About the Psych tests.

Cecil's Columns/Staff Reports
1

Erm. I briefly worked in Human Resources for the transit agency of one of America’s largest cities, and when I first started our hiring for bus drivers was based on a psychological test somewhat similar to the Briggs-Myers. (Not entirely, but… somewhat.) Not long after I started they switched to a different sort of test, but I think it’s worth noting that people were being hired based on this sort of profile.

It’s only harmless until someone who ought to get a job can’t. And lord knows, HR people can take all kinds of dumb things way too seriously. I think they must have just felt a lack of some sort of qualitative criteria but had no other option than this.

2

I’ve had to take these tests but they are always so poorly written that there’s no way to answer accurately. “Do you feel it’s ever possible to steal for a good reason?” How do you answer that? Obviously the answer should be no (if you want the job), but if you were in Nazi Germany, there could easily have been times when the answer would change. And they ask the same questions over and over again in slightly different ways to try to trip you up in case you are answer what you think they want to hear. And there’s no grey area; it’s all absolutes which don’t exist in real life. I have refused taking these tests in the past, but nowadays I don’t have that luxury. I helped my teen daughter reason through one of these employment screening tests at Target.com. I was embarrassed for them and her.

3

Which test have they switched to?

Many of them have a strong ethical component - but others don’t. The Myers-Briggs, for example, should never be used for recruitment - it’s for team building and personal development. Others can be, however. (And also some HR departments ignore the ethics.)

This is quite deliberate. They’re not necessarily asking about the specific circumstance in that specific test, they’re asking a series of similar questions to get a cumulative effect. And the results can definitely be staggeringly different between different candidates.

Properly practiced these tests can in fact be very nuanced. The “good” ones (properly researched and validated) are accompanied by interaction with a psychologist, and the candidate is the only person who can conclude their personal preferences, regardless of what the test result says.

That said, there is a bunch of bullshit out there, where the test has been pulled out of someone’s ass and is backed up by no research at all. If you’re in HR, it behoves you to choose a decently validated one - because a bad one can hurt not only the candidate, but the company too in rejecting good candidates and suggesting bad ones.

4

Oddly enough, everything you’ve said above matches with what my wife has said about psychics (there’s a lot of bullshit out there, so you need to get a “good” one; a tarot card reading/I Ching reading has to be accompanied by interaction with a professional, etc.).

-hogarth, INTP

5

Agree with Osomatic. I think the problem is the column was missing Cecil’s UC (unflinching cynic).

6

Looks like it’s up to me to provide the article link per the rules.

7

I want to note that a major problem with these tests is that they are more of a test of self perception than testing an objective reality. Assuming the person is answering honestly (the ability to game these tests is really not that difficult), the person’s self perception might not be very telling.

In some ways it’s like telling a person to think of a number between 1 and 10 and then asking several questions about whether it’s odd or even and between 1 and 5 or 6 and 10. Soon the test can determine the number the person had in their mind and say, “you’re the type of person who chooses [insert your number here]”. Wow! It knows you so well.

I told it I’m the life of the party and like interacting with all kinds of people and it says I’m an extrovert.

8

Mrs. Cad is currently in this position because on of the hiring guys thinks that work is a social event and she scored low on the sociability section which apparently outweighed her 20 years in the field, highest level of certification in her profession, and previous experience in a similar position. Oh and its for a management position too.

9

Cecil seems to confuse Introversion/Extroversion with Shyness/Outgoingness. The continuum of the former is harder to define than the latter.

10

Something I’ve noticed about the Myers Briggs personality test is that the INTJs can’t shut up about taking the Myers Briggs personality test.

Serious. INTJ is supposed to be 3%; the only people I’ve ever heard talking about it are INTJ.

11

Allow me to introduce myself as a mild Myers-Briggs convert.

Like many, I ridiculed the hell out of the test for years. Every time I got stuck taking one (high school, college, other times), I scored differently each time. When you look at just the “basic” test, where it assumes rigid either/or categorizations, the whole thing appears ridiculous. Plus, at its heart, it comes off like a psychic/astrological reading: so blandly generic that the descriptions apply to anything.

Bottom line: I’m skeptical about everything, especially M-B.

That said, last year I took the test again for a month-long training course, and count me as a mildly redeemed skeptic. First off, the test in 2009 was far longer than the ones I remember taking (larger sample size?). Second, the interpretative material was far more robust-- sure, you still get your type out of 16 choices, but the real meat of the test was where those 16 choices overlapped. Also, a lot more time is spent on differentiating between strong and weaker scores, e.g. I’ve always floated between Extrovert and Introvert in past tests, but this one showed where I fell on the scale (in the middle) and explained how often one can move back and forth between them (i.e., the test no longer assumes rigid, never changing types but allows for movement over time).

Long story short: the test is far improved than it used to be. I’d still be hard-pressed to call it scientific in any way, but for the first time I will say it accurately described my personality (I asked friends & co-workers what type they thought I was without showing them the test results, and mostly it was a match. Of course, if my friends know me that well, why’d I have to take the test? ;-).

Anyway, as a “team building” exercise, it’s harmless fun-- BETTER harmless fun than it used to be, IMO. As anything more, I’d never use it, and poo poo any HR or Student Life guru who swore by it for anything more rigorous.

12

The thread I started recently had some discussion of this (as was my original intention).

13

When I ran into this test, my thought was the old joke:
“There are two types of people in this world - those who shoehorn people into categories and those who don’t.”

