Tableaus, 1982, 169, 4-7.


John Ray


That part of the USA closest to the equator, "The South", is commonly characterised as more conservative, more romantic and more authoritarian. In recent years there has evolved a tendency to refer to Australia's northernmost state (Queensland) as the "Deep North" in analogy with this. Queensland is said to be not only warmer, but also more conservative and authoritarian. Scales to measure economic, moral and social conservatism together with the Munro-Adams attitude to love scale and the "Directiveness" scale of authoritarian personality were therefore administered to random postal samples of 219 Queenslanders and 158 people from the more southern state of N.S.W. Queenslanders were found to be more conservative on moral and social issues, but were not more economically conservative or authoritarian. They did show greater belief in the power of love. Radicals on moral issues and conservatives on economic issues were found to be more authoritarian.


Although it was for a time considered a rather suspect field of study, the discussion of national characteristics has once again begun to take a place in the psychological literature commensurate with the place it has always had in popular discussion (8,17,18). If national characteristics are of interest, however, it must surely follow that regional characteristics within one nation can also be of interest. The most widely documented example of regional differences would appear to be the difference between the North and the South in the USA. Even if only as part of a wider "sun belt", this regional distinctiveness of the old South does not seem destined to vanish.

The immediate historical causes of this distinctiveness seem at least at first sight to be fairly well-known. The institution of slavery alone is often felt to be sufficient explanation for the origin of differences which have persisted into modern times. Such an explanation, however, immediately moves one to ask why it was that the South and the North had such different attitudes to slavery from very early times. Could it not be that slavery was a symptom rather than a cause of underlying attitudinal differences? One observation that would seem to lend a little plausibility to this hypothesis is the role of climate in influencing behaviour. There seems to be some strength in the generalisation that the nearer one moves to the equator, the more conservative and authoritarian the people become. In Europe, the more northerly areas (Britain, Scandinavia, Holland and N. Germany) seem to be the traditional home of democracy, whereas the more Southerly areas (Italy, Spain, Greece, Portugal, Austria, Bavaria) have been the progenitors of some rather notorious Fascist regimes in recent times. One is inclined to hypothesise, therefore, that, the South is more conservative simply because it is warmer.

Such an explanation, however, must be seen as rather tendentious. There are many other possible reasons that might be advanced for the differences between the North and the South of Europe (relative poverty, racial origin, etc) and some at least of these may be equally applicable to the American deep South. Even if we do accept the generalisation, how is it to be explained? Why does heat make you conservative? It seems to be now widely accepted that there is an association between midsummer weather and upsurges of rioting and racial strife, but even whether the causes of this are to be sought in physiological or in sociological processes is far from decided.

Nonetheless, it is a very curious thing that far away in the other side of the world in Australia, a very similar differentiation seems to have arisen in recent years. This is despite the fact that the first white settlers arrived in Australia only in 1788 and one would think that climate ought not yet to have had much of a chance to influence the pattern of attitudes in the country. Most of Australia's population is on the Eastern seaboard of the continent and it is the most Northerly of the Eastern States (Queensland) which has become commonly referred to at least among journalists in recent years as "The Deep North". Being in the southern hemisphere, the North is climatically warmest in Australia and the term "Deep North" is a deliberate and conscious analogy to America's "Deep South". Queensland, in other words, is felt to be authoritarian and conservative when compared with other Australian States.

Although there has been some discussion of this difference in the academic literature (11,19), it is one that has not yet been substantiated by academic research. It is the purpose of the present paper to remedy this. If it is true that Queensland is more conservative, it could have considerable impact on the explanation of North-South differences in the USA. This is because many of the alternative explanations for greater conservatism that bedevil explanations of European or American differences do not apply in Australia. Queensland does not have a vastly different history, a vastly different racial composition or a vastly different level of poverty from the rest of Australia. In all obvious respects it is in fact almost a natural experiment -- with a very high level of control for all differences other than climate.

As background to the experiment, it might be worthwhile to outline some of the reasons why Queensland has come to be referred to as "The Deep North". Aside from its warmer climate and the fact that it grows cotton and peanuts, the main reasons reduce to the policies of the present State government and the strong support Queenslanders give to the conservative side of politics in Federal elections.

Queensland's Premier -- the 70 year old Mr Bjelke-Petersen -- is well-known throughout Australia for the gerrymander that supposedly keeps his government in power and for his strong stand against what are regarded elsewhere in Australia as basic civil rights. The first of these beliefs is in fact little more than failure to understand the implications of Australia's system of preferential voting. While it is true that Mr Petersen's own party (the National Party) gets only about a quarter of the vote at State elections, Mr Petersen's government is in fact a coalition of the National Party and the Liberal Party and the total preferred vote of these two conservative parties always exceeds 50% of the votes cast. In fact Mr Petersen's government has the record for the highest percentage of the popular vote (58%) gained at a general election by any Australian government -- State or Federal.

