Assessment is not a verdict

There is a way of using psychological assessment that we do not practise, and it is common enough to be worth naming. In it, a student sits a test, a report is generated, and the report announces what they are suited to. The test has functioned as a verdict. The student's part in the process was to supply the answers.

We think this is a misuse, and not a small one.

What the instrument actually produces

An instrument produces evidence about how a person performed on a defined set of tasks, compared with a defined group, on one occasion, with a known margin of error. That is a genuinely valuable thing to have, and it is not obtainable any other way — no amount of conversation with a student yields a comparison with a thousand others of their age.

But it is evidence, and evidence is an input to a decision. Every step from evidence to conclusion requires something the instrument does not contain: what the person is trying to achieve, what they will actually work at, what their family can support, what is available where they live, and what they are willing to give up. None of that is measurable and all of it decides the outcome.

A score tells you something true about a person. It does not tell you what they should do, because what they should do depends on things a score cannot see.

Why the verdict model is attractive

It is attractive because it removes the hardest part. A verdict ends an argument. In a household where four people disagree about a child's future, a printed report that names an answer is enormously appealing, and there is a commercial incentive to supply it.

It is also attractive because it relieves the student of responsibility for the decision, which feels like kindness at fifteen and does not at twenty-five. A decision handed to you cannot be revised by you, because you do not hold the reasoning it was built on.

What we do instead

We measure, we interpret, and we sit with the person while they decide. The report is read with them, not delivered to them. The uncertainties are stated rather than smoothed over — where the scores are close enough that the difference does not mean anything, we say so. Where the evidence points somewhere the student does not want to go, we say that too, once and clearly, and then we discuss it rather than repeating it.

The measure of a good session is not that the student leaves with an answer. It is that they leave able to explain their reasoning to someone who disagrees with it.

What this costs

This position has costs and it would be dishonest to omit them. It is slower. It requires a qualified person's time, which is the expensive part, and it cannot be automated as cheaply as a report generator. It does not produce a satisfying single sentence for the family. And it sometimes ends with a student who has decided something we would not have chosen for them.

We think that last one is the point rather than the failure. The decision belongs to the person who has to live inside it. Our work is to make sure they made it with better information than they would otherwise have had.