**************************************************************************************

The Journal of Social Psychology, 1983, 120, 91-99.

THE BEHAVIORAL VALIDITY OF SOME RECENT MEASURES OF AUTHORITARIANISM*



University of New South Wales, Australia

J. J. RAY AND F. H. LOVEJOY

SUMMARY

The Rigby and Rump attitude to authority scale, the Hogan symbolic authoritarianism measure, the Ray Directiveness scale, and a balanced F scale were administered to a group of people selected and rated by students. An achievement motivation scale was also included because of recent findings suggesting that achievement motivation may be one source of authoritarian behavior. The responses of the 82 Ss showed the five scales as adequate in reliability. The composition of the sample was representative by basic demographic criteria. Ratings of 11 attributes important in the authoritarianism literature were gathered for each S. The highest correlation with rated overall authoritarian behavior was recorded for the achievement motivation scale followed closely by the Ray Directiveness scale. Equivalent correlations for the other three scales were nonsignificant. It was concluded that future research into authoritarianism should restrict itself to the use of behavior inventories as measuring instruments.

A. INTRODUCTION

Although it appears to be a common impression that authoritarianism is not now the popular research topic that it once was in social psychology, books and articles on the topic (S, 6, 10, 11, 12, 14, 31) and new measuring instruments for the construct continue to be published (2, 7, 10, 19, 20, 29). The proliferation of new measuring instruments is at least partly a reflection of the now widely-known inadequacies (3, 19) of the Adorno (1) F scale. One would therefore expect that the new scales would have been subjected to rigorous trials of their adequacy before publication. Unfortunately, this is far from being universally true. While most have undergone some checks, their ability to predict actual authoritarian behavior is unknown. The present paper therefore takes five instruments published from 1970 onwards with a view to assessing their relative power in predicting everyday authoritarian behavior.

One of the five instruments (23) is in fact not overtly an authoritarianism scale at all but is an achievement motivation scale. Its inclusion springs from the often-neglected theoretical point that the connection between attitudes and behavior need not be a direct or obvious one (9). Authoritarian actions may often be the outcome not of authoritarian attitudes but of some entirely different motive. Ray (23) has presented theoretical arguments to show that achievement motivation could be such a motive and has also given some preliminary evidence that his "AO" scale does in fact provide the prediction required.

A second instrument is the Rigby and Rump (28, 29) General Attitude to Institutional Authority (GAIAS) scale in its 32-item short form. This is a very careful inventory of acceptance/rejection of authority in four major areas (the law, the armed forces, the police, and teachers). As such, it is superior to the earlier Ray (16) Attitude to Authority scale which had a preponderance of items concerned with the armed forces.

The third instrument is the Hogan (7, 8) "SF" test of "symbolic authoritarianism." It was included because of the wide and international use it has received. Although Hogan's own summary of research (8) with the scale reveals that it usually is not significantly correlated with the F scale, the poor validity of the F scale itself (19, 33) prevents this from being a necessarily fatal flaw. It does have the attraction of being a nonverbal test, although it could perhaps be more accurately referred to as a measure of preference for simplicity rather than of authoritarianism per se.

The fourth instrument is the Ray (19, 22) "Directiveness" scale in a short form of 14 items. This is a behavior inventory of interpersonal authoritarianism which has in the past shown better predictive validity than attitude scales (19, 25). Its conception of the authoritarian was in fact the same as that which lay at the basis of the Adorno et al. (1) work: "Someone prone to behave as the Nazis did -- in an aggressive, domineering, and destructive way towards other people" (19, p. 307). This is one scale that in the past has undergone extensive tests of its behavioral validity. It was included on the present occasion, therefore, primarily for purposes of comparison.

The fifth scale is a version of the original F scale in which half the items are F originals and half versions of F originals designed to require a "Disagree" answer to yield a high score on authoritarianism. Since the iconoclastic paper by Rorer (30) declaring acquiescent response style to be a "myth," the use of such "balanced" scales may seem unnecessary, but Peabody (15) has pointed out that even Rorer did not deny the potential confounding influence of acquiescence with particularly ambiguous items. Peabody then also showed that the F scale in particular was indeed ambiguous. There are several potential balanced F scales to choose from, but not all are equally attractive. The scale by Christie, Havel, and Seidenberg (4) shows little correlation between its supposedly opposite halves and the Lee and Warr (13) scale proves on closer inspection to have very few items traceable to the F original. The Cherry and Byrne (2) scale was considered, but raw data from one of that scale's administrations supplied by one of its authors revealed on analysis a correlation between the supposedly opposed halves of only .257 for males and .321 for females. The scale used is, therefore, a 14-item short form of the Ray (17, 20) "BF" scale. Even in its short form, this scale has shown a correlation between its two halves (before reversals) of -.5.

