Over the last few months, I’ve often heard that polls can’t be trusted. In particular, I have heard that they can’t be trusted because each one usually involves study of only about a thousand individuals. I have even heard that argument from a retired quantitative linguist.1 So I’ve put together this essay in order to explain how polls work, why a random sample of a thousand should usually be considered sufficient, and why the results should be treated as informative even though they do not enable us to predict precise numbers of votes (which is a particular problem when the results are going to be close — because then, precise numbers can make all the difference).
Last month, I came out to my friends as a non-supporter of Jeremy Corbyn, the leader of the Labour Party. Since that time, I’ve been public about it in my own small way. It took me a while to admit that my party was led by someone who did not have its voters’ interests at heart, but the key moment came when I was delivering leaflets for the Labour Party’s In for Britain campaign and I ended up talking to a man who had voted both for the Labour Party under Tony Blair and for the Conservative Party under David Cameron. A man – in case this isn’t obvious – from outside my middle class bubble in which people will express solidarity with Jeremy Corbyn even while being unsure that they really agree with him, and in which there are probably more anarchists than conservatives. I picked up that he wanted a Labour Party that he could vote for again. For Jeremy Corbyn, he had two devastating words: ‘Student politics.’ Those words stuck with me.
It wasn’t just that he was a swing voter in a marginal constituency, and therefore exactly the kind of person that Labour needs to be reaching out to if it ever wants to be in government again (if it ever wants to be in government again… a point that I’ll return to).
It was that he was right – and I knew it.