Tuesday, September 14, 2004

Jumping through hoops

It's the season for reading: having ploughed through The Wisdom of Crowds yesterday, and The Da Vinci Code and Angels and Demons in my mini Dan Brown marathon last week, I've just finished How Would You Move Mount Fuji? (subtitled "Microsoft's Cult of The Puzzle").

In a nutshell, the book examines the types of logic and lateral thinking puzzles that Microsoft is famous for asking candidates during job interviews. The title is one example. Others include:

  • Which way should the key turn in a car door to unlock it?
  • How many times a day do a clock's hands overlap?
  • How would you locate a specific book in a big library, without a catalogue or a librarian to assist you?

It tries to both analyse the history of how puzzles came to be part of the Microsoft interview (and in so doing gives a short history of IQ testing), as well as giving examples of the kinds of puzzles that are used, and a critique of them. (an excerpt for those who'd like to read more)

One of the myths/mis-conceptions debunked throughout the book is the notion that the ability to solve these puzzles has anything to do with predicting the potential employee's future job performance - in fact, the (reasonable) conclusion it reaches is that these logic puzzles serve at best to highlight unsuitability - i.e. if you can't solve the puzzles, you're likely to not do well in the job, but if you do solve them, or solve them well, it doesn't correlate to better performance either.

I must confess to being intrigued by these kinds of puzzles - I like the challenge of solving them, even though I only do moderately well on lateral thinking puzzles. (I've got better over the years though, which highlights another flaw with the assumption of IQ/intelligence testing as a measure of "raw" or "innate" intelligence - trainability) The use of IQ, logic, or lateral thinking puzzles is just another expression of the assumption that selection for jobs/placement in schools/etc can be done by making candidates jump through increasingly difficult hoops, until the required number are left. (I had to jump through one of these hoops when I was 12 to get into my secondary school, and again at 18 on the way to getting my scholarship)

The problem is that hoop-jumping has little or nothing to do with what the candidate is being selected for - unless your future job is to solve IQ puzzles full-time, then your performance in solving these very abstract tests is mostly irrelevant.

It would be somewhat relevant if the tests measured generic intelligence, but that concept is itself fraught with pitfalls and traps. We're each intelligent to varying degrees in so many different ways that it's a poor fit to label someone as "dumb" or "smart", and the world we operate in has so many different requirements for these varying expressions of intelligence that actual performance in life depends more on context than on IQ - i.e. being in the right place at the right time (such that your talents are relevant) is more important than being high-IQ (which explains why success in life doesn't correlate to IQ, and the existence of high-IQ plumbers as well as professors).

As a teenager, I was fascinated by IQ tests: they provided another measurable component in a highly competitive academic environment, and I (and quite a few of my friends) actively sought them out. Over the years though, they've become increasingly irrelevant when I realised that they didn't have anything to do with the really important things about a person - like integrity and soundness of character, for example. Or whether that person was nice. Or had leadership ability. Or had huuuge ..... tracts of land (depending on what you were looking for =) I still like to try these puzzles though, but I view them as mental gymnastics more than anything else. A similar motivation lies behind the wargaming as well.

(Wargaming is more fun though - pit your wits against someone else, and play with toy tanks: could it get any better?)

No comments: