Make your own free website on

The Journal of Social Psychology, 1980. 112, 215-218.



University of New South Wales, Australia


Kerlinger's theory that liberal and conservative "criterial referents" are different and that liberalism and conservatism are hence unrelated is tested. Balanced scales containing only "conservative" referents and only "liberal" referents were constructed in the short "catchphrase" item format popularized by Wilson. There was some difficulty in deciding to which political group an item might most plausibly be a "referent." The two scales actually produced were adequately reliable and correlated --.64. The sample was of Australian Army conscripts (N = 110). The results represent, then, a marked failure to replicate Kerlinger's findings. Radicalism and conservatism were found to be highly opposed.


Kerlinger (1, 2) has put forward the unusual theory that, contrary to normal appearances, liberalism and conservatism are not in fact opposed. He claims that the things which concern the political Left and the political Right are simply different: i. e., those of concern to a liberal are matters of indifference to a conservative and vice versa.

Clearly wrong though this account must seem as a description of the actual course of Congressional and Parliamentary debate, it must be acknowledged that the reasons (for example) that conservatives have for opposing government-funded health services focus on quite different issues from the reasons advanced in favor of such health services by liberals. The liberal does not want the poor to lack care while the conservative does not want the taxpayer's funds wasted on bureaucratic inefficiency. The conservative is not apparently against care for the poor. He is more likely not to care much either way. Similarly, the liberal is not against efficient provision of services: Efficiency just does not seem very important in his scale of values. Thus although opposing positions might be taken on a particular issue, liberals and conservatives are in fact for the most part talking about different things. They have incompatible values rather than opposing values.

There are a number of problems with Kerlinger's view: The most obvious is that acquiescent response set could alone have produced his findings. Acquiescence causes items to be responded to similarly; it is an influence inducing them to correlate positively. If, however, the influence of item meaning is to cause items to correlate negatively, the two influences could arithmetically summate (cancel one another out) and thus produce the observed orthogonality. To eliminate this possibility, separate balanced scales to measure both liberalism and conservatism should be used.

This is not as impossible as it might at first seem. Kerlinger's view of separate criterial referents does not in fact seem necessarily to entail that all criterial referents of importance to our typical liberal or conservative will be positively viewed, although he gives that impression. There may be some items a liberal favors and some he opposes, but both should be matters of indifference to conservatives. Another problem with Kerlinger's findings is his use of student samples. Obviously, what is true of students, may not necessarily be true of the population as a whole. The present study was designed to provide some corrective to these problems.


Since Kerlinger first described his theory, an attractive way of testing it directly has become available: Wilson's "Catchphrase" format for attitude scale items (8). Wilson has shown that the core of most attitude scale items can be expressed as a single word or phrase and that people can meaningfully express an attitude by responding to that word or phrase alone. Kerlinger himself (2) has acknowledged the relevance of Wilson's method to his concept of a "criterial referent." It seemed appropriate, then, to write two balanced scales in the Wilson format which would embody the criteria] referents or catchphrases peculiar to the two political poles characteristic of the English-speaking world.

How to characterize these two poles was an initial minor difficulty. In the U.S.A., those on the political Left are wont to describe themselves as "liberals." In Australia and the U.K., the same people would nowadays describe themselves as "socialists." This probably reflects the greater radicalism of the political Left outside the U.S.A. Under the circumstances it was decided that the two scales should be conceived for Australian conditions as measuring "Radicalism" and "Conservatism."

The construction of the two scales presented other difficulties. How does one decide whether an item (referent) should be of greatest concern to a radical or a conservative? Kerlinger himself appears to suggest no general a priori criteria or procedures for decision. An initial foundation for the study was, then, the Wilson (8) Conservatism scale. To it were added a large number of extra items in the same format and the classification of all items was decided by reference to the profusely exemplified and very extensive definitions given under "Liberalism" and "Conservatism" by the unabridged Oxford English Dictionary. A categorized list of the items used is available gratis from the author to those readers who may wish to decide for themselves on the adequacy of the categorization. As examples, however, "Sabbath observance" and "My country" were classed as positive conservative referents with "Pot-smoking and" and "Communism" as negative conservative referents. "Mercy-killing" and "Civil rights" were classed as positive radical referents with "Defense-spending" and "Racial discrimination" as negative radical referents. There was produced, then, a 64-item "Radicalism" scale with 29 "pro" and 35 "anti" items and a 62-item "Conservatism" scale with 31 "pro" and 31 "anti" items.

The battery of items was administered under instructions similar to those recommended by Wilson (8). The sample has already been described elsewhere n. Briefly, it comprised 110 Australian Army conscripts (all males). Being drawn from the entire population by a random birthdate ballot, they do represent a good cross-section of the Australian population with age and sex controlled.


The Reliability ("alpha") of the two new scales was fairly high for a previously untried set of items. For the Left/liberal scale it was .74 and for the Conservatism scale it was .78. The two scales correlated -.64. This correlation is clearly a far different result from that found by Kerlinger; it supports the more usual view of the political spectrum: viz., that the Left and the Right are very much opposed not only on policies but also on what they basically value. To value Left-wing shibboleths is to reject Right-wing ones.

Clearly, then, Kerlinger's theory has not been validated under more closely controlled replication. Control for acquiescence has made a very considerable difference to the relationships observed. It would appear that Kerlinger has mistaken the usual tendency towards orthogonality produced by acquiescence (3, 5) for a real (substantive) effect.

One objection that might be urged against an interpretation of Kerlinger's original findings as an acquiescence artifact is the work of Rorer (7), who purports to show that a systematic tendency to acquiesce does not exist. Recent work reported elsewhere (4, 5, 6) has, however, shown that a consistent tendency to acquiesce can be demonstrated if a "within scales" rather than a "between scales'" approach is used. This undermines the evidential basis of Rorer's account and makes attention to the effects of acquiescence very necessary indeed.

A final objection that might be urged to the above conclusions is that Kerlinger's theory has not really been tested. Not liberal and conservative criterial referents, but two conservatism scales were correlated. If this is so, however, it points to the indeterminacy of Kerlinger's theory. How does one decide when an item is a liberal or conservative referent? Kerlinger's own procedure so far seems to amount simply to observing on which factor an item falls and hence fails to allow for the influence of acquiescence. The correlations on which the factor analysis is based will incorporate any distortions resulting from acquiescence. Acquiescence must be eliminated by experimental controls before an analysis is undertaken. Producing a balanced scale according to a predecided conceptual schema seems at the moment to be the only way of doing this. If the definitional resources used in the present study are accepted, Kerlinger's theory is not supported by the evidence. If they are not, his theory may be, at least so far, indeterminate.


1. KERLINGER, F. N. Social attitudes and their criterial referents: : A structural theory. Psychol. rev. 1967, 74, 110-122.

2. KERLINGER, F. N. Similarities and differences in social attitudes in four Western countries. Internat. J. Psychol., 1978, 13, 25-37.

3. RAY, J.J. (1972) Are conservatism scales irreversible? British J. Social & Clinical Psychology 11, 346-352.

4. RAY, J.J. (1979) Is the acquiescent response style not so mythical after all? Some results from a successful balanced F scale. Journal of Personality Assessment 43, 638-643.

5. RAY, J.J. (1979) Is the Dogmatism scale irreversible? South African Journal of Psychology 9, 104-107.

6. RAY, J.J. & PRATT, G.J. (1979) Is the influence of acquiescence on "catchphrase" type attitude scale items not so mythical after all? Australian Journal of Psychology 31, 73-78.

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

8. WILSON. G. D. The Psychology of Conservatism. London: Academic Press, 1973.

Go to Index page for this site

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