Saturday, January 20, 2007


By John J. Ray | August 27, 2003

Like most college and university teachers in the social sciences and humanities, academic psychologists are overwhelmingly Leftist in their orientation. So it will be no surprise to hear that at least since the 1950's psychologists have been doing their best to find psychological maladjustment in conservatives. To anyone with a knowledge of history the results have been quite absurd (See here) but psychologists rarely seem to know much about history so that has not disturbed them.

I spent 20 years from 1970 to 1990 getting over 200 articles published in the academic journals of the social sciences which subjected the various politically relevant theories of psychologists to empirical test. The only test that psychologists normally give to their theories is to seek the opinions of their students on a variety of issues and present THAT as evidence about how the world works. My consistent strategy was to do the same sort of test among random samples of people in the community at large. I found that people in the community at large are not nearly as accommodating to the theories of psychologists as psychology students are!

My non-conformist behaviour in actually doing a serious test of these theories won me no kudos, however. I appear to have had far more articles on political psychology published in the academic journals than anyone else and so would therefore -- by conventional academic criteria -- normally be considered the No. 1 world expert on the subject but in fact my writings have always been comprehensively ignored. My findings did not produce the RIGHT CONCLUSIONS, you see. In fact my findings showed the theories concerned to be wrong in almost every respect.

So it was no surprise to me at all to read the latest effort in the long line of attempts by psychologists to discredit conservatives. The article "Political Conservatism as Motivated Social Cognition" was published recently by John Jost and his collaborators at Berkeley in The Psychological Bulletin -- one of the premier journals of academic psychology. The "powers that be" at Berkeley were so pleased with this article that they put out a press release that was designed to publicize the findings of the article as widely as possible.

The result was great derision from conservative political commentators. The study was so obviously one-eyed that it was very easy to deride. Their claim that Stalin was Right-wing, for instance must be some high-point of twisting the evidence. If the most prominent Communist of the 20th century was Right-wing, who on earth would be Left-wing? Black might as well be white. Here is what Jost and his crew actually said:

"There are also cases of left-wing ideologues who, once they are in power, steadfastly resist change, allegedly in the name of egalitarianism, such as Stalin or Khrushchev or Castro (see J. Martin, Scully, & Levitt, 1990). It is reasonable to suggest that some of these historical figures may be considered politically conservative"

It is hard to know where to start in commenting on this breathtaking statement. To say that the instigator of huge (and disastrous) changes in almost everything in Russian life resisted change is incomprehensible. And to call Communists of that era conservatives is equally perverse. One has to say that "conservative" obviously has a pretty strange meaning in the ivory towers of Berkeley. In their world even Stalin can be blamed on conservatism.

Apparently as an attempted explanation of their perverse definitions, they go on to say that the worldwide legion of Communist tyrants that they allude to are not typical of Leftists. The fact that Communists at their height controlled nearly half the world is not apparently enough to get them counted as typical Leftists.

The reason they get themselves into this awful muddle is that they have no idea what conservatism actually is. To define what it is, they consulted dictionaries and social science encyclopaedias and their fellow Leftist writers but they do not appear to have consulted a single conservative! And they have certainly studied no political history. They have fallen into the simplistic trap of equating conservatism with rejection of change. It is of course true that conservatives DO reject some changes -- the foolish changes advocated by Leftists in particular -- but it is also true that conservatives WANT change -- change that will get the government off their backs in particular. If Jost and his colleagues had talked to almost any conservative they would have found out very rapidly that there are a HEAP of things about our present society that conservatives would like to change. It is only Leftists talking to one another who think that conservatives are motivated only by a rejection of change.

If anybody would like to look at what has historically defined conservatives, a summary of the history concerned can be found here. Briefly, however, conservatives have always resisted attempts to curtail their rights and liberties -- particularly from the incursions of tyrannical governments. THAT is what has often made them resist change -- as governments do have a habit of trying to curtail the rights and liberties of the individual. But to say that conservatives resist change for its own sake is something that only a Leftist would believe. Conservatives believe in liberty. Leftists believe in the State.

So the Jost group in effect fell at the first hurdle. If they could not get straight what conservatism is, the chances of their saying anything reasonable about it were pretty slim. But let us look a little further at what they did say anyway:

The first thing to note is that the article concerned is not in any sense new evidence about anything. It is just a summary of what other Leftist psychologists have said on the subject over the years. And as computer people have always warned: GIGO (Garbage in, garbage out). And it is a very strange summary at that. Although the authors express some pride that they "cast a wide net" in looking for articles to include in their summary, they in fact leave out practically everything that does not suit them. They look only at articles that they like. It is NOT in any way a comprehensive, objective survey of the literature on the subject. For instance, how many of my articles did they cite? Just two. So just in ignoring the great bulk of my articles on the subject they ignored half the relevant literature. But Leftists always have been good at ignoring evidence.

