Here are the oddest questions included in the political polls
Ever wanted to know more about the anonymous people whose opinions shape our forecasts for the General Election on May 7? Well, wonder no more. The answers can all be found in the poll data.
Let's start with the latest poll from Lord Ashcroft, which showed the Conservatives leading Labour by 34% to 30%. But just who are these 1,002 respondents from whom we are getting our electoral insights?
Over the past three years 61% of them have gone abroad on holiday:
And 12% of them live in households with three cars or more:
Or how about we look at the most recent YouGov poll, which polled 1,675 people and showed Labour leading by one point at 35% to the Conservatives' 34%.
The pollster broke its respondents down by the type of newspaper they buy:
Although the questions may seem rather odd, in fact they serve a serious purpose. These questions have been deemed important by the pollsters because they help them weight respondents' answers to ensure that they are more representative.
As YouGov explains (emphasis added):
YouGov draws a sub-sample of the panel that is representative of British adults in terms of age, gender, social class and type of newspaper (upmarket, mid-market, red-top, no newspaper), and invites this sub-sample to complete a survey...[This] Active Sampling ensures that the right people are invited in the right proportions. In combination with our statistical weighting, this ensures that our results are representative of the country as a whole.
That categorically does not mean that each individual poll represents an accurate snapshot of the nation's sentiment, but it does mean that as the polls continue through the political cycle they help to build a more accurate picture of people's voting intentions than the raw data would as it reduces the likelihood of unrepresentative samples skewing the results.
Moreover, by publishing the questions that they ask respondents and how the samples are weighted polling companies are giving us an insight into how they come to their headline results. And that's important, because the process of weighting especially by things as apparently idiosyncratic as the frequency of foreign holidays, the number of cars people own or the type of newspaper they buy is not without controversy.
Political forecasters can then weight the polls based on which methodologies they believe provide a more accurate reflection of voting intentions and provide the public with their analysis of where the parties stand. In an election as close as the one Britain is facing, that insight can help shape our understanding of what to expect after the General Election and, potentially, allow us to vote in a more informed way.
So in the end, knowing how many times people go on holiday may be much more helpful than it first appears.