Several days ago I got into a debate with JV over the validity of the Myers-Briggs personality type indicator. In fact our debate was more over the countless web manifestations of it, most or all of which are of questionable quality and relation to the original test. JV had presented me with one of these and his own opinion of which type I was. The home of this version, which claims to be well-researched and not just another novelty, is here:

Myers-Briggs according to Craig

My problem with all the web personality tests, and I don't think this can be seriously disputed, is that they are for novelty purposes and unscientific. They seek a reaction of serendipity when, after answering some seemingly innocuous questions, you receive a block of text that describes you with incredible accuracy. It's as if the test understands you! The effect is produced by two methods. One is that of course the questions partly feed into the answers, but they are disguised well enough that you don't notice. The other is that the type description that you see at the end is written in a certain style, the same style used by astrologers and psychics and such. This style has been well-documented by skeptics. It's done primarily by telling people what they would like to believe about themselves. The rest is taken care of by our inherent suggestibility and resulting willingness to fill in the blanks, and remember the accurate parts while forgetting the less accurate parts.

A related aspect of the descriptions of the different types at the above site that bothers me is the feel-good nature of them. The one JV pointed me to for myself (Ok, it was INTP) was quite long and went on and on about what my type would do in different situations and how they would handle different parts of life. I had to admit to JV that plenty of things in the description seemed accurate to me. But it was also all very positive--telling me how smart I am, what a good memory I have, and so on. How can I disagree? About the worst thing in there is that I might "become [an] intellectual diletante as a result of [my] need to amass ideas, principles, or understanding behavior." Yes, you see, I'm just too dedicated, it's a terrible strain, really. The more I thought about it, the more equivocations I found in the description. Take this nugget for example: "They are very adaptable until one of their principles is violated. Then INTPs are not adaptable at all!" Yep, that's me...and most other people (or at least we'd like to think so).

I started reading some of the other type descriptions to get a feeling for how much more accurate my text was than theirs. But there was so much of it, and it was so exhausting to read, that it was hard to get a sense of the big picture. I did however find some more rather flexible sentences, such as these: "ENTPs have little patience with those they consider wrongheaded or unintelligent, and show little restraint in demonstrating this. However, they do tend to be extremely genial, and quite charming, when not being harassed by life in general." Does that describe you? Personally I'm the opposite! I love those wrongheaded people, but when life stops harassing me, watch out.

Then there's the obligatory list of famous people of this type, all of them great guys like Socrates, U.S. Presidents, and uh...the Olsen twins. Now it's one thing not to say that any of the types are the "jerk" type or the "homicidal maniac" type, that's defensible. But surely the more infamous historical figures do have types like the rest of us. I don't see the harm in including them. (I'll leave aside the question of how exactly it was determined that Socrates was an INTP.)

Now, all of this is indeed a separate issue from the validity of Myers-Briggs itself. As it happens, a New Yorker article by Malcolm Gladwell that I saw shortly after this argument addressed just that subject. He brings up this tasty statistic: "according to some studies, more than half of those who take the test a second time end up with a different score than when they took it the first time." The reason is twofold: first, our personality characteristics fall on a continuum rather than neatly in one category or another, so someone on the borderline gets simplified down and the result may not be reliable. Second, "personality is contingent, not stable, how we answer is affected by which circumstances are foremost in our minds when we take the test."

The other problem he brings up is that no one can determine objectively that the four axes used in the test are the four ways to talk about someone's personality. Anyone can come up with their own set of types, and there's no proof that one set is better than another, except that people's types come out the way you generally expect them to, which kind of tells you that you didn't need a fancy test in the first place.

To clear up one common misconception that I was glad to find out about from the New Yorker piece, the official test was not developed by doctors or professional psychologists. It was invented by a housewife (not that there's anything wrong with that) named Katharine Briggs who sought to understand the strange personality of her daughter's boyfriend, Clarence Myers. She started reading psychology books and came upon the theories of Carl Jung, which she then adapted into the test and the types. Jung never endorsed the idea that each individual was one type or another for life, and called the test a childish parlor game.

I went to the test's official site to try to determine how embellished or altered the type descriptions from that other site were. The only descriptions I found here were very short ones that basically regurgitated the names of the types.

The heart of JV's side in our debate was that he simply found the theory to be useful. Is it? I don't think it should be used as some shortcut to understanding others, as seems to be one major selling point. The fact that lots of big companies pay to have their employees take it doesn't convince me at all; big companies will do anything to feel like their decisions are somehow validated by an authority, preferably the same one all the other companies are using. It seems doubtful to me that it could be terribly reliable in determining who will do better in different business positions, as that's a whole other logical leap beyond what's in the types themselves. But, as with most all other pseudoscience, it's very difficult to convince people who believe that, despite what I might say, it is benefiting them in their lives. And sometimes (but not always) I think it's better not to try.

I remain convinced that Jung and the countless novelty websites have seen the test for its true (if not most profitable) nature: a parlor game.