So they say that their Leftist approach to the study of politics "has withstood the relentless tests of time and empirical scrutiny" and go on to cite 13 articles that support their approach as evidence for that assertion. If they had been fair and objective, they could also have quoted 100 articles of mine that upset that assertion. But in good Leftist fashion it is only people who support their views that they cite. No tolerance of ambiguity there!

Their ignorance of what conservatism is also shows in the means they use to measure it. Psychologists measure traits by use of "scales" (sets of interrelated questions). The more "conservative" statements you agree with, for instance, the more conservative you are said to be. The big hitch, of course, is in deciding what is a "conservative" statement. And psychologists have arrived at some strange conclusions in deciding that. Two of the most popular scales that psychologists use in studying conservative politics are the 'F' scale -- put together in the late 1940s by the (Marxist) Adorno and his group -- and the RWA scale put together by Altemeyer. Yet in general population samples Leftist voters are just about as likely to get high scores on both scales as Rightist voters are! (See e.g. here) So once again we have psychologists seriously telling us that a lot of Rightists are Leftists. Apparently Stalin is not so atypical after all!

That their measures of "conservatism" do not correlate with anything of current political relevance was even acknowledged by one of the authors of the scales concerned -- the RWA (Right-wing Authoritarianism) scale. Altemeyer, on p. 239 of his 1988 book Enemies of Freedom makes the bald statement that "Right-wing authoritarians show little preference in general for any political party". So in what sense are the statements in the scale "right-wing" if right-wingers are no more likely to endorse them than Leftists are? Leftist psychologists are like something out of Alice in Wonderland where words can mean anything that they say they mean. I myself do think that the F and RWA scales measure some sort of conservatism (though what sort would be hard to say) but it is certainly not political conservatism. My best shot so far at guessing what they measure is an old-fashioned outlook -- but note that both Leftist and Rightist voters are equally likely to be old-fashioned in the sense concerned.

But you would never guess from what the Jost group say that their scales are so hopelessly invalid ("invalid" = "does not measure what it is supposed to measure"). Of the RWA scale, for instance, they say "Scores on the RWA Scale have been found to predict a broad range of attitudes and behaviors related to social, economic, and political conservatism as defined in the general culture at the time. For instance, the scale has correlated reliably with political party affiliation" Weasel words! They do not even seem to have read Altemeyer himself. The shred of justification they have for that statement is that among SOME people (e.g. politicians) those identified as conservative are slightly more likely to get higher scores on the RWA scale. But it still remains true that in the community at large Leftist and Rightist voters are roughly equally likely to get high scores on the thing.

The central weasel word above is of course "reliably". Psychologists have the strange habit of taking a correlation seriously as long as it is "statistically significant". But ANY correlation will be statistically significant if the sample size is large enough! So we have the weird phenomenon of correlations as low as .15 (meaning that there is only a 2% overlap between two groups of people) being taken as proof of something ( See e.g. here). What this means in practice is that if 51% of voters for the conservatives get a high score on the RWA scale and 49% of voters for the Leftists get a high score on it (which is roughly what happens), psychologists say: "Aha! We told you so! This scale measures conservatism!"). And if that happens often we get statements such as the one that the RWA scale "reliably" measures conservatism. That roughly half of their "conservatives" vote Leftist does not seem to bother anyone at all!

Leftist beliefs are obviously so rigidly held to among psychologists that any evidence at all will do to support them.

I could go on to point out many, many more holes in this caricature of serious research (e.g. by pointing out that Hitler was a socialist, that Mussolini was a Marxist, that scores on the Dogmatism scale are meaningless and that intolerance of ambiguity is a "unicorn" concept) but I think that it would be flogging a dead horse to do so. I have written a fair bit more about it on my blog for those who want to look a little further and if anyone wants to look at the academic issues involved in detail there are all my published academic journal articles on the subject to refer to. They are all now online.

One of my academic colleagues -- Prof. James Lindgren -- has also been looking closely at the Jost work. He has gone to the trouble of finding out what the public opinion polls reveal about conservative voters. A sample of his findings is here. He does however have in preparation a point-by-point refutation of the Jost claims. Practically everything the Jost group claims about conservatives on the basis of what students say turns out to be the reverse of the truth according to the public opinion poll data. When Prof. Lindgren has his work ready I will of course publicize the link on my blog.

Post publication addenda

I understand that Prof. Lindgren has decided to report his research in the form of a book but that it has not yet been published.

