"Are you the life and soul of the party? Do you think it is acceptable to beat a turnip?"
I'm writing an article about the use of psychometric tests in the workplace and I'm finding it hard to remain objective. In fact, it's driving me crazy. The tests themselves are like the grown-up version of the personality questionnaires that you find all over the internet (and which, prior to the advent of the internet, existed solely in the pages of magazines for teenage girls). I can't be bothered to find a new way of explaining how they work, so I'm going to plagiarise my draft article:
Psychometrics practitioners contend that their discipline is backed by a great depth of research - the biggest tests have been on the market for decades and are supported by thousands of past results carried out on sample populations. These past results provide “norm groups” for the tests and allow analysts to fit distribution curves to individual answers - so that responses can be seen to fall within different percentiles of the population. Some responses, and patterns of responses, are more common than others - and each test is backed by psychological theory which claims to explain what these patterns of answers mean. One pattern might mean that a candidate scores highly for “extroversion” for example. Another might mean that the candidate has a low rating for “altruism”. Testing firms try to demonstrate the validity of their tests by showing, for example, strong positive correlation between a high score for extroversion and a successful career in sales.
This is maddening in so many ways, but my main gripe is with the "reliability" of the tests - a term which, in this context, means that any candidate would give more or less the same answers regardless of when the test was taken. This rests on the assumption that the personality is something concrete and fixed. It also rests on the even more fundamental assumption that each question means the same thing to each person.
I don't see the personality as something immutable. I'm not even convinced that we have anything fixed enough to be described as a personality - meaning something singular and fairly easily-defined. People don't always behave the same way in the same situation - yes, person A may tend to be more chatty, or less defensive, than person B. But, in certain circumstances, A might well be less chatty or more defensive. Over time, for various reasons, the number of occasions when A is less chatty might drift. People change.
One of my more cynical contacts has described the discipline as being run by "charlatans and snake-oil salesmen". I agree. So, I'm not sure how to play this: one of the firms I'm talking with has offered to let me take their test, purely so I've got a clearer idea of what's involved and can see how the process works. I'm going to take them up on it, but there are a few paths open to me. I could ...
a) Answer as "honestly" as possible
b) Answer as "dishonestly" as possible
c) Answer as though I were a job candidate trying to provide the "best" answers
I'm really tempted to try and demonstrate (in a non-scientific, purely subjective way) that the tests are a load of rubbish. But how best to do that?
Psychometrics practitioners contend that their discipline is backed by a great depth of research - the biggest tests have been on the market for decades and are supported by thousands of past results carried out on sample populations. These past results provide “norm groups” for the tests and allow analysts to fit distribution curves to individual answers - so that responses can be seen to fall within different percentiles of the population. Some responses, and patterns of responses, are more common than others - and each test is backed by psychological theory which claims to explain what these patterns of answers mean. One pattern might mean that a candidate scores highly for “extroversion” for example. Another might mean that the candidate has a low rating for “altruism”. Testing firms try to demonstrate the validity of their tests by showing, for example, strong positive correlation between a high score for extroversion and a successful career in sales.
This is maddening in so many ways, but my main gripe is with the "reliability" of the tests - a term which, in this context, means that any candidate would give more or less the same answers regardless of when the test was taken. This rests on the assumption that the personality is something concrete and fixed. It also rests on the even more fundamental assumption that each question means the same thing to each person.
I don't see the personality as something immutable. I'm not even convinced that we have anything fixed enough to be described as a personality - meaning something singular and fairly easily-defined. People don't always behave the same way in the same situation - yes, person A may tend to be more chatty, or less defensive, than person B. But, in certain circumstances, A might well be less chatty or more defensive. Over time, for various reasons, the number of occasions when A is less chatty might drift. People change.
One of my more cynical contacts has described the discipline as being run by "charlatans and snake-oil salesmen". I agree. So, I'm not sure how to play this: one of the firms I'm talking with has offered to let me take their test, purely so I've got a clearer idea of what's involved and can see how the process works. I'm going to take them up on it, but there are a few paths open to me. I could ...
a) Answer as "honestly" as possible
b) Answer as "dishonestly" as possible
c) Answer as though I were a job candidate trying to provide the "best" answers
I'm really tempted to try and demonstrate (in a non-scientific, purely subjective way) that the tests are a load of rubbish. But how best to do that?