Jobcentres have been foisting a bogus 'personality test' on the unemployed at the behest of Downing Street, bloggers uncovered  earlier this week. The tests spat out random 'personal strength'’ to jobseekers who were forced to take part; traits like 'originality' and 'love of learning' – with the feedback apparently having no bearing on the responses people gave.
The ‘My Strengths’ test came from the much-hyped No.10 'Nudge Unit ' (officially the Downing Street Behavioural Insight Team), which tries to use discoveries from the behavioural sciences to improve various parts of government. Why would they give the unemployed fake personality test?
One possibility that’s been floated  is that they’re pseudoscientific snake oil salesmen with no idea what they’re doing. This is entirely possible. Equally likely, the problems with the seemingly pointless test could also be down to an IT screw-up – it wouldn’t be the first time.
But there is one explanation that I can think of that does make some sense. It’s possible Nudge Unit was trying to use the power of stereotypes and framing to help people get jobs.
There’s quite a lot of evidence to suggest that people 'play up' to stereotypes they have imposed on them. The most famous related experiment is the Stanford Prison Experiment : a random group of subjects are taken and divided into guards and prisoners, then made to staff a pretend prison. After a few days, both groups, picked at random, internalise their roles and the guards are lording it over the prisoners – people play up to the roles you give them.
One piece of research  from  the University of Canterbury on gender stereotypes illustrates what Nudge Unit's thinking might have been. A group of participants were asked by the Canterbury researchers to do mental arithmetic, and were paid according to the number of questions they got right. They could choose to be paid in one of two ways: either proportionally, according to the number of questions they got right, or in competition with others. The second option would lead to higher pay offs for those successful but with more risk of being left with nothing.
The twist is that before choosing between the two pay approaches, the subjects were given one of two questionnaires. Half were given a questionnaire about their career, the other half one about gender and family issues.
The results are interesting: In the group given the career questionnaire, men and women were equally likely to choose the competitive approach – 25 per cent of each. But given the questionnaire which highlighted gender issues, there was a big gender gap: 37 per cent of men went for the high-risk approach, and only 7 per cent of women.
One plausible conclusion from these results is that even something like a questionnaire is enough to ‘prime’ people’s decision-making and get them to play up to a stereotype. So it’s not inconceivable that Nudge Unit thought that a ‘personality test’ exercise that emphasised personal strengths could have a positive effect on an unemployed person who believed themselves to be unemployable. Their plan was probably to get people to internalise the strengths given to them in the questionnaire and to play that role as someone with a ‘love of learning’ or ‘perseverance’, or whatever made-up strength they decided they want jobseekers to think they have.
So there’s a bit of evidence behind this, and the fake tests might not be entirely cranky – but that doesn’t mean it was a good idea. It still involved the government wilfully lying to people, which most people would probably think was pretty unethical. Less 'nudging' than 'making things up'. And giving people an inaccurate picture of their strengths probably isn’t very healthy anyway – overconfidence is just as much of a problem  as under-confidence; both are people not understanding their capabilities and lead to bad choices. One doesn’t necessarily compensate for the other.
Then there’s the fact that Nudge Unit seem to have quite ham-fistedly tried to take a laboratory finding and apply it to the real world without much skill. The questionnaire in the gender experiment 'primed' women to act less or more like their gender stereotype in the actions they were about to take, by changing the framing and context in which imminent decisions were made. But there’s no evidence it somehow permanently stripped them of their gender identity or made them think in more gender neutral terms for anything except the thing they were about to do. Unless the Jobcentre test was given to unemployed people directly before a job interview (it wasn’t) it’s difficult to see what the point would be, as it would be unlikely to have any long-term effect.
The transition from the laboratory to the field also failed in that by choosing to use fake tests, the initiative was set up to implode. It was only a matter of time before people figured out the tests were crap, and once it was common knowledge that they weren’t credible they’d look very silly and any benefits would disappear. This might not have been a factor in a controlled environment, but it is when you go out in the real world.
But worst of all, the whole approach of going to work on jobseekers’ egos is very arguably complete rubbish. It treats unemployment like an individual failing to be fixed at a personal level, rather than a problem of arithmetic where there are fewer jobs than the number of people who want them. This approach cannot work  as a strategy to bring down unemployment: if this survey ever did somehow boost someone’s self-image and land them work, someone else who would have otherwise got the job would just lose out instead. The idea that we could fix unemployment if only we would just believe in ourselves is a dangerous fantasy, and to be fair to Nudge Unit, it’s not one exclusive to them.