Debunking the IQ myth

[napoleon_in_rags]

2. I do a little math.

And I can tell you a thing or two about it.

Suppose you have a 6-side dice. in the language of probability theory, what you have is "random variables" 1-6, (numbers on the sides of the dice) associated with a "probability density function", which ascribes a 1/6 probability to every variable, so a flat line if you plotted it.

But now suppose you have 2 6 sided dice, and your counting what they add up to when you roll them. Now random variables 2-12, and a distribution that looks like a triangle: low values for 2 and 12 (only one way to get each of those, snake eyes or double 6s') but a lot of different ways 2 die can add up to 7, so a peak there for probability.

Now suppose you have 3 six sided die. Random variables 3-18, starting to round up near the peak, flatten near the extremes.

And now suppose you have more. The shape you end up with as you add more die, converges to the "normal distribution", also known as the "bell curve". That's the shape that describes the distributions of IQ's. Yet by scaling the model and calculating the variance, you can always see any bell shaped curve as the sum of a number of independent variables, like counting the number of die that make it up. Thus there is a natural dimension of any bell -shaped distribution. what it represents its the n-dimensional space, where each test subject is evenly distributed in an n-dimensional space.

So what does that mean in practice?

I believe it means there is no such thing as IQ as a singular entity. It means what we call IQ is actually the sum of a number of totally independent variables. What could they be? Nutrition? Childhood inspiration? Its definitely worth finding them out. But until we know those independent dimensions for sure, IQ will always be a weak abstraction.

Peace!