Skip to content
Join our Newsletter

The science of voting

Everyone should vote. It is a privilege and a right that we have within our country. It allows us the freedom to elect who will govern. It is not an onerous task. Every vote counts.
col-whitcombe.13.jpg

Everyone should vote. It is a privilege and a right that we have within our country. It allows us the freedom to elect who will govern. It is not an onerous task. Every vote counts.

Why write a science column about voting?

It turns out that voting is actually quite scientific. It is based on statistics and probabilities. It is mathematical in nature. And there has been a lot of research done on everything from why people vote to how they vote to what are the best methods of voting.

A quick search of Google Scholar turns up over 1,000,000 articles on the subject of voting science.

In its simplest term, voting is an enumerated count of all of the individuals within a group that are eligible to vote. It should involve nothing more than counting. If there are 21 people in a group and everyone votes, then ideally whoever wins will have a majority of support.

In practice, it doesn't tend to turn out that way.

This is particularly true when there are more than two candidates involved in the election. In that case, the winning candidate could be preferred by less than 50 per cent of the voters - sometimes way less. For example, in a three candidate race with 21 voters, the winning candidate only needs the backing of eight people assuming the remaining vote is split. A total of eight out of 21 is only 38 per cent of the voters.

This is one of the arguments that is used for a two party system or a series of run-offs in which the lowest polling candidates are progressively removed from the list until just two remain. Then the winner must have a clear majority. Or so you would think.

But what happens if not everyone votes? Or if after the first ballot, your favorite candidate is no longer eligible? Both situations can easily lead to an elected official lacking in popular support.

For example, in the last municipal election, only 28 per cent of the population of Prince George voted. The vast majority of people in our city did not participate. In the last election, our mayor only received 6,969 in a city with 60,000 or so eligible voters. Much more than the runner up with 5,332 but not even close to a majority of voters.

From a statistical point of view, 28 per cent might represent a sufficiently large enough population sample that there would have been no statistical deviation between the individuals polled and the results for the whole population. The results would be the same and additional voters would simply be redundant. We see this when elections are called by the major news networks long before all of the polls are counted.

Or maybe the results might not be the same because at 28 per cent, it is still just a poll. And polls of late have been notoriously inaccurate.

To put this in perspective, it is like saying that the polls before the last provincial election were sufficient and should have been used to generate the composition of the provincial government. If that had been the case, the B.C. Liberals would not be in power.

A poll depends on taking a large enough group that the results approximate the real answer. However, there is always some error which is why pollsters say: "accurate to plus or minus 3.5 per cent, 19 times out of 20."

What they are saying is that the pollster thinks that 95 per cent of the time their results are within 3.5 per cent of the real answer.

But aside from the pure mechanics of the voting process, there is also the question of why do people vote or, more accurately, why do some people not vote?

Political theorists have wrestled with this one for a long time. In a paper by John Ferejohn and Morris Fiorina, titled: The Paradox of Not Voting: A Decision Theoretic Analysis, the authors try to set the choice of voting or not voting within a framework. They reference work that states: "a citizen would vote rather than abstain if his [or her] vote value exceeds zero, where the vote value equals the utility difference between the two candidates discounted by the citizen's probability of affecting the outcome."

While that is a mouthful, essentially it comes down to saying that people vote when their voice matters and when the cost of voting is not high. In our democracy every vote matters. If they didn't, we could just leave it all to one person to decide or maybe even the flip of a coin. And the cost is only a few moments of every citizen's time.

Hence, everyone needs get out and vote. Anything less is simply a poll. Everyone voting is the only way to ensure that the results of the election truly represent the will of the people.

Every vote counts.