B. METHOD

The five scales were combined into a questionnaire administered by members of a small class of sociology students to people they knew, under the sole constraints that nonstudents and people in the humbler occupations were to be preferred as Ss. Both males and females could be interviewed and the venue for the study was the Australian city of Sydney. The students were also told to be sure that they could in fact rate the person chosen on traits having to do with authoritarianism and to seek assistance from third parties who knew the interviewee concerned if any doubt about a rating arose. After the S had completed his questionnaire under guarantees of anonymity, the student-raters also filled out a sheet of 11 ratings of the person. The ratings and the questionnaire thus provided independent sources of information on the S's traits. Peer ratings would generally appear to have the best claim on providing information about an S's everyday or long-term behavior patterns, as distinct from behavior in a laboratory, which is intrinsically of unknown generalizability. All validation studies tend to have particular problems with sampling and there can be no pretense that the present Ss form a randomly selected sample. The constraints of preferring nonstudents in the humbler occupations, however, were designed to counteract the usual middle-class bias of student-selected S groups and it may be noted that such constraints have in the past succeeded (19, 25) in producing a "sample" that was in fact representative in terms of basic demographic criteria. Each student-interviewer on the present occasion obtained an average of just over five people each to produce a final N of 82 for the study.

The demographic composition of the sample was surprisingly representative under the circumstances. The mean age was 32.56 years (SD 13.79), the mean score on sex (scored 1 = male; 2 = female) was 1.48 (SD .50), the mean score on occupation (scored 1 = manual; 2 = nonmanual) was 1.76 (SD .42) and the mean score on education (scored 1 = primary; 2 = half-secondary; 3 = full secondary; 4 = tertiary) was 3.06 (SD .85). These figures compare with 35.19 (SD 16.42), 1.46 (.50), 1.71 (.47), and 2.60 (.96) in a recent doorstep study (N = 102) using random cluster sampling carried out in the Sydney metropolitan area (24).

C. RESULTS

The basic findings are given in Table 1. All scales were shown to have some validity -- but in different areas -- and to have satisfactory reliability. The alphas were as follows: Directiveness .80, AO .79, BF .78, GAIAS .93, and Hogan SF .76. The four subscales of the GAIAS also held up well, with the following reliabilities: .87 (Police), .89 (Army), .77 (Law), and .80 (Teachers). All subscales of the GAIAS were also highly correlated with one another. Their intercorrelations ranged from a low of .53 to a high of .65. This finding supports the notion of attitude toward conventional authority as being a general trait. The Directiveness scale correlated .40 with the AO scale, .05 with the BF scale, and -.004 with the GAIAS; hence those reporting themselves as behaving in an authoritarian way were also ambitious but did not show any special respect for authority. The correlation of .38 between the Hogan and BF scales suggests that high scorers on the F scale did indeed tend to favor cognitive simplicity.

The above results were obtained with the attitude scales being scored 5 to 1 for each answer (from "Strongly Agree" to "Strongly Disagree") and with the personality scales being scored 3 to 1 for each answer (from "Yes" to "No"). The Hogan scale answers were scored "1" (asymmetrical) or "2" (symmetrical). The response options for the peer-ratings were in each case "Very true" (scored 5), "True" (4), "?" (3), "False" (2), and "Very False" (1).

The only significant correlations between the demographic variables and the scales were correlations of .28 and .35 between age and the BF and GAIAS scales respectively, of .32 and -.27 between education and the AO and BF scales respectively, and .33 between the AO scale and occupation.