The other charges against Mr Petersen, however, do have more substance. Mr Petersen and his family openly hold shares in companies with which his government does business; He does send in baton-wielding police against anti-Apartheid demonstrators; He does administer legislation governing the State's Aboriginal population that must at least be described as paternalistic; He has effectively outlawed demonstrations and street marches by radical groups on the grounds that they "disrupt traffic"; He does support foreign investment at the expense of "the environment" and he does censor as pornographic even old-fashioned girlie magazines such as "Playboy".


The only existing data on attitudinal differences between Australian states is in the form of single-issue public opinion poll questions. Such data does not of course permit inferences about stable character traits or attitude constellations. Attitude to abortion (for instance) may reflect general conservatism in some cases, but such attitudes are not necessarily representative of conservatism as a whole. Whether attitudes to abortion do go together with other attitudes characterised as conservative must be the subject of empirical proof. It cannot be assumed. If we wish to speak about conservatism in general, it is conservatism in general that we must measure.

The usual way of measuring such general traits is of course by way of the multi-item attitude scales familiar to most psychologists. One problem with such scales in the field of political issues, however, is that they tend to date very quickly. The salient political issues of today are different from those of even five years ago. The Wilson and Patterson (22) Conservatism scale, for instance, lists as issues "Computer Music" and "Pajama Parties" among others. Whatever their importance was in 1968, the relevance of such items to a definition of conservatism today is surely very slight. The present research then, uses freshly-constructed conservatism scales, but ones for which internal consistency can be demonstrated.

A wide range of existing conservatism scales were examined and discussions held to find out what seemed to be topical political issues. The statements of attitudes so gathered were grouped under three headings: Economic, Moral and General Social. The Economic Conservatism scale, for instance, consisted of statements for and against greater government involvement in the economy and more equal distribution of wealth. The moral Conservatism scale discussed issues of greater permissiveness or restraint in matters of sexual morality. The General Social Conservatism scale included all those issues not previously included in the other two scales.

All three scales were combined into a single questionnaire together with the Ray Directiveness (12,15) scale and the Munro-Adams (9) "Attitude to Love" scale. The Directiveness scale is a measure of authoritarianism in behaviour inventory format and was included because of the tendency to confound discourse about conservatism with discourse about authoritarianism (e.g. 19, 22). In such circumstances, it was thought desirable to measure the two entities separately. It could thus be shown that (for instance) even if Queenslanders are more conservative, they are not necessarily also more authoritarian. It must be stressed that the Directiveness scale as used here, however, measures authoritarianism only in the narrow sense of "The desire or tendency to impose one's own will on others", rather than in any broader sense of rigidity, aggressiveness, etc (1,12,15). The Munro-Adams scale was included because of the apparent importance of moral issues in Queensland. Moral campaigners usually claim to be defending "the family" and condemn their opponents as depraved. It was desired to test this claim by examining if attitudes to relationships with the opposite sex were in fact different in Queensland. The question was, in other words, whether greater moral conservatism does in fact lead to a more loving and satisfactory perceived relationship between the sexes.

Another reason for including the Munro-Adams scale was a general impression that one of the minor currents in perception of the American South is an impression of the South as somewhat romantic. This may reflect nothing more than a nostalgia for the graceful and aristocratic life of the old Southern gentry, but it did nonetheless seem worthwhile to see whether attitudes in Queensland also might not be somewhat more romantic than elsewhere.

The questionnaire containing the above scales was mailed out to 700 randomly chosen addresses in Queensland and 500 addresses in New South Wales. As Australia's oldest and most populous state, N.S.W. was thought to form a contrast of some interest with Queensland. In each case the addresses used were chosen from the respective State Electoral Rolls. Because of Australia's system of compulsory voter registration, the sampling frame was thus unusually comprehensive.

An issue that was expected to arise in interpreting any results of the survey was the effect of urban/rural differences. As Australia's most decentralised State, Queenslanders are much more likely to live outside the Capital city than are any other Australians. Any finding that they are more conservative could thus plausibly be attributed to the greater rural weighting in the sample. A simple way to preclude this possibility is to examine responses emanating from the two capital cities separately. To give an expected final n for each capital city of approx. 100, therefore, more questionnaires had to be sent out in Queensland than in New South Wales (assuming a 25% - 30% response rate). According to official statistics (3 -- pp. 79,80), Brisbane has 46.5% of the State's population whereas Sydney has 63%.

