That is the science of voting prediction, well many regard it more as a "black art". It has been notoriously wrong on a number of occassions, the election where Kennet got ousted in Victoria for example. The polls were predicting he would be returned with an increased majority.

One of the most obvious reasons for this is that people don't end up voting the way they said they would. It may be lying but more often it is probably they were not as firm in their opinion as they said they were.

Further to the previous post about telling when someone is lying Drazen Prelec of the Massachusetts Institute of Technology recons he has come up with a way that he believes eliminates the effect of lying and indecision by people on the forms and questionaires they fill out.

The problem is, according to Drazen, that people tend to answer in a way that they believe is expected of them, particularly when they are undecided. With Drazens approach a second question is asked of the responded along the lines of what they think other people will say. This question is used to alter the weighting of the response to the first question.

Here is how it works:

For example, to find out whether people think Picasso is a great artist, you ask people whether they like Picasso, and also how many people in the population they think are likely to answer yes or no to that question.

A person who believes more people believe the same as them is more likely to be hold a firm opinion and be telling the truth. This is because you hold the opinion for reasons you believe to be valid. Because you believe your reasons are valid it is a logical extension to believe many other people will have come to the same conclusion. Given this you tend to overestimate how many other people share your opinion.

If you are less sure you are likely to pick a lower percentage of people who share your opinion.

Drazen recons it works best when there is little information about what the public, as a whole, thinks about a topic. It would work well immediately after a televised debate for example.

If he is right and his system works then he stands to be a very wealthy young man very quickly.

Read more in Nature.