A poll published in The Times today (subscription required) once again told us that the public are an angry lot who would, if they had their way, lock up everyone who broke one of the ten commandments for a very long time indeed. The Times is a paper that has to make money so I won’t spill all the beans here. But I will let myself use some of their material to comment on the polling company Populus who asked excellent questions on their behalf – that is if the purpose of surveys is to give the person who commissioned the survey exactly the results they want to hear….
Let’s take the first question first:
Q. Judges are out of touch and politicians should make sure they do not let criminals off too lightly.
A. 85% agree, 15% disagree
Well, surely such resounding agreement is definitive proof that the public think British Justice Secretary Ken Clarke has gone mad by suggesting that locking up more and more people each year without reforming them (or rather than preventing crime in the first place) is probably a monumental waste of public money? Mmm… or could it be that this is possibly the worst survey question I have seen since ‘Scotland has its own parliament. Should England have it’s own parliament too?’ in the run-up to devolution. First, there are two questions here (sorry, Populus, but last time I checked, that’s not exactly survey best practice). Second, the second part of the statement contains a value judgement ‘too lightly’, which must surely skew the responses to an enormous degree. It’s a bit like me asking ‘If teachers are allowing too many children to fail in school, should the government intervene?’. Erm…
There’s another question, that I’m not sure about – but may be told is perfectly acceptable. Views VERY welcome.
Q. The government should not increase the number of prison places because it would mean further cuts to public services elsewhere or more borrowing
A. 45% agree, 55% disagree
I like what the questioner is trying to get at here but just worry that the number of variables in the question is too high and the wording a bit confusing. I basically can’t figure out whether the answers to this question are judgements on the importance of prison, the importance of deficit reduction, or a reflection on the logical statement that more prison means more cuts elsewhere. No doubt it is a judgement on all of these – and hardly a stark judgement at that.
Then there’s an excellent question about what the public thought would do most to reduce crime, followed by some options. Except, wait! The options are the most unmitigated pile of nonsense I’ve ever encountered in my life.
Q. Which one would have the single biggest impact on reducing crime in Britain
a) Making punishments harsher to deter offending (40%)
b) Scrapping the human rights act (10%) (TG: what?! what?! I had to read this about 6 times to work out whether it was serious and to consider how 10% of respondents actually thought that this might be the best way of reducing crime in Britain)
c) Putting more emphasis on rehabilitating offenders to reduce the likelihood of reoffending (16%)
d) Recruiting more police (17%)
e) Giving the police more powers (17%) (TG: invisibility perhaps? walking on water?)
The fact that people preferred tough punishments to more police/ more police powers (40%, 17%, 17%) flies in the face of the vast bulk of academic theory and evidence, which generally shows that certainty of punishment is a better deterrent than changing the harshness of the punishments if caught. (It’s actually pretty obvious that you only think about punishment severity if you actually think there’s a reasonable likelihood you’ll get caught – if you’re thinking about consequences at all). But, I’m not going to go too much into why the public always get things like this wrong. Instead, suffice to say that these limited options (which respondents were forced to pick from) are clearly bonkers. What about more parenting classes (an approach most evidence supports)? What about better education (something the public might go for, based on other polls)?
My point here is that bad polling gets you bad results – or rather that the question you ask dictates your answer. Pollsters know this, as do journalists, and so they are rather good at conducting polls that make headlines. But you see, I think I could make my own headline, proving that the public hate prison and want a preventative approach to crime policy… watch this space…