Australian & New Zealand J. Sociology, 1972, 8 (June), 96-102.
JOHN J. RAY*
The great body of work in political sociology has received no more notable contribution from social psychology writers than the work of Adorno et al. (1950) -- in which the California 'F' scale was presented. This scale has become the standard measure of authoritarianism and the relationship of scores on this scale to social class and political preference has been widely noted. See especially Lipset (1960) and Brown (1964). Nonetheless the adequacy of this scale has been widely questioned. Not the least problem is the fact that scores on it are almost totally unrelated to authoritarian behaviour (Titus, 1968). A second problem is the one-way wording of the scale -- meaning that acquiescent people automatically get high scores. Any observed relationship (such as the one with social class quoted above) can hence always be attributed at least as much to the characteristics of acquiescent people as to the characteristics of authoritarian people. This latter problem appears now to have been solved by the successful production of a balanced version of the scale (Lee & Warr, I969). The new version however, retains the earlier conceptualisation of authoritarianism given in Adorno et al. (1950) and hence fails to discard an approach which is known to have had only very little success in predicting actual behaviour (cf. also Kirscht and Dillehay, 1968).
In his discussion of this failure, Titus (1968) speaks of 'the classical conception of authoritarianism' and notes that the 'F' scale cannot be taken as measuring this. What he means by the 'classical conception' is presumably that which was recorded in the Oxford dictionary long before the work of Adorno et al. (1950) loaded the term "authoritarian" with psychodynamic implications and correlates. The Oxford definition is 'Favourable to the principle of authority as opposed to that of individual freedom'. That the 'F' scale should measure something different from this conception should not, however, be surprising. Its authors did stress from the beginning that it was intended to be an indirect measure and indeed a measure of "pre-Fascist' (i.e. right-wing and ethnocentric) attitudes rather than of authoritarianism per se. In fact the central notion behind their use of the word 'authority' had to do with a neo-Freudian concept of authority in the family and the projection of this upon the political scene of those times -- not with authority as more usually defined.
This being so, there is a clear doubt about how authoritarians in the general sense might be expected to behave. With the conceptual restrictions and predictive failure of the 'F' scale, a new instrument seems called for. With a new instrument we could re-examine the relationship between authoritarianism and ethnocentrism. It could be that the well-known correlation between ethnocentrism and 'F' scale scores is attributable to the paranoia and other psychopathological tendencies
deliberately built into the 'F' scale by its authors. It does not follow that classical authoritarianism will correlate similarly with ethnocentrism. Perhaps it is only the peculiarly psychopathological brand of authoritarianism promoted by Adorno et al. (1950) which is related to ethnocentrism.
The items for the new `A' scale were written with the characteristics in mind that might reasonably be expected of a person well adapted in an authoritarian environment (such as the army). These may be described in terms of a short list of costs and gains:
1. Restrictions of liberty.....................1. Clarity of role and custom definitions
2. Lack of participation in deci- .........2. Release from individual respon-
3. Acceptance of aggression..............3. Acceptance of aggression
towards the self .......................................toward others
4. Flexibility.........................................4. Order
The typical authoritarian, then, is expected to accept and justify the costs while affirming the desirability of the gains. The anti-authoritarian would do the opposite. Formally, then, the authoritarianism that the 'A' scale aspires to measure is defined as: A desire for a form of social organisation similar to military institutions and procedures - with the restriction of liberty, lack of participation in decision making, lack of individual responsibility, acceptance of aggression and clarity of role definitions that this implies. It thus includes, but is not limited to, simple 'attitude to authority'.
A pool of 43 items was administered to a group of 114 first year psychology students at the University of Sydney, along with the Eysenck (1954) 'R' scale, the Rokeach (1960) 'D' scale and the California 'F' scale. Political party preference was also ascertained.
The product-moment correlations of all items with the total score on the scale were calculated and all items showing a correlation (uncorrected) below .300 were arbitrarily dropped. This left a scale naturally balanced, with 12 positively scored items and 12 negatively scored (see Table 1). Its reliability ('alpha') was .85. See Cronbach (1951) for the rationale underlying this statistic. Split-half reliability is a special case of 'alpha'. The correlation between positive and negative halves of the scale was -.609.
