How normal are you?

12 11 2009

You’ll have run across psychometric tests – the Stanford-Binet IQ test, the Myers-Briggs Type Indicator. Increasingly, they’re used in job selection and assessment, career guidance, and university entrance. Maybe you’ve taken one, or had to review the results. Were you confident of what they were testing?

Psychometrics have had something of a bad reputation, due in part to the interest taken in them during World Wars I and II for classifying recruits, and the inevitable existence of numerous poorly-constructed “pop” psychology tests in the media and elsewhere. Properly constructed and validated psychometrics aren’t, of course, perfect, but they have been thoroughly and repeatedly studied and validated, and can’t be dismissed as uninformative.

What are psychometics, really? Simply, they are ways of measuring constructs, like “intelligence”, or “preference”. Crucially, they don’t measure against some abstract standard of perfection. Essentially, all psychometrics are based on the bell-shaped normal distribution, where the majority of the population are clustered in the middle of a distribution, with increasingly fewer results at both the top and bottom of the curve. The majority of people, therefore, will always fall somewhere in the middle of the distribution in any population. What a psychometric can do is to tell you where – according. of course, to the result you get on any particular day.

(A note on the word “normal”. In this, its most common use in psychology, it doesn’t mean “good” – it just describes the pattern seen across a population. It is an endless fascination to me how almost all terms meaning “common” or “usual” come to mean “good”, and “unusual” bad.)

A psychometric might measure intelligence, but it doesn’t measure what we all understand, abstractly, as the quality “intelligence”. It measures a precisely defined, constructed idea of “intelligence”, and as mentioned above, the crucial aspect is that it only measures your results on any given day. Anything you can define, you can measure – but when you define something you necessarily limit it. No psychometric definition of “intelligence” can capture all the facets that most people understand in that term. And most qualities aren’t fixed – nobody has just one level of “intelligence”. Your performance on a test depends on how motivated you are at the time, not to mention your existing knowledge, and maybe even how tired you are on the day. All tests are also filtered through the language they’re applied in, the culture they’re written for, and the assumptions they depend on, which is one of the reasons tests should always be taken in the participant’s native language.

Psychometrics can’t tell you how intelligent you are. But they can tell you where, on a given day, you rank in the population according to a necessarily imperfect definition that misses some things out. Maybe.




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: