(For the less geeky, the post title is “Models” — a play on Numb3rs)
For someone who deals with statistics only when I absolutely have to (the formulae make my head swimmy), I still have a fondness for doing comparative measurements. Most of the online personality-type tests are an absolute waste of time (I’d much rather work out a Sudoku), but once in a great while one will catch my attention long enough for me to actually complete it, such as the nerd test. Okay, so at a 93% I’m not as nerdy as Bug Girl, who earned a “Nerd God” score of 99!
On the other hand, last time I took the AQ Test (Autism Quotient), I scored a 45, which is pretty dang high … especially for a female.
And for a third item, my Myers-Briggs result is the rare “INTJ” (Introverted iNtuitive Thinking Judging), also known as “the scientist”. This is my favourite quote from the description:
Perhaps the most fundamental problem, however, is that INTJs really want people to make sense.
~ Marina Margaret Heiss
Yeah. I’m still working on that making sense-of-people business. Judging by the numbers of different kinds of tests out there for personalities and cognitive skills, so’s everyone else. Of course, any test we come up with is going to be a highly artificial construct, rather than something that truly describes what another person is really like.
That’s because all tests are based upon schematic models, and all models, whether they are from statistics, ecology, digital graphics, meteorology or any other field, are in many ways nothing more than carefully constructed lies. Models only set up values for a limited number of variables in a set. When we run a model simulation, we can throw in different values to see what the results are, but those results are in no way guaranteed to be “real”. Because all models are artificial, the validity of the results are limited to whatever factors we think to put in them. You cannot use a model to figure out what kinds of unknown factors you need to have; you can merely get a very different result than the reality, and then sit back and say, “Gee, what-all did we leave out?” Likewise, tests do not define people, but rather describe a particular facet, and how one person compares to others.
This is not to say that tests and models are useless. They give us transferable schema; little world-view models that we can share with one another and pass along to describe and more-or-less measure things. Being able to measure, describe and share information is important to communication and to the ongoing development of culture.
But no matter what sort of tests, measurements and models we use to describe special learning needs or autism or disability or imagination or intelligence or any other constructs, we need to remember that they can be excellent tools, and also that they are inherently limited. We want to avoid “maya”, or foolishly believing that our mental maps are the same as the territory or world that they model.
Enough of all that high-brow analysis — what’d you get?