The items of the 'A' scale -- scored from 5 (St. Agree ) to 1 (St. Disagree). ('R' = reverse scored). Conscript sample mean was 70.30 with an S.D. of 9.04.
1 Human beings are more important than efficiency. R
2 If the army allowed more room for individuality it might be a better
3 People should be guided more by their feelings and less by the rules. R
4 In the army soldiers should not obey an order if it is obviously morally
5 Patriotism is just a glorified name for national selfishness. R
6 Individual freedom is a basic human right. R
7 To be efficient is bad if you hurt people in doing it. R
8 At the moment it is more important to build good roads than to increase pensions.
9 When Mussolini made Italy's trains run on time, that at least was an important achievement.
10 There is something wrong with anybody who likes to wear military uniform.
11 People should be made to be punctual.
12 Schoolchildren should have plenty of discipline.
13 The only way Australia will win respect in Asia is by building up strong armed forces.
14 There is always a better alternative for nations to take rather than going to war.
15 Children should be seen and not heard.
16 Before you give children freedom you have to be careful that they will not abuse it.
17 One of the first things children should be taught is to obey their parents at all times.
18 The Armed Forces are largely unnecessary.
19 It is a duty to bring your children up properly.
20 Children should not answer back to their parents.
21 There is far too much regimentation of people nowadays.
22 War is a purifying force.
23 You know where you're going when you have an order to obey.
24 People should not be expected to conform as much as they are today.
Some important validation evidence is available via the correlations with this scale of variables included in the original study. The 'A' scale correlated .733 with Radicalism, .362 with Dogmatism, .664 with 'F' scale score and .432 with conservatism of political party choice.
Of more particular relevance here, however, is the question: What behaviours does it correlate with? There seemed to be two types of behaviour measure that might be of importance in answering this question: direct measurement of actual behaviour and rating of habitual behaviour by peers. The latter, of course, is the method championed by Titus (1968).
Instances of each of these two measures were obtained in an intensive study of fifty-two second-year psychology students at Macquarie University. The reliability of the scale on this sample was .87.
In this study Kelly's repertory grid technique (Bannister, 1966) was employed in an adaptation for group testing. A list of the following twelve roles and institutions was put up on the blackboard: The army, your father, your mother, yourself, children, the lecturer you like most, policemen, university students, members of the opposite sex, politicians, young Liberals, and labourers. Kelly's normal procedure was then described to the subjects and they were each instructed to act as their own testers until they had obtained ten constructs which they found personally most useful in discriminating between the twelve roles. If they could not think of ten constructs they were allowed to seek suggestions from their neighbours. When this was done, each subject filled out a grid wherein each role was rated for the presence (1) or absence (0) of the characteristic named by each construct. Separate principal components analyses were then conducted on each grid handed in. In line with Kelly's suggestions, the size of the largest eigenvalue of each matrix was taken as the subject's directly measured cognitive complexity score (the person who has the largest resource of independent constructs gets the lowest first eigenvalue). This statistic is a very common factor analytic index of the degree to which any set of variables go together. As the authoritarian is expected to have a desire for simplicity and order, we might then expect a correlation between the 'A' scale score and this direct measure of how much people do in fact impose simplicity and order on their conceptual worlds.
With forty subjects, the correlation obtained was -.401, which is highly significant but in the direction opposite to that expected. Does this mean that authoritarians are more cognitively flexible? No. A look back to the roles required to be rated reveals a simpler explanation. The roles contain a heavy preponderance of authority figures and institutions such as 'the army', 'policemen', 'your father' and 'the lecturer you like most'. Occurring in such a context, the authority aspects of other roles such as 'your mother' and 'politicians' were also probably very salient.
What has been demonstrated, then, is that authoritarian people show a finer discrimination between various authority figures and roles than do others. They do not just lump all examples of authority into one basket with some label generally. meaning 'bad'. In this light, the correlation obtained does represent clear validation for the scale.
