Two key lessons about surveys:
Surveys often 'merge' data. The bus company survey covered in the previous post had bundled together the people who answered the survey as 'more and the less inclined', and the people who are classified as 'definites' and the 'probables'.
Blurring the middle rankings in a survey is common tactic, even in some serious surveys. Imagine a survey on the support for President Obama. Completely hypothetical, but let's say it asks if you are "pro" or "anti". There are three choices
1: wholly for,
2 wholly against,
3 partly for and partly against.
The result is that one third are wholly for, one third wholly against and one third in the middle. If you want to make it look like a vote for, you report that 67 per cent were "wholly or partly" in favour of President Obama. If you want to make it look like a vote against, you report that two thirds ("well over half") were "wholly or partly" against President Obama. It's common to try to amalgamate the undecided or half-hearted opinions with those who favour your stance on the matter in order to make the 'statistics' show that you have a clear majority on your side.
Another problem with surveys that often puts a spanner in the works is who they ask. In Northern Ireland a few years ago the government asked people to compile a voluntary survey on the police force. Who was surveyed? A cross-sample of the population. NO. How did they do the survey? By internet and post, with people who 'wanted' to take part. Not so good.
Even with a little thought you can see that those who most likely wanted to comment on the police force were not that happy with them. At that time one third of the community was against the police due to political allegiances. Anyone who thought logically about the situation would have seen the recipe for bias was in place before the survey began. All surveys can suffer from bias - deliberately or not - by picking up or missing too many of a vital groups and simply extrapolating data - as if everyone in the country thought the same way as the people who completed the survey. Good surveys require excellent planning......