Wax and Wane

Trust in the government did not ‘plummet’.

An article in The Observer claimed a poll showed “public trust plummets”. The Telegraph published an article about the lock-down. This article asserted the public was “increasingly supportive of a staggered exit”.

Both claims derive from the same poll by Opinium. Neither claim is accurate.

The Observer

Opinium conducted an internet panel poll between 21–23rd April 2020. The company sampled 2,005 UK adults, on behalf of The Observer.

Toby Helm, political editor of The Observer, wrote an article with the headline:

Public trust plummets in Britain’s handling of pandemic, new poll reveals

The headline refers to Opinium’s regular question:

To what extent do you have confidence in the Government’s ability to handle the Coronavirus situation as it continues to develop?

Respondents can give one of five answers, including ‘neutral’. If people clicked ‘completely confident’ or ‘somewhat confident’, those answers were ‘confident’. The ‘not confident’ answers were ‘not very confident’ and ‘not at all confident’.

Opinium’s series does not show a plummet in public confidence:

These graphs are in the online article. (Image: The Observer)

Estimated public confidence waxes and wanes. The article’s link address suggests ‘wanes’ was first used in the article, rather than ‘plummets’.

This trend is like public approval in the government’s handling. In the article, Adam Drummond (Opinium) describes the trend in approval:

Overall approval has fluctuated and ticked downwards but is still stronger than before the lockdown came into effect.

The first sentence also asserts:

The public’s confidence in the government’s ability to handle the coronavirus crisis has fallen sharply in the past fortnight, with less than half of voters now having faith in decisions made by ministers, according to the latest Opinium poll for the Observer.

Based on this poll, we cannot say “less than half of voters” have confidence.

The headline statistics are for all UK adults in the sample, not just voters.

Surveys provide estimates, subject to many sources of potential error. For a poll of 2,000 adults, sampling error is large enough that an estimate of 49% may not be a minority.

In the same way, 51% may not be a majority. (Image: Pew Research Center)

The Telegraph

Gordon Rayner, political editor of The Telegraph, wrote:

An Opinium opinion poll, published on Sunday, showed that the public is increasingly supportive of a staggered exit from the lockdown, with more than half of people wanting restaurants, offices, shopping centres and schools to reopen as soon as new infections decrease, although a majority want sports stadiums to remain closed until there is a vaccine.

This sentence refers to the following question in Opinium’s survey:

Assuming that a vaccine isn’t available until next year, please tell us what conditions you think need to have been met for each of the following things to re-open?

The response options were:

  • Should re-open as normal from May 8th regardless of the situation;
  • Should re-open once the number of new infections starts to go down;
  • Should re-open once the number of new infections starts to go down, with limitations on the number of people allowed inside, but closes if the number of infections starts to rise again;
  • Should not re-open until a vaccine is available;
  • Don’t know.

This is the first time that Opinium asked this question. Without a point of comparison, we cannot say support is ‘increasing’.

Opinium estimated the third option (“with limitations”) had majority support for some facilities.

Opening ‘with monitoring and restrictions’ generally had the most support. (Image: Opinium)

Rayner’s phrasing may impress the second option was the most popular. This would be a false reading of Opinium’s poll.

Readers should know how survey researchers word questions. They should also know what the response options were.

Surveys provide estimates, with uncertainty. Journalists should take care when reporting survey research, and convey that uncertainty.

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