garote: (Default)
garote ([personal profile] garote) wrote2007-08-15 08:47 pm

Why IQ testing and MENSA can bite me

I saw this question once, on a test from MENSA:
Which one is most different: A, F, N, H, E
It was a multiple choice question. Apparently there was one correct answer, and four that were wrong.
I considered each letter in turn for a couple seconds:
  • A, because it’s the only letter that creates an enclosed space
  • F, because it’s the only letter that doesn’t have symmetry
  • N, because it’s the only letter that can be drawn in one stroke
  • H, because it’s the only letter with symmetry on both axis
  • E, because it’s the made up of four lines (all the others use three)
This question neatly encapsulates everything that I think is wrong with using a standardized test to judge ""intelligence"". Regular tests are all about recalling a fact, or applying a procedure to arrive at a solution. But an ""intelligence"" test is supposedly designed to test precision and cleverness, and be as independent of cultural bias as possible. Instead, every ""intelligence"" test I've come across has been an exercise in divining which approach, among many, the entity scoring the test expects you to use. Which can easily become a matter of cultural bias.

What is the point of questions like this? Of tests like this?

To clarify:

[identity profile] ex-zeugma416.livejournal.com 2007-08-17 01:06 am (UTC)(link)
My objection was to your characterization of MENSA as being "just like" the examples you mentioned, and to nothing else. Political influence is a non-issue, and I'm not sure why you've brought it up in response. MENSA more resembles a country-club type of organization than (say) a beer-brewing group or a congregation. To equate the former type of organization with the latter two is to miss an important distinction. That is my single point. Whether they do good or ill is nothing to the purpose. I stand well corrected on the testing issue, but the error hardly invalidates the point.