Poll Fatigue

Once upon a time opinion polls were conducted by people with clipboards stopping strangers on the street. Each clipboard-wielder had a quota of ‘types’ of person to question, and back at the poll HQ the ‘sample’ would be carefully balanced and weighted to reflect the composition of the electorate.

In theory they still do that today. In practice it is more complicated. There are more polling companies conducting more polls than ever. The number crunching is faster – instantaneous, even, for some polls, and samples are still carefully weighted. And yet….

There’s a divergence between internet polls and telephone polls (by which I mean cold calls made by phone, rather than polling done via a smartphone app). Phone polls give a markedly higher rating to the Conservative Party. Given that when questioned, both pundits and the public think that the Tories will be the largest party after the election, despite polls of opinion suggesting a dead heat, should we conclude that internet polls aren’t worth heeding?

I’m on the YouGov panel. Like all those on the panel, I’ve chosen to be involved. It means that I’m already politically engaged, a heavy internet user, and a smartphone owner. Like an awful lot of people I know, I’m also someone who no longer answers the landline. My guess is that I’m especially useful to pollsters, as I’m a middle-aged, ethnic minority woman, and I’m assuming that fewer of us/them are volunteers for such panels.  But it also means, I’m not ‘typical’.

So, as a mere scrap of poll fodder, what do I make of polls?

Firstly, I think they are likely to be broadly accurate as a barometer of opinion.  But they are only as good as the data they put in, and so the methods they use will skew results.  Phoning people at home will increase participation by older people who still answer landlines. They’ll also under-represent the poor, who, increasingly, do not have landlines at all, but do have pay-as-you-go mobiles.  But older people turn out to vote in high numbers, and poor people don’t. Could the two balance out? Maybe.

But are they right about the likely outcome of the election?  There I hesitate.  This still feels like the weirdest, least predictable election of my life.  I wouldn’t be at all surprised if this didn’t turn out to be the election when the people called the pollsters’ bluff.  But how?  Now there’s the question….

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s