Recently, the New York Times wrote about the Catholic Church's attempt to ferret out certain types of gay candidates for the priesthood, using psychological testing.
This attempt is obviously something they feel strongly compelled to do, given their long history of failures to expelling known pedophiles, but it is a method doomed to failure.
I'm no psychologist, and I usually try to write about things within my area of competence, but I did spend an extensive amount of time looking into psychological testing during my Ph.D. studies in Law, Policy and Society. I was looking for psychological tests that could reveal the presence of a transgender or transsexual personality. I had heard about a number of tests that purported to do so. I took an advanced psychology course and chose, as my subject, the nature and construction of psychological personality testing. I spent a lot of time meeting with the professor outside of class to discuss the psychological tests I had researched. I hoped to use the information in my dissertation.
I scrapped the whole subject as a bad business, and not worth my time. It's a pretty fascinating idea, just like those quizzes you can take in Cosmo that tell you your ideal romantic partner. And about as useful, for the purposes of the Church.
What I learned about psychological testing startled me. It also suggests that the Catholic Church is not likely to learn much from these psychological testing methods. But it will give them pretty good cover for rejecting applicants that appear to them like the image of a gay man in their heads that they learned from TV and movies.
The MMPI, the Minnesota Multiphasic Personality Inventory, is one of the oldest and most highly regarded tests for psychopathology. It's basically a survey that asks the taker a lot of questions about their feelings -- a lot of questions, about 600. The format of the questions is true/false. It takes an hour or two to complete, depending on your reading level.
An analysis of the results can supposedly reveal the likely presence of mental illness or certain personality traits.
Now, if answered truthfully, I have no doubt that the results reveal something useful for a trained mental health professional.
But since when have people ever answered a survey truthfully? Especially if they know the test-giver is looking for specific "problem" traits? They might as well give a survey that asks candidates "True or false, do you like to fondle little children?" The reliability and validity of this test rank up there with the polygraph lie detector machine, which is to say that the test can be faked out pretty easily.
Ah, but the MMPI is no ordinary survey. First of all, after a while -- say about the 100th question -- the answering of the questions gets mechanical, and you have to think harder to lie.
Second, they have a "fake out the fakers" mechanism built in. That means that the test-makers, who have been working on this test for the past seventy years, have a pretty good idea of what the answer profile is going to look like. Some people want to "fake bad," that is to say, they want to appear ill in some way. Maybe they are taking the test to try to back up their claims of disability. There are certain questions salted throughout the test that even those with serious mental illnesses would mark "No." But the fakers, malingerers and people with factitious disorders show a pattern of marking yes. Or at least, that's the theory.
There are also people who want to "fake good."
In other words, they have mental illnesses or personality problems, say, the desire to fondle little children. But they want to appear as if they don't. Those people say "No" to everything that even hints at psychopathology. But the normal person, telling the truth, would actually mark "Yes" to a lot of those questions. Those who wish to "fake good" show a pattern of marking no. That's the theory.
In the best tradition of "your lips say no, no, but your eyes say yes, yes," psychologists who believe in the MMPI also believe that they can evaluate the answers to certain questions given by the millions of people who have taken the test, and reliably decide that they indicate faking. But the people who really want to fake don't have to look far to find information on how to fake properly. I mean, all you have to do is look at the DSM, memorize the symptoms to your favorite mental illness, and answer everything truthfully, except for the questions that invoke your chosen illness.
In addition, people with certain mental illnesses seem to invariably have the same answer to certain questions, even though the question itself seemed totally irrelevant to psychopathology. For example, I recall one question that was deemed to reliably indicate the presence of mental illness. That question was:
True or False: My stomach often hurts.
Surprising, right? But when the test-maker ran the results through various mathematical analyses, they found that this was one question in common with the vast majority of people with certain categories of mental illnesses. Who know why? Maybe people with those illnesses have a high degree of anxiety that makes their stomachs hurt? Maybe they need attention? Maybe their psychological symptoms manifest as somatic symptoms? For the purposes of the test-makers, it doesn't really matter why. It just works.
And that's a lot of how the MMPI works. They looks at millions of test results, determine various patterns, and use those patterns to judge individual test-taker responses.
It's very, very cost-efficient. And that's the main thing about the MMPI that makes the psych people love it. After all, it's pretty obvious to everyone that if a trained mental health professional sat down with an individual and asked them 600 questions, the mental health professional's judgment would likely be better than a mechanical analysis of 600 true/false questions by someone with no personal knowledge of this test-taker. But how much time and money would that require? A lot. More than any organization cares to spend on candidates.
There are other tests that purport to be better than the MMPI, like the California Personality Inventory, but most of the other tests are shorter and have less research data on reliability and validity. The MMPI is still the gold standard in psych testing.
And the MMPI people clearly make the point that the test is only as good as the interpreter. In their directions to the test-giver, they note that the results can only be interpreted by a properly-training mental health professional who actually works with the individual to address areas suggested by the test. And that requires a patient who wants to get better, not someone who's trying to get out of being identified as mentally ill.
But the idea that the MMPI can reliably root out pedophiles, or gay men, or transsexuals, or whatever, is based in nothing scientific. Its mystique, however, is likely sufficient for the purposes of the Church. They can say they're working to root out pedophiles, or gay men, or whatever, and that they have this magic wand that can tell them who is who.
It's the same reason government agencies continue to use the polygraph lie detector machine. It has a lot of dials and buzzers and lights and stuff, and even though it's easily faked out, it can induce a lot of confessions from people who are afraid of it.
Fear is the best truth-detection tool. It worked pretty well for the Inquisition, it works pretty well at Guantanamo and Baghram, and, truth be told, the 15th century really isn't all that far away.