-group situation.

A questionnaire was prepared which contained four personality self-report scales: the Ray ( 1976, 1980) 14-item Directiveness scale -- one of the dominance scales used in previously reported research (Ray & Bozek. 1980), the Ray (1979a) AO scale -- a 14-item scale also as used previously; the 10-item 'Freneticism' scale, produced and reported on as the third JAS component in Ray & Bozek (1980), and an 18-item short form of the JAS itself. This short form was a further reduction of the 24-item short form reported in Ray & Bozek (1980) based on the item analysis from the third study there reported. The further reduction was designed to increase reliability by deleting less central items. It may be noted that it is quite usual for only a small number of the full set of JAS items to be scored (Matteson & Ivancevich, 1980).

This questionnaire was administered to a standardization sample of the first two weeks' attenders at clinic and any actual CHD victims separated out. This gave a control group sample of 201 people. In subsequent weeks only CHD patients completed the questionnaire. Over a 12-month period a total of 112 questionnaires from actual CHD sufferers were accumulated.

Confirmation of the diagnosis of CHD (previous myocardial infarction or angina pectoris) was made by history and by assessment of the Rose & Blackburn (1968) Cardio-vascular Questionnaire.


Means and standard deviations for the two patient groups were calculated for each of the four scales. The means of only two scales differed significantly across the two groups. It was found that the CHD sufferers had significantly lower scores on the modified JAS (t = 2.29; p < .05) and significantly higher scores on the Directiveness scale (t = 3.05; p < 0.01). As in previous research it was also found that the JAS correlated significantly with scores on the Directiveness scale: 0.463 among the CHD cases and 0.358 among the controls.

The finding that the JAS correlates with CHD in the opposite direction to that usually reported seems at first odd but is in fact nothing more than an artifact of the correlation of the scale with age -- the one variable that does without dispute increase one's CHD risk. This is most clearly seen if the findings are re-expressed in terms of correlations -- with CHD incidence scored as a 'dummy' variable (0 for being in the control group; and 1 for being in the victim group). Among the 313 subjects we then find that the correlations with CHD are 0.173 for Directiveness, -.130 for the JAS and 0.488 for age. Age in turn correlates 0.013 with Directiveness and -.263 with the JAS. Age then is far and away the main variable predictive of CHD and, as one gets older, one gets lower scores on the JAS (i.e. one slows down). We must then control for age if we are to examine the true impact of the A-B variable alone. This is most easily done by means of partial correlation. When this is done we find that the correlation of the JAS with CHD drops to --0.002 -- totally non-significant. Directiveness score, on the other hand, is unaffected by age.


Although generically a dominance scale, the Directiveness scale was in fact written to measure the authoritarian style of dominance (Ray, 1976). It has in fact been used in a wide variety of studies of authoritarianism (e.g. Heaven, 1980; Ray, 1980). It differs from the California 'F' scale (Adorno et al., 1950) in that it is a good predictor of actual authoritarian behaviour (Ray. 1976). It is perhaps ironical that although Adorno et al in their pioneering work termed authoritarianism a disease, it was in fact a form of mental abnormality they had in mind. Just when this allegation has been fairly convincingly laid to rest (Masling, 1954: Roberts & Jessor, 1958; Elms, 1970: Richek et al., 1970: Crabbe. 1974; Ray. 1979b) it turns out that authoritarianism is in fact implicated in illness after all -- but illness of a physical kind.

It is not hard to surmise why authoritarianism might lead to heart disease. A life devoted to imposing one's own will on others must be one filled with all kinds of
stresses -- particularly emotional. Since the heart is automatically affected by many kinds of emotional stress, it is plausible to see it as just simply harder worked among authoritarians.

The failure of the JAS or any of its other 'components' to predict CHD is not of course new. In fact many researchers outside the Jenkins--Rosenman--Friedman-Zyzanski group report what Rime & Bonami (1979) politely call 'inconsistent' results in studies of the personality correlates of CHD. Where the American group reports CHD sufferers as very active, European studies in particular show some tendency to find precisely the opposite -- that CHD victims are particularly passive. Rime & Bonami have summarized some of the findings of this type and from their own work conclude that CHD victims are passive, dependent and ego-defensive. Siltanen et al. (1975) report similar findings for their group of 'mild' CHD sufferers.

The present findings do of course look as if they fit precisely in this 'European' mould if we look at the correlation between the JAS and CHD before controls for age are applied. This may suggest that the 'European' pattern of results is an artifact of failure to control for age also and that the real picture is simply one of no relationship.

In conclusion, then, it would appear that the JAS and the construct it measures (A-B) have been far too general and inclusive, and that where positive results have in the past been found it has been due to only one of the many components embedded in the larger measure -- and that component is authoritarianism. This does in fact constitute something of a belated vindication of the work of Dunbar (1943). In her pioneering work, Dunbar described the coronary personality as compulsive, dominating and aggressive. As Matteson & Ivancevich (1980) point out, this view has tended to lack current acceptance because of a variety of methodological problems in the original work. The present confirmation of Dunbar's theories from a quite different starting-point and a quite independent source must then be seen as giving them fresh claims to careful consideration. At the very least, the Directiveness scale would seem to have been given a strong claim for inclusion in future research into the personality precursors of CHD.


The authors would like to thank Robert Bozek for assistance with the earlier stages of this project.


Adorno,T.W., Frenkel-Brunswik, E., Levinson, D.J. & Sanford, R.N. (1950). The authoritarian personality New York: Harper.

Crabbe, B.D. (1974) Are authoritarians sick? In J.J. Ray (Ed.) Conservatism as heresy. Sydney: ANZ Book Co.

Dunbar, F. (1943) Psychosomatic diagnosis. N.Y.: Hoeber Press.

Elms, A.C. (1970) Those little old ladies in tennis shoes are no nuttier than anyone else, it turns out. Psychology Today 3, 27-59.

Heaven, P.C.L. (1980) Authoritarianism, prejudice and alienation among Afrikaners. J. Social Psychology, 110, 39-42.

Masling, M. (1954) How neurotic is the authoritarian? J. Abnormal & Social Psychology 49, 316-318.

Matteson, M.T. & Ivancevich, J.M. (1980) The coronary-prone behavior pattern: A review and appraisal. Social Science & Medicine 14a, 337-351

Ray, J.J. (1976) Do authoritarians hold authoritarian attitudes? Human Relations, 29, 307-325.

Ray, J.J. (1979) A quick measure of achievement motivation -- validated in Australia and reliable in Britain and South Africa. Australian Psychologist 14, 337-344.

Ray, J.J. (1979) The authoritarian as measured by a personality scale Solid citizen or misfit? J. Clinical Psychology 35, 744-746.

Ray, J.J. (1980) Authoritarianism in California 30 years later -- with some cross-cultural comparisons. Journal of Social Psychology, 111, 9-17.

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

Richek, H. J., Mayo, C.D. & Puryear, H. B. (1970) Dogmatism, religiosity and mental health in college students. Mental Hygeine, 54, 572-574.

Rime, B. & Bonami, M. (1979) Overt and covert personality traits associated with coronary heart disease. British J. Medical Psychology 52, 77-84

Roberts, A. H. & Jessor, R. (1958) Authoritarianism, punitiveness and perceived social status. Journal of Abnormal and Social Psychology, 56, 311- 314.

Rose, G. A. & Blackburn, H. (1968). Cardiovascular Survey Methods. Geneva: WHO Monograph Series. no. 56.

Siltanen, P., Lauroma, M., Nirrko, O., Punsar, S., Pyorala, K., Tuominen, H. & Vanhala, K. (1975) Psychological characteristics related to coronary heart disease. J. Psychosomatic Research 19, 183-195.

Simons, L. A. & Jones, A. S. (1978). Coronary risk factor screening and long-term follow-up. Year One of the Sydney Coronary Heart Disease Prevention Programme. Medical Journal of Australia, 2, 455-- 455.


No comments: