Really unfortunate

Written by

in

ng-left: 40px”>It said there needed to be more transparency about data and how predictions were being made.

The projections were out of date and over-estimated deaths, it has emerged.

A forecast made by Public Health England and Cambridge University said the country could soon be seeing more than 4,000 deaths a day.

The projection was made weeks ago and had forecast there would be 1,000 deaths a day by the end of October when the average was actually four times less than that – a fact that was known at the time of Saturday's TV briefing.

What is more, the model had already been updated to predict a lower estimate, but this was not used in the briefing fronted by chief scientific adviser Sir Patrick Vallance and chief medical officer Prof Chris Whitty, alongside Prime Minister Boris Johnson….

Sir David Norgrove , chair of the UK Statistics Authority, said: "I recognise the pressures faced by all those working on decisions related to coronavirus.

"But full transparency of data used to inform decisions is vital to public understanding and public confidence."

The use of the data has also drawn criticism from former prime minister Theresa May, who abstained from the lockdown vote in parliament on Wednesday. The vote saw MPs agree to the four-week restrictions in England.

Sir David Spiegelhalter, one of the most respected statisticians in the country, said the whole saga had been "really unfortunate".

He said the situation with Covid was sufficient to warrant "radical action", but he did not believe the data made available was enough to fully support a lockdown, saying other wider factors needed to be taken into account as well.

Really unfortunate? Well, that's one way of putting it – if you're one of the most respected statisticians in the country and unlikely to be looking for work anytime soon. Others might phrase it a little more strongly.

Comments

  1. Richard Powell Avatar
    Richard Powell

    Agreed. Tangentially, I loathe the formulation “x times less” which I take to be an indication that whoever cobbled the piece together is at best semi-numerate. What does it mean? What would “one time less” be? No deaths at all? Half the number of deaths? “Two times less” – 500 deaths, 333 deaths, minus 1,000 deaths, minus 2,000 deaths? An argument could be made for each of these. The construction has not legitimate place in statistics or news reporting.

  2. Dom Avatar
    Dom

    I took “4 times less” to mean one-fourth. I never heard it said like that before. Yes, it’s a very strange construction.

Leave a Reply

Your email address will not be published. Required fields are marked *