The Left are amusing: The "authoritarianism" explanation for Trump support

On Feb 25 I put up some comments on the accusation that Trump supporters are "authoritarian".  The article I critiqued was based on some research by a PR man named Matthew MacWilliams.

The claim is that "authoritarians" are fear-motivated and that Trump panders to those fears.  That authoritarians are fear motivated is a claim that goes back a long way.  Erich Fromm asserted it in the '50s and various subsequent authors have made feeble attempts to prove it, e.g. Sales, S. (1973). Careful researchers would use a measure of fear motivation and a measure of authoritarianism and try to show that the two were correlated.  But MacWilliams and his gurus skipped that awkward step as far as I can see.  They just defined authoritarianism in their own way and noted that it showed correlations with some fears.  That their measure of authoritarianism was in fact a measure of anything authority-related was not shown.  As far as I can see it at most measures old-fashioned thinking. But even the academic work that MacWilliams relies on -- work by Feldman & Stenner (1997) -- concedes that there is no direct relationship between authoritarianism and threat/fear.

And the research made elementary mistakes -- indicating a profound ignorance of the precautions that psychometricians normally take when doing survey research.  Recently, however, a new and much expanded article based on the MacWilliams research has emerged -- under the title "The rise of American authoritarianism", written by Amanda Taub.

I think my previous comments were sufficient to show that the work is a lot of hokum but maybe I can add a few more comments.

I might initially expand my comments about the naive nature of the questions they used to assess authoritarianism. They were "forced-choice" questions.  A typical question was

"Please tell me which one you think is more important for a child to have: to be considerate or to be well-behaved?".

The option you chose was supposed to indicate whether you are authoritarian or not.  But what if you thought that BOTH attributes are important?  What if you wanted a kid who was BOTH considerate AND well-behaved?  The form of the question prevents you from saying that.  So the answers given might not well represent what the person actually thinks.

So is that naive form of question construction actually misleading?  It is. If the people don't like the choices they are offered, what is likely to happen is that a "Donkey vote" effect will result: If the choices in a forced-choice scale are labelled "a" and "b", the Donkey voter will, at the extreme, simply tick all the "a"s.  And I showed in my own survey research years ago that forced choice questions can push the results in a direction more or less opposite to what occurs with more straightforward questions

I think that alone invalidates their conclusions but "Wait! There's more"! -- as the steak-knife salesman said.

The "authoritarianism" researchers say that Trump appeals to "authoritarians" as defined by them and that Trump is the ideal candidate for authoritarians. So virtually all authoritarians should support him, it would seem.   So I was amused to read this about their research findings:

"Trump has ...  a full 52 percent support among very high authoritarians."

What a laugh!  Even high authoritarians split roughly 50/50 in support for Trump.  Nearly half of these sad people DON'T support Trump.  Where does that leave Trump as the ideal candidate for authoritarians?  As is common in Leftist researchers, they can't even read their own data.  They conclude what they want to conclude, regardless of their actual findings.  Keeping reality out is an essential skill for Leftists.

So they have a lovely theory but it happens to be wrong.  So they might have to accept that there really is something rotten in the state of American politics, and Donald Trump is bringing that to the fore.  The fault may lie with the political establishment, not with the personal inadequacies of Trump supporters.


Sales, S. (1973) "Threat as a factor in authoritarianism".  J.  Personality & Social Psychology, 28, 44-57.

No comments:

Post a Comment

All comments containing Chinese characters will not be published as I do not understand them