The aim of a final n around 100 needs some explanation. The function of sample size is often misunderstood and the practice in commercial public opinion polls of using samples of around 2000 may make the present target samples seem foolishly small. It should be noted however that public opinion polls generally have as their prime purpose the high probability estimation of some precise parameter - e.g. the percentage of the population who intend to vote for a particular politician. For this purpose, samples in excess of 1000 are indeed necessary. The social scientist's prime purpose, however, is generally to determine whether a particular effect (difference, correlation) exists at all. Although he may be very interested (for instance) in whether working class people are more authoritarian than others, it is a matter of some indifference to him whether they are 49.9% more authoritarian than some other group or 51.1% more authoritarian. To a politician, by contrast, the same difference would be crucial. As one might expect, the sample size required for the academic's task is much smaller. A glance at the tables in the back of any statistics book shows that the sample sizes are often not given above 60 or 100. Samples as small as 60 or 100 are capable of detecting the existence of such weak effects that further tabulation is seldom necessary. Only if the social scientist were interested in the existence of very weak effects indeed, would he require samples in excess of 60 to 100. For instance with an n of 95, correlations explaining less than 5% of the variance would be shown as significant ( 5 -- p. 362). Weaker effects than this are not generally of great interest. A further exposition of the issues involved here may be found in Lykken (7).

Another issue that arose in the design of the present study concerned the projected response rate. "One wave" postal surveys in Australia seem to give response rates of as little as 25%. This can of course be improved by sending out a second wave of questionnaires to those who failed to return the first one. To do this, however, we has to keep records of names and addresses and this makes truthful guarantees of anonymity impossible. As many of the questions in the present survey were necessarily of current political importance, the guarantee of anonymity was thought indispensable on the present occasion. A second wave could not therefore be sent out.


For each State, the same proportion of questionnaires was returned -- 31%. This useful uniformity does at least preclude debate about possible effects of differential response rates. Thus, although such a response rate must raise some questions about the samples' representativeness, it is at least shown that whatever biases affected the results, did affect the results from both States equally. With sample biases thus controlled, the observed differences should be due to genuine inter-State differences. It must be re-iterated that it is the purpose of the present study to compare States, rather than to obtain exact estimates of population parameters.

The mean age of the Queensland sample was 42.9 years (S.D. 17.07) vs 41.58 (S.D. 16.77) for the N.S.W. sample. The difference is not significant. The N.S.W. sample contained 79 males and 79 females; Queensland had 108 females and 111 males. Again the difference is not significant. Education was scored on a 4-point scale from 1 (primary) to 4 (tertiary) with 3 indicating matriculation standard and 2 some high school education. On this scale the Queensland mean was 2.29(S.D. .99) while the N.S.W. mean was 2.88 (S.D. 1.00). The difference is significant < .05. Occupation was scored 1 for manual and 2 for nonmanual (vide 10). With this scoring the N.S.W. mean was 1.69 (S.D. 0.47) while the Queensland mean was 1.57 (0.53). Again the difference is significant < .05.

Another demographic comparison of some interest in passing was the comparison between the means on the same four variables for the Sydney sub-sample of the above survey and another doorstep sample (n = 206) gathered at much the same time in the Sydney metropolitan area. The means for the random doorstep survey and for the random postal survey were in fact virtually identical. There were certainly no significant differences. This would tend to suggest that the postal survey method did not introduce any unusual biases.

Broken down by area of origin, the postal survey responses originated as follows: Sydney n=98; Brisbane _n=104; Non-Metropolitan N.S.W. n=60; Non-Metropolitan Queensland n=115. As proportions of the Statewide samples, the Sydney and Brisbane ns are virtually identical to those given earlier derived from official census statistics of the Australian population. The more rural character of the Queensland sample would then easily account for the lesser education and lower occupational status of Queenslanders. Country people are of course usually less educated and have fewer openings for non-manual employment.

The coefficient "alpha" reliabilities (4) for the respective scales were as follows: Social Conservatism (30 items) .71 in Queensland and .81 in N.S.W.; Moral Conservatism (15 items) .80 in Queensland and .85 in N.S.W.; Economic Conservatism (12 items) .72 in Queensland and .78 in N.S.W.; Directiveness (30 items) .87 in Queensland and .86 in N.S.W.; Attitude to love (26 items) .87 in Qld and .81 in N.S.W. All are reasonably satisfactory and indicate that coherent traits are being measured. They indicate that there is some generalizability between attitudes -- enough to warrant concepts as general as "Conservatism".

