The DWP's bogus psychometric tests are dystopian, but randomised trials aren't the problem

The solution is more trials, not fewer.

New Statesman

The DWP is under fire for mandating benefit claimants to carry out bogus psychometric tests. The Guardian's Shiv Malik reports:

The test called My Strengths, devised by Downing Street's behavioural insights or "nudge" unit, has been exposed by bloggers as a sham with results having no relation to the answers given.

Some of the 48 statements on the DWP test include: "I never go out of my way to visit museums," and: "I have not created anything of beauty in the last year." People are asked to grade their answers from "very much like me" to "very much unlike me".

When those being tested complete the official online questionnaire, they are assigned a set of five positive "strengths" including "love of learning" and "curiosity" and "originality".

However, those taking the supposed psychological survey have found that by clicking on the same answer repeatedly, users will get the same set of personality results as those entering a completely opposite set of answers.

The aim behind the "test" is apparently to prevent the claimants with motivating statements about themselves, in an effort to encourage them to find work. The behavioural insight team itself describes the process as:

Building psychological resilience and wellbeing for those who are still claiming after 8 weeks through ‘expressive writing’ and strengths identification.

But it's worth being clear about why "My Strengths" is so unpleasant. The problem with it is that unemployed people are being lied to in an effort to psychologically manipulate them into a mental state where they will do what the Government want. That sounds like a sentence from a dystopian future, but it's a fairly straight description of what is happening.

But Steve Walker, the blogger who broke the story, goes further, attacking the test because it's part of a government trial.

Walker writes that the trial breaches principals established by the EU's body for co-ordinating R&D which state that the medical rule of informed consent "remains valid for any other kind of research".

Principles aren't laws, so there's not a legal case to answer here; but even if we're just arguing morality, establishing medical-grade requirements of informed consent in social research would be a terrible idea.

The rise of randomised controlled trials (and, to a lesser extent, randomised trials and controlled trials – not everything can be both randomised and controlled) in social research is one of the most promising trends of the last decade. It's good not only because it can increase our knowledge of how to fight social problems in employment, education and crime, but because it's barely different from what's being done anyway.

Most providers of the government's work programme – which involves more than just the unpaid mandatory labour that has made it infamous –  have neither the staff, time or money to offer every service to every person who comes through their doors. When they're trying to decide who gets to go on the CV workshops and who doesn't, the obvious thing to do is to chart which groups improve the most when given the workshops. This is doubly true if you don't actually know whether CV workshops help or harm; keeping a good eye on the results is invaluable.

When it comes to trials, the biggest crime of the BIT isn't that they did them, but that they didn't do them enough. As a report from December shows, the team aggregated together three completely different changes:

  1. Making sure every customer talks about getting back to work on their first day (not after 2 weeks) by cutting down and reorganising processes;
  2. Introducing stretching commitment devices which focus on what the job seeker will do for the whole of the next fortnight. This replaces the present system where advisors ask if job seekers have done three job search activities in each of the previous two weeks;
  3. Building psychological resilience and wellbeing for those who are still claiming after 8 weeks through ‘expressive writing’ and strengths identification.

Overall, apparently, "job seekers in the treatment group are 15-20 per cent more likely than those in the control group to be off benefits 13 weeks after signing on" – but the trial is so badly designed that we can't see which of these treatments helped the most. It could well be the case that the My Strengths test actually hindered jobseekers, but the effect was hidden by the benefits of "cutting down and reorganising processes".

The government needs to behave ethically when dealing with its citizens, and that goes whether or not it is done in the framework of a formal trial. But if it does try a new ethical way of helping jobseekers, the last thing we want to do is discourage it from measuring the results.