Brexitometer Boards and Selection Bias

‘Brexitometer’ boards from various campaign groups are often shared on social media.

This article examines the multiple deficiencies in treating these boards as measures of public opinion.

Rhagfarn hunan-ddethol

I don’t think they want Brexit.

In which people attend a campaign stall, there are two types of sampling bias at work.

Firstly, there is coverage error: many adults in Wales would have been unaware of or unable to go to that particular stall on that particular day. The failure to adequately cover the whole population (Welsh adults) can lead to error in opinion estimates.

Secondly, there is self-selection bias. The people running these stalls do not randomly select passersby to give their view. People choose whether to participate. Those who volunteer their views tend to over-represent strong opinions.

Social reinforcement can also play a part in that self-selection bias, as people with countering opinions may see no purpose in participating.

Arolwg barn

Opinion is somewhat more balanced than the Liberal Democrat polka-dot poster would impress.

The YouGov poll of 1,071 Welsh adults conducted on 23–26th July estimated that:

  • Whilst opposition has narrowed, 46% opposed a second EU membership referendum, with 43% in favour.
  • Excluding those who would not vote or were undecided, 53% would vote Remain in a hypothetical referendum.

Analysts and scientists should be forthright in the fatal limitations of using campaign boards as measures of public opinion.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store