The means on the attitude and personality scales are given in Table 1 for Qld. and N.S.W, separately. The means for Brisbane only are given in a third column. This third column shows that the Qld. and Brisbane means are virtually identical -- thus disposing of any possibility that the differences between the States could be simply an artifact of the greater proportion of rural respondents in the Queensland sample.

It was found then that Queenslanders were in fact significantly more conservative on social and moral issues, but were not more conservative on economic issues. The fact that economic issues are responded to differently is in accord with Lipset's (6) well-known thesis concerning "working-class authoritarianism" and well justifies the decision not to treat conservatism as a unitary concept. Queenslanders were also significantly more likely to have a positive attitude to Love. The t for this comparison was on the very borderline of significance (1.989) and it was clearly on the Romantic Power sub-scale of-the overall Munro-Adams scale that most of the difference occurred. Queenslanders tended to believe more in the power of love. Finally, it was shown that Queenslanders were not significantly differentiated by their degree of personal authoritarianism.

The correlations between the variables revealed that the "Directiveness" scale of authoritarian personality did not correlate significantly in either State with Social conservatism, but did correlate with moral (rs of -.273 & -.213) and economic conservatism (.174 & .231). Moral conservatism and Attitude to Love also correlated significantly in both States (.320 & .436).

In view of the fact that variables differentiating Queenslanders and New South Welshmen were themselves correlated, the question arose as to what degree the different tests furnished some element of independent prediction. To examine this, a multiple regression was carried out using State of residence as a "dummy" (scored "1" or "0") criterion variable. With an r of .219 General Social Conservatism was the major predictor. Nonetheless the multiple R obtained was .317 - which is significant <. 01 and is substantially greater than the r obtained with General Social Conservatism alone. The Moral Conservatism and Romantic Power subscales did then make some independent contribution to the prediction.


Means and S.D.s on personality and attitude variables for two Australian States and their two capital cities as sub-samples. Significant (<. 05) differences between columns 1 and 2 and between columns 3 and 4 are marked by an asterisk against the rightmost mean of each pair.


Social Conservatism...69.01(9.49).......72.84(7.56)*.......71.92(7.39)........68.78(9.40)*
Moral Conservatism....26.41(7.36).......29.53(6.73)*.......29.41(6.60)........25.70(7.15)*
Econ. Conservatism....25.85(5.49).......25.19(5.02).........25.24(4.88)........26.26(5.18)
Attitude to Love............65.30(8.14).......67.10(8.99)*.......66.46(9.04)........64.22(7.56)*
...Romantic Power.......19.86(5.31).......21.46(5.08)*.......21.12(5.16).......19.23(5.21)*
...Romantic Love..........24.70(2.89).......24.89(3.14)........24.84(3.31)........24.63(2.65)
...Conjugal Love...........20.72(2.88).......20.74(3.10)........20.49(3.28).......20.35(2.90)


Multiple regression analysis of Queensland and N.S.W. attitude and personality scale data. Predicting State of residence (Qld.=1; N.S.W.=0).

Variable...........................Correlation.....Beta Weights

Social Conservatism............ .219............... .236
Moral Conservatism............. .215............... .117
Economic Conservatism......-.063...............-.191
Conjugal Love...................... .003...............-.085
Romantic Love...................... .030...............-.070
Romantic Power................... .150............... .105
Directiveness........................-.043............... .030


It has been confirmed that the greater conservatism of Queensland politics reflects more than the personality of the present State government. Queenslanders generally are more conservative on moral and other social issues. It has furthermore been shown that these differences apply even as between Sydney and Brisbane. The differences cannot solely be accounted for as an outcome of the more rural nature of the Queensland population. Even Qld city-dwellers are more conservative in outlook.

It has however been confirmed that whatever might be true of Queensland conservatism, Queensland overall is not (contrary to Thomas (19)) an authoritarian society. The degree of personal authoritarianism evinced by the average Queenslander in the present survey was no different to that observed in N.S.W. This conclusion is founded upon answers to an authoritarianism scale of demonstrated behavioural validity (12, 15). In this respect the Directiveness scale differs from the California F scale (12, 20, 21).

The finding that Queenslanders and New South Welshmen differ in attitudes but not in personality mirrors a similar finding -- also obtained with the Directiveness scale -- as respects the differences between the English and the Scots (14). It may be that the Directiveness scale taps something more deep-seated in the personality which is hence harder to change than do scales of political attitudes. This is however not to say that mean Directiveness scores are invariant across populations. Recent work in one of the traditionally submissive cultures of Asia (Philippines) has shown especially low mean scores there (16).