The second sort of validation sought on this sample was via peer ratings. Each person was given three rating sheets and told to write on each his name or identification. He was then instructed to hand them to the three people in the class who knew him best. These people then completed the ratings and handed them back to the experimenter directly. The ratings were on six items of behaviour as follows:
1. Always modifies his behaviour to fit the circumstances of the situation;
2. Is characterised by fixed opinions;
3. Tends to follow instructions without critical thought;
4. Likes to push others around;
5. Is inclined to trust the motives of others;
6. Does not have to be told what to do all the time.
There were seven response options from 'very true' to 'very false'.
The mean score over the three raters was taken for each item and correlated with the 'A' scale scores. Only two were significantly related to the 'A' scale: Nos. 1 and 3 - which correlated .469 and .333 respectively. In other words, high 'A' scale scorers were seen as: 'Always modifying their behaviour to fit the circumstances of the situation' and 'tending to follow instructions without critical thought'.
The second correlation needs little comment other than noting that, as in Titus' (1968) findings with the 'F' scale, the 'A' scale seems here to be sensitive to authoritarian submission. The first correlation, however, was with an item originally intended to tap one of the opposites of authoritarianism, i.e. flexibility. It appears not to have been successful in tapping this and the submissive aspects of the item again would appear to have been more salient.
While the above does represent some validation for the 'A' scale, it is of course of greatest interest for the comparison it makes possible between this work and the work of Titus (1968) with the 'F' scale. It has in fact independently been shown here that, using an entirely different scale, authoritarianism has submissiveness as its most important behavioural correlate. In other words, even when the defects of the 'F' scale are removed (i.e. the one-way wording and the indirect nature of the items) the findings of Titus still stand.
Attitude to Innovation
Yet a third set of data was gathered from this sample. This time the question being investigated was the much touted relationship between authoritarianism and conservatism (Adorno et al., 1950). It was felt that an attempt might profitably be made to tap the basic aspect of conservatism, i.e. the level closest to personality. This was seen as particularly important among students, whose attitudes to particular topical issues might be more in a state of flux than at later times in their life.
The basic element of conservatism, then, was taken to be attitude to innovation (Lentz, 1935) and a set of thirty items were written to tap this construct . These items were subjected to scale analysis and a twenty-two-item scale with a reliability of .74 was produced (eight negative and fourteen positive items) (see Table 2). This was termed the 'I' scale and it was hypothesised that the authoritarian would be characteristically opposed to innovation.
The items of the 'I' scale (Attitude to Innovation). ('R' = reverse scored)
1 If society is to progress, newer solutions to problems are essential. R
2. I would rather visit Europe than America.
3 People in general were much happier in former times than they are
4 I enjoy modern music better than music written years ago. R
5 The people in this community must continually look for new solutions to problems rather than be satisfied with things as they are. R
6 Steam trains are just as good as diesels.
7 I would like to spend some time going back over all the places I knew as a child.
8 People are so busy keeping up with the Joneses nowadays that they have forgotten how to live.
9 Even if the newer ways conflict with the way things were done in the past, they are absolutely necessary and desirable. R
10 Changes are desirable even if they do not seem to contribute as much as one might expect.
11 Australia's ties with Great Britain should not be reduced.
12 I prefer old masters to modern art.
13 Changes in political policy should always be gradual.
14 We are better off today than our parents were. R
15 The churches are out-of-date. R
16 The best buildings are those whose architecture is functional and without much ornamentation. R
17 The woman's place is in the home.
18 Thinkers in ancient civilisations probably knew more than we do on many important topics.
19 A community should not accept programs which upset the settled way of doing things.
20 People who are dissatisfied with the way things have been done forget that doing things in a new way may bring about even worse conditions.
21 The most reasonable approach toward social development is to accept changes which do not substantially alter the established order.
22. While changes are desirable, they should never be implemented at the cost of our past values and traditions.
The correlation obtained was .335, which is significant and in the expected direction. Validation for the 'I' scale itself was found in its significant correlation of .379 With conservatism of political party preference. The 'A' scale itself correlated .439 with political party preference.
