Hacker News Reader

Comment by sanxiyn | Hacker News Reader

Summarizing a complex property into a single numerical value is very useful and not silly. For example, temperature is a numerical summary of huge number of molecular motions.

More on this here: http://slatestarcodex.com/2014/08/11/does-the-glasgow-coma-s...

View on HN

I didn't say that all attempts to summarise a complex system in a single number are silly - just that intelligence is such a multi-faceted and ill-understood area that, in my opinion, evaluating people based on a single number is silly.

NB I say that as someone who got a very high IQ test result - which didn't exactly convince me that IQ tests are a good idea....

What is the name of the IQ test you took?

Intelligence really isn't "multi-faceted" (read Gardner's own admission that his theory never panned out) and it isn't ill-understood (refer to the Nature or Nurture interview with Nancy Segal on YouTube).

There's two reasons people say that. One is, they fared badly on a test and want to dismiss it, and the other is, they fared well on a test and are bashful about it.

Also, IQ tests are meant to measure a person's intelligence, not to convince them that IQ tests are "a good idea" - for that you would have to study Psychometrics.

IQ tests are somewhat ridiculous to begin with. The first tests that eventually became IQ tests were not intended to measure intelligence. They were intended to measure school readiness of rural children whose birthdates were often not known with certainty. The lack of clear birthdates meant an age cut off could not be used and rural children faced cultural differences from kids in the big city (namely: Paris) that created inherent challenges to them fitting in and doing well in school.

Intelligence is neither well defined nor well understood. It is fairly controversial stuff.

"it isn't ill-understood"

I worked on AI research for a number of years and my strong opinion is that our understanding of general intelligence is, as some wit put it, "pre-Newtonian".

What we mean in common language by intelligence may not be a single property. It might be more analogous to smell than temperature. An inherently multi-dimensional phenomena that can't be accurately described by a scalar measure.

It's actually not that difficult to turn a multidimensional measurement into a scalar quantiry; just represent the measurement as a vector composed of the deviation from median on each axis, then find the magnitude of the vector

But you're right that this doesn't tell you much about any of the individual dimensions. Maybe adding a variance across dimensions, so two numbers, would be more useful.

Of course, but you first you want to find the most accurate representation, then reduce it to something simpler as required.

Reply on news.ycombinator.com