The greater belief in the transforming and overcoming power of love among Queenslanders may perhaps give some support to morals campaigners. There does seem to be some association between moral strictness and more loving attitudes. Whether either is realistic in any sense, however, is of course another issue. Not only the difference between means, but also the correlations, support the association between more love-oriented attitudes and moral conservatism.

The pattern of the American South is then rather strikingly replicated. Queensland is not only warmer but also more conservative, more morally impermissive and even more romantic. This latter finding is in fact something of a surprise in that it does not form part of the stereotype of Queensland generally held in Australia (2). The finding is therefore some testimony to the heuristic power of the theory that climate has similar effects on people worldwide.

As even laboratory experiments are seldom able to exclude all extraneous variables that could have influenced the results, it cannot be hoped that the natural experiment reported here will be free from criticism. Unlike the American South, Queensland has never been at war with the rest of the country; like the rest of the country, it has only a very small proportion of blacks (1%-2%); like N.S.W. it was first settled as a penal colony; like the rest of Australia its population is about 80% of Anglo-Celtic origin and there are also only the most minute differences in inter-State standards of living. These are very impressive uniformities, but the possibility of there being some subtle differences other than climate cannot of course be dismissed. It can only be said that the hypothesis of a climatic influence on attitudes has been given much stronger support than has previously been available.

An incidental finding of some interest was that some types of conservatism were significantly correlated with scores on the Directiveness scale of authoritarianism. This runs somewhat counter to the U.K. findings reported in Ray (13) - where it was shown that there was no association between Directiveness and conservatism. The direction of the correlation was particularly interesting. On economic issues it is the conservatives who are more authoritarian, but on moral issues it is the radicals (advocates of permissiveness) who are more authoritarian. The picture is still very different to what Adorno at al (1) would lead us to expect. At least one type of Left-wing authoritarianism has at last been detected. In neither State, however, was there any significant association between vote and authoritarianism.


1. Adorno, T.W., Frenkel-Brunswik, E., Levinson, D.J. & Sanford, R.N. The Authoritarian Personality N.Y.: Harper, 1950.

2. Berry, J.W. The Stereotypes of Australian States. Aust. J. Psychol. 1969, 21, 227-233.

3. Cameron, R.J. Year Book: Australia. No. 63 Canberra: Aust. Bureau of Statistics, 1979.

4. Cronbach, L.J. Coefficient "alpha" and the internal structure of tests. Psychometrika 1951, 16,297-334

5. Edwards, A.L. Experimental Design in Psychological Research N.Y.: Holt Rinehart, 1960.

6. Lipset, S.M. Political Man N.Y.: Doubleday, 1960.

7. Lykken, D.T. Statistical significance in psychol. research. Psychol. Bulletin 1968, 70, 151-159.

8. Lynn, R. Personality and National Character Oxford, England: Pergamon, 1971.

9. Munro, B & Adams, G.R. Love American Style: a test of role structure theory on changes in attitude to love. Human Relations 1978, 31, 215-228.

10. Ray, J.J. (1971) The questionnaire measurement of social class. Australian & New Zealand J. Sociology 7(April), 58-64.

11. Ray, J.J. (1976) Authoritarianism and racial prejudice in Australia: A reply to Thomas. Journal of Social Psychology, 99, 163-166.

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

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

14. Ray, J.J. (1979) How different are the Scots and English? Contemporary Review 234, 158-159.

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

16. Ray, J.J. (1981) Achievement motivation and authoritarianism in Manila and some Anglo-Saxon cities. J. Social Psychology 115, 3-8.

17. Howse, T. Australian Liberalism and National Character Melbourne Australia: Kibbel Books, 1978.

18. Terhune, K.W. From national character to national behaviour: a reformulation. J. Conflict Resolution 1970, 14, 203-263.

19. Thomas, D.R. The relationship between ethnocentrism and conservatism in an "authoritarian" culture. J. Social Psychology, 1974, 94, 179-186.

20. Titus, H.E. F scale validity considered against peer-rating criteria. Psychological Record 1968 18, 395-403.

21. Titus, H.E. & Hollander, E.P. The California F scale in psychological research: 1950-1955. Psychol. Bulletin 1957, 54, 47-64.

22. Wilson, G.D. & Patterson, J.R. A new measure of conservatism. Brit. J. Social & Clinical Psychol. 1968, 7, 264-269.


Other articles on the topic above are as under:

Ray, J.J. (1982) Climate and conservatism in Australia. Journal of Social Psychology, 117, 297-298.

Ray, J.J. (1983) Race and climate as influences on anxiety. Personality & Individual Differences, 4, 699-701.

Ray, J.J. & Hall, G.P. (1994) CONSERVATISM IN THE DEEP NORTH: Trends in Queensland attitudes. Unpublished.

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)