Don't forget that it helps to get the answer you're looking for if you ask the questions the right way. If I was doing a survey on speed enforcement and I wanted to be sure it'd broadly support scameras I might go about it like this:
How many children are in your family? This is simply to make people start thinking about their kids, or other young relatives. In itself it's irrelevant, but it's important that people they care strongly for are in their minds when they're considering later questions.
Do you believe child casualties are too high? Most normal people would consider one child fatality as one too many. The only people who're going to think that casualties aren't high enough will be regular starit jacket wearers. This question is asked simply to get the respondent to start agreeing with the questioner. It's actually about as relevant as asking if you prefer noodles or starvation.
Are you concerned about the level of serious injuries among children? Very similar to the previous question, and designed to provoke a second successive agreement on the part of the person being questioned.
Does the number of children currently being killed or injured by speeding motorists worry you? Now, there's no mention of what the number of children killed or injured by speeding actually is. Nor does the question ask the respondent to think about what this figure might be. We're simply trying to get another positive response here, and again one is one too many for most people so we're likely to get it. Whether the numbers involved are significant or not just doesn't enter into it. At the same time we're linking child casualties with speeding for the first time, having first got them to think of their own children (if any) and got them to agree with us twice before.
Do you feel that enforcement of speed limits can help reduce child casualties? Well, all the people who've just said yes to the previous question are not likely to change their minds at this point and say that enforcing limits is useless, so they're probably going to say yes again. Inconsistency won't make you look too bright, and nobody wants to seem stupid to a stranger.
Do you feel that police and/or safety cameras near schools, playgrounds etc would help enforce speed limits in areas where children are vulnerable? Again, fishing for a positive response. No one is likely to go past a school at flat chat with a police car about, so logically this is going to help enforce the limit. And now we're equating the presence of police with the presence of cameras. We're also trying to get emotions going by talking about children and vulnerability.
Do you believe that the risk to children is reduced by police use of safety cameras? Again, we're tying the police and the cameras together, and again we're hoping that people who answered yes to the previous question will do so again out of consistency. It helps that the question is almost the same as the last one, just worded differently. We're also keeping the emotions stirring by using the phrase 'risk to children'.
Do you feel that police use of safety cameras should continue? Having got tentative support for scameras in the last couple of questions we're looking for fairly unequivocal support with this one. By asking a series of questions that people are likely to say yes to, hopefully more people will say yes to the last one than would be the case if the survey consisted of this question alone.
No real information is given out as part of the questions, so we can rely on public misconceptions supplying the answers to a degree. Also, by using words like feel, believe, worry, concern, none of the questions are actually asking people to think - they're just asking for people's feelings. And I wouldn't bother publishing the whole survey either. The last question and the percentage of yes answers is what I'd put on the web FAQ, along with maybe the question that deals with being concerned about speeding and child casualties. Hey presto, public support for cameras.
Now, I'm not suggesting that this is actually done. The only time I get asked to do a survey tends to be when the phone rings and dinner's on the table, so I usually tell them to get lost without even waiting to find out if the survey is about speeding or my favourite yoghurt. I've no idea what sort of questions are really asked in camera support surveys. But if I was doing it I'd be tempted to think very hard about what sort of questions will lead to the result I want. Then again, I am a sneaky *******.