It’s like the old “when all you have is a hammer, everything looks like a nail”. If you are determined to push people into one of 16 pigeonholes, then one way or another, you will define things so everyone fits somewhere. Each of the 4 scales is a continuum, and your mileage may vary day to day.

I suppose - if the questions are answered honestly - that it is indicative of what personality you have. But as Cec points out - everyone is a winner. Thief, saboteur, brainiac, expert or leader are as likely as not to be any of the categories. Some may be better suited for some jobs; I am curious what they though was suitable and unsuitable for bus driver.

Also, as Cec points out, your behaviour may vary depending on the situation. I hate social settings, I can’t remember names, I probably have moderate Aspergers’, I prefer computer programming - yet my boss initially pegged me as extrovert before the test, because with a group I’m familiar with (like co-workers) I can be the biggest joker around… although a girl back in high school pegged me pretty fast - “you use humor to hide your shyness…”

Think of psych tests as like an indicator. Let’s call them “quack tests”. You know - “if it waddles, like a duck, quacks like a duck, and looks like a duck - it’s probably a duck”. All the psych test gives is a quiet ‘quack’. To be sure of anything, you need a lot more testing. However, the first thing humans learn when they mature (most of them) is how to fake sincerity, so all the test will tell you is whether they care enough about getting the job to lie.

14

:dubious: You know psychics who actually run double-blind tests to see if their predictions are accurate?

15

Which is why you shouldn’t be using a psychological test for a job search. You need a test specifically designed (and tested to work) for the purpose you are using it.

In this one, I’d suggest one that has trick questions designed to trip up the brown nosers who just say what they think they are supposed to.

16

hogarth, just because the claims that psychics make match up to some fo the claims that are made about psychometrics, doesn’t mean they’re the same thing.

This is what got me over my initial skepticism. Cecil sites “one study” that shows a correlation between tests; there are in fact hundreds of similar validity tests out there. The descriptive reports are absolutely not Barnum statements, though I at first thought they were. And there are clear clusters of “type” around certain professions - the research has been done, published, and peer reviewed.

Sure, in the case of the MBTI, as Cecil says, the four dichotomies were pulled out of Jung’s arse, but that doesn’t necessarily invalidate them: to an extent, they’re just codified common sense. But the MBTI is just one of many, and many of the more recently devised tests are based on the “big five” model of personality, listed in the article:[ol][li]Openness to Experience[/li][li]Conscientiousness[/li][li]Extroversion[/li][li]Agreeableness[/li][li]Neuroticism[/ol]Do these things exist in human personality? Of course they do. Are they measurable? Of course they are. [/li]
Not just that, but most of these diverse tests’ scores stack up well against completely different (research-based) psychometric tests performed on the same candidates.

Like one of the guys training me said, “if you slice the salami of personality one way, you get a cross section of certain aspects; if you slice it a different way, you get another cross section”.

17

The result of an individual Myers-Briggs assessment indicates the result of an individual Myers-Briggs assessment. That is all it indicates, that is all it can indicate, that is all anyone should say it indicates.

If you want to believe it correlates with anything else, or that it reliably informs you about something (such as your personality or your suitability for a given role), then you will see evidence to support this belief. If you do not, then you will not. In other words, the utility of an MB assessment is as real as you want it to be, no more and no less. The same goes for its purpose or reliability.

The fact that some consultant and industrial psychologists use MB and other assessments does not mean that these assessments tell you anything reliable. It means these psychologists know they can be paid to administer these assessments, lecture on them, report on them and compile ‘reports’ based on them. That’s all.

There is no good evidence or good reason that supports any alternative contention. If you doubt that this is so, or if you think this is an extreme viewpoint, you may like to read the following.

I am friends with one of the UK’s leading industrial consultant psychologists. He works for a company that, among other things, lectures on the efficacy and usefulness of MB and other types of tests and assessments. He also receives significant fees for administering these kinds of tests/assessments and producing reports based on them.

Three years ago I offered him a friendly bet, challenging him to produce one single shred of credible evidence that a company that uses these kinds of tests is in any way better off than one that does not. For example, can it be shown that the company achieves lower staff turnover, higher ‘job satisfaction’ ratings, better management performance, better market share, better profitability or some other such sign? Three years have passed, and I am still waiting. Nothing. And this is a guy who has access to all the relevant research, including research that isn’t public domain.

18

Another nitpick with the article.

This is incorrect with regard to the MBTI proper. While the ethics of the MBTI are indeed non-judgemental, the “champion”, “inventor” etc. definitions quoted here are derived from the Keirsey Temperament Sorter. This is an independent methodology that overlays those defnitions onto MBTI character types. But they are absolutely not from the MBTI itself. Can I request a correction?

19

Where did jjimm say anything about double-blind tests in his comment?

I think you’re misunderstanding me: I’m not saying the the MBTI is useless like a psychic reading is. I’m saying that a comment like “90% of [DUBIOUS_THING] is bullshit, you need to get the real McCoy” is a terrible way of defending something. It’s the “No True Scotsman” fallacy in a nutshell.

20

Which should itself be evidence that the test means something– If nothing else, it’s able to pretty accurately identify the sort of folks who tend to talk about psychometrics.

And for those saying that it’s possible to “game” the test in order to get the “right” answers, you’re completely missing the point. There are no “good” or “bad” personality types (at least, not as indicated by Myers-Briggs), so you shouldn’t be trying to get one of the “good” ones. The correct response is the one that actually describes you, so by gaming the system, all you’re doing is making your results worse.