There have been lots of subsequent uncritical mentions of the Jost et al. pseudo-meta-analysis but a widely read one appeared in "Psychology Today". A brief comment on the article concerned is therefore offered below. Below that is offered a brief comment on another article much relied upon in the "Psychology Today" article -- a quite ludicrous article by that old warhorse of the field, Jack Block.


Psychology Today has put out a big article on the psychology of politics. It purports to explain why people hew to the Left or the Right.

It bases its conclusions, however, on research that has no scholarly worth whatsoever. The two main pieces of research it relies on to support the differentiations it proposes are by Jost et al and Block et al. Both pieces of research are within my area of academic specialization so the gaping flaws in both are more evident to me than they might be to others. I have some time ago pointed out how intellectually discreditable both studies are -- reproduced above and below. I was actually one of the authors whose research the Jost et al. article purported to summarize!

Jost et al. showed vividly the usual Leftist talent for ignoring large bodies of evidence that did not suit their preconceptions. Their article must in fact rate as one of the most dishonest meta-analyses ever reported. They were very clearly guilty of "motivated cognition" -- exactly what they accused conservatives of. Leftist projection and leftist denial do usually go hand in hand.

Michelle Malkin and Jim Lindgren note the absurd "sampling" in the Block study. It seems unlikely that there were any actual conservatives among the individuals studied! That is not the first time Leftist psychologists have perpetrated such an absurdity in their mendacious claims to "explain" conservatism, however. See, for instance, here.

Given the very dubious basis of the generalizations offered in the Psychology Today article, the generalizations offered by the article should detain no-one. Whether any of the statements made are true is simply unknown.

Sadly, many in the blogosphere have assumed that there is some respectable evidential basis for the article. Cinnamon Stillwell and Fausta sum up some of the blogospheric reactions. Iron Shrink also has a detailed debunking of the Jost et al. study.

For an account of the psychology of conservatism based on the actual history of conservatism, see here.


Starting in 1950, psychologists have been trying to prove that conservatives are psychologically maladjusted --an effort that was summarized by Jost et al. and demolished above. They never give up trying, however, and the study of 95 children from Berkeley, California by Jack Block and his wife is the latest episode. It's a bit like shooting fish in a barrel to demolish the "research" concerned so I will just mention a few basics.

What the authors found was that Berkeley children who were rated as unhappy in their kindergarten years turned out as adults to be conservatives. So what do we conclude from that? Do we conclude that conservatives are intrinsically unhappy people? Hardly. The opinion polls repeatedly show conservatives to be the happiest (e.g. here). So what DO we conclude? How about saying that conservatives in ultra-Leftist Berkeley feel uncomfortable in that environment and pass on some of that discomfort to their children? But the children (as most children do) grow up to share their parents' politics so they turn out conservative in later life too. So conservative parents who were SITUATIONALLY unhappy have conservative children -- big deal!

But that is actually putting the best face on the study. It is not nearly as good as that:

1). The children concerned were not a representative sample of any known population. They may not even have been representative of Berkeley, let alone anywhere else. And if you don't sample, you can't generalize.

2). The correlation between personality and ideology was often very weak (e.g. .27 or 7% shared variance for the rating "Is self-reliant, confident") so there were nearly as many confident children who turned out Rightist as Leftist. The reported correlation could in fact have turned on the responses of just one child. That certainly weakens ALL causal inferences from the study. And many of the stronger correlations involve obvious value judgments. For instance "Is visibly deviant from peers" is said to characterize conservatives but why not turn the value judgment around and conclude that conservatives tend to be independent?

3). The measure of conservative ideology is suspect. Block does not list the actual attitude statements he used but, as I have shown above, Leftist psychologists in general don't have a blind clue what conservatism is -- and what they regard as a measure of conservatism is usually a caricature of the real thing -- generally a collection of ignorant and aggressive statements that very few real-world conservatives would assent to. That it was a caricature in this case is suggested by the fact that it showed conservatives as less intelligent. In the general population it is Leftists who are less intelligent.

I could go on to mention the Rosenthal effect etc. but what's the point?


Conservative commentators sometimes return the sort of compliment offered above by writing swingeing analyses of Leftist psychology. You can find one example here by forensic psychiatrist Lyle H. Rossiter, Jr, MD.

What a contrast! The article above about Leftists is based on what actual Leftists actually do — a vast difference from the Leftist psychologists who failed to interview any actual conservatives — and who certainly did not base their observations on what conservatives actually do, as distinct from what they say. And no psychologist should need to be told that attitudes are a poor guide to behaviour and it is behaviour that counts.