We are a testing society. We use tests in schools and in the workplace and as a means to self-improvement. We like the scientific certainty that tests imply. In a funny way it’s comforting to “scientifically” know that we or others are of a certain personality type, have a certain IQ, that we or they have this or that aptitude, that we or they are dyslexic, have ADHD, are hyperactive or are high level functioning autistic.
It may be comforting to “know” these things about ourselves and others but all is not what it appears to be.
I wish people could see the problems with personality diagnostic tests, inventories, aptitude assessments, and the like. It’s not very difficult. These instruments are designed by someone or some committee, based on some theory they have adopted about characteristics that can be identified by means of counting and/or interpreting subject responses to the instrument. These responses are said by the designers to be indicative — to have predictive power.
The use of such instruments is fascinating. The principles of testing are predicated on the idea that by means of sampling, more general predictions can be made. The test is thought to be both “valid” predictively and “reliable” when administered repeatedly.
We can see how the methods we use to test “things” can be useful. It’s not too difficult to come up with criterion referenced indicators that can be said to be valid and reliable. But what happens when we apply this method to human beings? How well do we do when it comes to identifying valid and reliable indicators that predict human behavior going forward? By what method can we decide that our instruments actually predict outcomes going forward? Or is it possible that our instruments actually create outcomes going forward?
In point of fact, the predictive powers of testing used to classify and categorize individual aptitudes, styles and capabilities are built up as houses of cards, from the classification schemes employed to the methods used to sort individuals into bins.
Consider these tests, from IQ tests to personality inventories, and consider the methods used to determine that they are in fact, predictive. The conditions for assessing the predictive efficacy of these instruments over time are hopelessly complex.
Now think about the quacks who employ these instruments whose validity and reliability have no solid evidentiary basis. If the interpretations and predictions drawn from the use of such test are believed, then the future course of the person so reduced will be changed even if that person is the only one besides the measurer who is privy to the test outcomes and interpretations.
If I administer an instrument to you and you perform as required and then I make an interpretation, I am steering you forward based on the scheme I have concocted — based upon some theory for which I have no hard evidence.
I say this instrument tells us that you are artistic by nature. You are controlling by nature. You have an aptitude for math. (When they tested Einstein in the 8th grade he was deemed to have no aptitude for math!). Your IQ does not suit you for higher learning. You interpersonal skills need development. This control over what people become goes on and on, perpetuated by those who claim to be able to predict what others can or should become. Where do these tests and inventories and diagnostics really come from? Mostly nonsense sold to unwary subjects.
Now think deeply on this and consider the rites of passage that are virtually universal in human cultures. These are tests that mark the emergence of children into adulthood. Are these rites designed as tests that pass or fail people? Are they tests used to pigeon-hole individuals into certain bins? Or is it that by virtue of taking the test itself, the doorways to adulthood and the becoming that will follow are opened to the taker of the test? This is a very different notion of testing in which the taking of the test itself is enabling.
When it comes to predicting the future performance of individuals, our first principle should be “Do no harm”. If we want to use instruments to predict the future of individuals then we’d better have solid evidence that our instruments work and then solid evidence that the long term effects of our predictions make a better world.
The claims we make for the tests we create and administer shape the future of real human beings going forward. We’d better be damn sure we know what we are doing. In most cases we haven’t gat a clue.