TABLE 1

THE SIGNIFICANT (p < .05) CORRELATIONS BETWEEN FIVE SCALES AND 11 RATINGS

Ratings....................................Ray Dir....... n-Ach......BF....GAIAS..........Hogan

Behavior:
This person tends to be:

Dominant over others................. .50............. .25
Aggressive................................. .46
Authoritarian: i.e., desires or
tends to impose own will on
others......................................... .32............ .34
Submissive to direction from
others........................................ -.34
Orderly in habits

Attitude:
This person:

Is respectful of authority............................................ .31......... .33
Is ambitious................................ .29............ .38
Tends to dislike immigrants
Tends to dislike Jews ................. .25......................... .22
Tends to dislike Aborigines......... .25......................... .22........................... .22
Tends to be rigid in thinking........ .23............ .22....... .31........................... .23

Scale means:.............................31.64.......... 31.69.....38.63.......93.98........24.32
Scale SDs....................................6.08..............6.34.......8.74......20.42..........3.91


D. DISCUSSION

The most surprising result in Table 1 is the finding that the best prediction of everyday authoritarian behavior was provided by a fairly conventional scale of achievement orientation. This is certainly vindication of the clearest kind for the theory put forward elsewhere (23) to the effect that achievement motivation may be one of the main motives behind authoritarianism. The finding serves to re-emphasize the complexity of the relationships between attitudes, behavior, and personality.

In terms of its ability to predict a wide range of authoritarianism-related behaviors it was clear that the Directiveness scale, however, has no near rivals. It was by far the best predictor of straight dominance and straight aggressiveness and was the only scale to provide any significant (inverse) prediction of submissiveness. Its prediction of rated authoritarianism was second only to the achievement orientation scale and it also predicted rigidity and racial prejudice. This latter finding would seem to be in some conflict with previous findings to the effect that Directiveness score and racial prejudice are unrelated (19, 26). It must be noted, however, that previous correlations have been with self-reported prejudice, not peer-rated prejudice. There may be some tendency to regard the peer-report as more trustworthy because of the lesser temptation of the rater to present the person in a socially desirable light, but this view could only be accepted with severe reservations. It could, for instance, also be that dominant people are more likely to be perceived as having negative views towards others even though they may not in fact do so. In a validation study one is trying to show that two types and sources of information converge, which does not necessarily mean that one of those information types is superior or more trustworthy in general. The attraction of "peer" ratings lies mainly in the potential they have for providing information about customary behavior, the main focus of the present study. Although they may also provide information about attitudes, they would appear to be limited in this regard by the fact that whereas behavior is in principle publicly observable, attitudes are not directly publicly observable and must be inferred.

The finding that the attitude scales predicted ratings of authoritarian attitudes only and that the personality scales predicted only ratings of authoritarian behavior confirms previous findings (19) about the lack of a direct relationship between attitudes and behavior in this area, as does the finding that the Directiveness scale correlates with neither of the attitude scales. This now brings to three the number of attitude scales of authoritarianism with which the Directiveness scale has been shown not to correlate (19). It may also be worth noting that ratings of respect for authority also failed to correlate with ratings of authoritarianism (r = .05; n. s.).

As was expected, the Hogan scale proved to be a predictor of rigidity rather than of authoritarianism as such. Even the F scale version did at least predict rated authoritarian attitudes if it did not predict rated authoritarian behavior. The Hogan scale predicted neither. Since cognitive rigidity however, has been a topic of considerable importance in the field of authoritarianism research, the demonstration that the Hogan scale has some validity as a measure of this may mean that much of the existing research with the scale may need only slight reinterpretation. The fact that the scale also predicts dislike of blacks does tend to confirm the view that rigidity is indeed a factor in racial attitudes. Some support is thus provided for at least one aspect of the Adorno et al. (1) theory.

Two ways in which the present results with the Hogan scale differ somewhat from the general run of results so far reported with that scale (8) are that the reliability was unusually low and the correlation with the F scale was unusually high. There was one occasion when a much higher correlation between the Hogan scale and the F scale was found but that was in the initial scale construction study where Hogan gave the two scales to his psychology classmates to answer. The opportunities on that occasion for a Rosenthal-type effect hardly need comment. The present study, then, serves to provide some independent validation for the Hogan scale of a sort that Hogan at least seems to regard as fairly central. What the BF scale itself measures, however, will be further discussed below. The lower reliability observed for the Hogan scale in the present work would appear to be no more than what is usually found when a scale developed on student samples is applied to a more heterogeneous population.

The Rigby and Rump (29) GAIAS, by contrast, did nothing to support any theory. It was shown to be valid as a measure of what it purports to measure -- respectful attitudes towards conventional institutional authority -- but had no relationship to authoritarian behavior or any associated construct. This negative result may, however, be quite important: it suggests that the central element (respect for authority) in the purportedly covarying constellation of traits central to the Adorno et al. (1) theory shows no tendency to covary at all when it is measured by a more straightforward scale than the F scale. Perhaps, therefore, it is only the in-built biases of the F scale that have caused respect for authority in the past to appear as the keypoint of the "rogues gallery" of attributes that Adorno et al. described as constituting "authoritarianism."

The result with the GAIAS does tend to limit the inferences to be made from the results obtained with the balanced F scale. At first sight, the finding that it is valid as a predictor of respect for authority and that it does predict antisemitism, dislike of blacks, and cognitive rigidity might seem quite good support for the Adorno et al. (1) account of things. In the circumstances, however, we must make the more cautious inference that it predicts these things only because it was specifically built to do so. It was designed to predict not only acceptance of authority but also acceptance of authority of a suspicious and fearful sort. The present results suggest that this aim was achieved, but at the same time they limit the significance of the correlations observed for purposes of theory. It has been shown that the correlates of the F scale are not also the correlates of respect for authority in general.

How are the present results related to the contention that the F scale measures nothing more than simple, traditional political conservatism (18, 22, 32)? Do not they vindicate it as being at least a measure of authoritarian attitudes? How indeed do they relate to another contention that authoritarianism is a personality variable only and does not exist as an attitude dimension (27)? The answer to these questions lies in the extensive demonstration given elsewhere (18) that respect for conventional sources of authority is a central part of traditional conservative ideology. A conservatism scale should predict respect for authority. The contention then is not that authoritarian attitudes do not exist but rather that they do not exist separately from conservatism. If this seems like something of an indictment of conservatives, we must also recollect the other now well-demonstrated fact that such attitudes have nothing to do with actual authoritarian behavior. Authoritarian personalities (i.e., behavior dispositions) are equally likely to be found on both sides of politics (21).

E. CONCLUSIONS

The two attitude scales and the Hogan symbolic measure were found to be not valid as predictors of authoritarian behavior. Two personality scales -- generically a dominance scale and an achievement motivation scale, respectively -- were found however to be valid by the behavioral criterion of peer-rated authoritarianism. If it is deeds, not words that concern us, then, it would seem that future research into authoritarianism should confine itself to the use of personality scales (i.e., behavior inventories) and that continued use of the California F scale in this connection will inevitably tend to be misleading.

REFERENCES


{Articles below by J.J. Ray can generally be accessed simply by clicking on the name of the article. I am however also gradually putting online a lot of abstracts, extracts and summaries from older articles by other authors so if an article not highlighted below seems of particular interest, clicking here or here might just save you a trip to the library}


1. ADORNO, T. W., FRENKEL-BRUNSWIK, E., LEVINSON, D. J., & SANFORD, R. N. The Authoritarian Personality. New York: Harper, 1950.

2. CHERRY, F., & BYRNE, D. Authoritarianism. In T. Blass (Ed.), Personality Variables in Social Behavior. Hillsdale, N.J.: Erlbaum, 1977.

3. CHRISTIE, R., & JAHODA, M. Studies in the Scope and Method of "The Authoritarian Personality." Glencoe, Ill.: Free Press, 1954.

4. CHRISTIE, R., HAVEL, J., & SEIDENBERG, B. Is the F scale irreversible? J. Abn. SOC: Psychol., 1956, 56, 141-158.

5. HARTMANN, P. A perspective on the study of social attitudes. European J. Soc. Psychol., 1977, 7, 85-96.

6. HEAVEN, P. C. L. Authoritarianism: South African Studies. Bloemfontein, South Africa: DeVilliers, 1980.

7. HOGAN, H. W. Reliability and convergent validity of a symbolic test for authoritarianism. J. of Psychol., 1970, 76, 39-43.

3. HOGAN, H. W. A review of research experiences with a symbolic measure of authoritarianism. In V. K. Kool & J. J. Ray (Eds.), Authoritarianism Across Cultures. Bombay, India: Himalaya, 1982.

9. KELMAN, H. C. Attitudes are alive and well and gainfully employed in the sphere of action. Amer. Psychol., 1974, 29, 310-324.

10. KOOL, V. K. Measures of Authoritarianism and Hostility. Bombay, India: Himalaya, 1980.

11. KOOL, V. K., & Ray, J. J., Eds. Authoritarianism Across Cultures. Bombay, India: Himalaya, 1982.

12. KREML, W. P. The Anti-Authoritarian Personality. Oxford, England: Pergamon, 1977.

13. LEE, R. E., & WARR, P. B. The development and standardization of a balanced F scale. J. Gen. Psychol., 1969, 81, 109-129.

14. MCKINNEY, D. W. The Authoritarian Personality Studies. The Hague, Netherlands: Mouton, 1973.

15. PEABODY, D. Authoritarianism scales and response bias. Psychol. Bull., 1966, 65, 11-23.

16. RAY, J.J. (1971) An "Attitude to Authority" scale. Australian Psychologist, 6, 31-50.

17. RAY, J.J. (1972) A new balanced F scale -- And its relation to social class. Australian Psychologist 7, 155-166.

18. RAY, J.J. (1973) Conservatism, authoritarianism and related variables: A review and an empirical study. Ch. 2 in: G.D. Wilson (Ed.) The psychology of conservatism London: Academic Press.

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

20. RAY, J.J. (1979) A short balanced F scale. Journal of Social Psychology, 109, 309-310.

21. RAY, J.J. (1979) Does authoritarianism of personality go with conservatism? Australian Journal of Psychology 31, 9-14.

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

23. RAY, J.J. (1980) Achievement motivation as an explanation of authoritarian behaviour: Data from Australia, South Africa California, England and Scotland. Chapter in: P.C.L. Heaven (Ed.) Authoritarianism: South African studies Bloemfontein: De Villiers.

24. RAY, J.J. (1981) The new Australian nationalism. Quadrant, 25(1-2), 60-62.

25. RAY, J.J. (1981) Authoritarianism, dominance and assertiveness. Journal of Personality Assessment 45, 390-397.

26. RAY, J.J. (1981) Explaining Australian attitudes towards Aborigines Ethnic & Racial Studies 4, 348-352.

27. RAY, J.J. (1982) Authoritarianism/libertarianism as the second dimension of social attitudes. Journal of Social Psychology, 117, 33-44.

28. RIGBY, K. A concise scale for the measurement of attitudes towards institutional authority. Austral. J. Psychol., 1981, in press.

29. RIGBY, K., Rz Rump, E. E. The generality of attitude to authority. Hum. Relat., 1979, 32, 469-487.

30. RORER, L. G. The great response-style myth. Psychol. Bull., 1965, 63, 129-156.

31. SUTHERLAND, S. L., & TANENBAUM, E. J. Submissive authoritarians: Need we fear the fearful toadie? Can. Rev. Sociol. & Anthrop., 1980, 17, 1-23.

32. SUZIEDELIS, A., & LORR, M. Conservative attitudes and authoritarian values. J. of Psychol., 1973, 83, 287-294.

33. TITUS, H. E. F scale validity considered against peer nomination criteria. Psychol. Rec., 1968, 18, 395-403.

The University of New South Wales P. 0. Box 1, Kensington, NSW, Australia 2033

* Received in the Editorial Office, Provincetown, Massachusetts, on June 8, 1982. Copyright, 1983, by The Journal Press.



POST-PUBLICATION ADDRENDUM

I should probably draw attention to the fact that some more of the NON-relationships reported above may be of interest. The fact, for instance, that neither occupation nor education correlated with the Directiveness scale, the GAIAS or the Hogan scale would seem to be a rather severe blow to the "Working class authoritarianism" theory.




Go to Index page for this site

Go to John Ray's "Tongue Tied" blog (Backup here)
Go to John Ray's "Dissecting Leftism" blog (Backup here)
Go to John Ray's "Australian Politics" blog (Backup here)
Go to John Ray's "Gun Watch" blog (Backup here)
Go to John Ray's "Education Watch" blog (Backup here)
Go to John Ray's "Socialized Medicine" blog (Backup here)
Go to John Ray's "Political Correctness Watch" blog (Backup here)
Go to John Ray's "Greenie Watch" blog (Backup here)
Go to John Ray's "Leftists as Elitists" blog (Not now regularly updated)
Go to John Ray's "Marx & Engels in their own words" blog (Not now regularly updated)
Go to John Ray's "A scripture blog" (Not now regularly updated)
Go to John Ray's recipe blog (Not now regularly updated -- Backup here)

Go to John Ray's Main academic menu
Go to Menu of recent writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page (Backup here)
Go to Selected pictures from John Ray's blogs (Backup here)
Go to Another picture page (Best with broadband)