The Relationship with Ethnocentrism
Although the 'A' scale had been designed primarily for use with student samples it was felt that the important topic of its relationship with ethnocentrism warranted a sample more nearly representative of the general population. To this end, the 'A' scale was included in a larger study employing Australian National Servicemen as subjects . Like all samples, the National Service intakes clearly do fall short of complete representativeness. They are however randomly selected on an Australia-wide basis and lack the normal volunteer artifact so prominent in most door-to-door surveys. The ninety-six recruits in the present sample were contacted in their first week of training at Kapooka.
Included with the 'A' scale was a short ethnocentrism scale extracted from the work of Beswick and Hills (1969). The items of this scale are given in Table 3. The reliabilities (coefficient 'alpha') of the two scales were .61 for the 'E' scale and .64 for the 'A' scale. This drop in 'A' scale reliability is of course to be expected in a scale applied to a sample unlike that on which it was standardised. The correlation between the two scales was found to be -.108. Not only is this non-significant but it is in the opposite direction to the correlations observed in using the 'F' scale.
The 'E' (ethnocentrism) scale:
1 Allowing educated Asians to immigrate benefits Australian society. R
2 The White Australia policy is a good policy because it keeps Australia
3 We must be careful not to let too many Asians into the country or they'll take over the place.
4 Asians should be allowed to migrate to Australia. R
5 The Japanese are a very productive people and should be allowed to settle in Australia. R
6 I would not like an Aboriginal to be my boss.
A valid and reliable scale has been presented which provides an alternative measure of authoritarianism to the California 'F' scale. Using this measure, the relationship of authoritarianism to submissive behaviour and to conservatism has been independently verified.
Most importantly of all however, authoritarianism as measured by the 'A' scale has been shown to be quite different to that measured by the 'F' scale -- in that 'A' scale scores are not related to ethnocentrism.
 They were partly drawn from Jacobs, Teune and Watts (1967) and partly written especially for this study.
 These data were collected by Miss J. Ross.
ADORNO, T. W., ELSE FRENKEL-BRUNSWIK, D. J. LEVINSON and R. N. SANFORD (1950) The authoritarian personality. New York: Harper.
BANNISTER. D. (1966) 'A new theory of personality.' In B. M. Foss (ed.), New horizons in psychology. Harmondsworth, Middlesex: Penguin.
BESWICK, D. G. and M. D. HILLS (1969) 'An Australian ethnocentrism scale.' Australian J. Psychology, 21: 211-26.
BROWN, R. (1964) Social Psychology. New York: Free Press.
CRONBACH, L.J. (1931) Coefficient alpha and the internal structure or tests. Psychometrika, 16: 297-334.
EYSENCK, H. J. (1954) The Psychology of Politics. London: Routledge.
JACOBS, P. E., H. TEUNE, and T. WATTS (1967) 'Values, leadership and development.' Paper delivered at the annual meeting of the Political Science Association.
KIRSCHT, J. P. and R. C. DILLEHAY (1967) Dimensions of authoritarianism: A review of research and theory. Lexington: University of Kentucky Press.
LEE. R. L.. and P. B. WARR (1969) 'The development and standardization of a balanced F scale.' J. General Psychology, 81, 109-129.
LENTZ, T.E. (Jr.) & Colleagues (1935) "Manual for C-R opinionnaire" St Louis: Washington Univ. Character Res. Inst.
LIPSET, S. M. (1960) Political man. New York: Doubleday.
ROKEACH, M. (1960) The open and closed mind. New York: Basic Books.
TITUS, H. E. (1968) 'F scale validity considered against peer nomination criteria.' Psychological Record, 18: 18, 395-403.
The work above was substantially updated here:
Ray, J.J. (1984). Half of all racists are Left-wing. Political Psychology, 5, 227-236.
Replication is one of the cornerstones of science. A new research result will normally require replication by later researchers before the truth and accuracy of the observation concerned is generally accepted. If a result is to be replicated, however, careful specification of the original research procedure is important.
In questionnaire research it has been my observation that the results are fairly robust as to questionnaire format. It is the content of the question that matters rather than how the question is presented. It is nonetheless obviously desirable for an attempted replication to follow the original procedure as closely as possible so I have given here samples of how I presented my questionnaires in most of the research I did.
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)