Attribution of an Amalgam

The Observer misattributed a comparison in The Guardian.

An article in The Observer claimed:

the Office for National Statistics (ONS) reported that the UK had the highest total of deaths from Covid-19 of any European nation.

This is false. The Guardian calculated the UK’s figure, using statistics from three statistical offices. Their article made the comparison: the ONS did not.

Constructing an incomparable figure

On 5th May, the headline in The Guardian was:

Calls for inquiry as UK reports highest Covid-19 death toll in Europe

The article also contains a visualisation:

The UK has the highest death toll in Europe from Covid-19

The star for the UK’s figure highlights it is different to the others. (Image: Guardian)

Different countries are counting their COVID-19 death figures in different ways. These differences constrain international comparisons. It is difficult for aggregator sites to contain all information on these differences.

On 4th May, total UK confirmed deaths with COVID-19 were below the Italian total¹.

The Guardian article sums death certificate counts in the four nations. These deaths certificates mention COVID-19:

The ONS produced the figures for England and Wales. The NRS publishes the Scotland statistics. NISRA does the same for Northern Ireland.

For the latter two statistics, this is the description in the article:

With the addition of the official death figures for Scotland and Northern Ireland, this was calculated to take the UK’s toll to 32,313.

This amalgam is incomparable to confirmed deaths elsewhere.

The counted death certificates mention COVID-19. That mention needs clinical suspicion or a positive test for COVID-19. Few countries publish suspected deaths with COVID-19 as part of their daily counts.

Total confirmed death records are up to 4th May. We should not compare that count to certified deaths occurred up to 24th April. The article recognises this temporal problem:

The true figure is likely to be significantly higher due to missed cases and a lag in reporting.

It would be necessary to calculate the same statistic for other countries on the same basis. Mortality statistics in other countries may be less timely than UK publications.

False attribution

In an article on 9th May, The Observer reports this claim as:

The findings come after a week in which the Office for National Statistics (ONS) reported that the UK had the highest total of deaths from Covid-19 of any European nation.

It should be plain the ONS did not make that report, and made no such comparison.

The Observer made a false attribution to the statistics office. That claim itself rested on a flawed comparison.

Uncertainty

Since 6th May, total confirmed COVID-19 deaths for the UK exceeded Italy’s total. These figures come from the health institutes in each country.

News organisations later made the same claim, based on these daily counts.

There is uncertainty over how many people have died with COVID-19. Testing affects the number of confirmed cases and deaths. There are demographic, social, healthcare, and political differences between and within countries.

Whilst we can make broader international comparisons, we should avoid overprecise claims.

The European Centre for Disease Control and Prevention collates these figures. (Image: Our World in Data)

¹Our World in Data uses stats from the European Centre for Disease Control and Prevention. There is a reporting delay of one day. We look for figures on 5th May in this data-set.

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