June 3. The puzzle is why Matt Hancock thought he could get away with it.
Everyone knew that his claims for the level of coronavirus tests included multiple swab tests for the same individual, posted tests, pregnancy tests, driving tests, eyesight tests (the last two another form of double counting in Cummings land).
Yesterday he received a magisterial rebuke from the chair of the UK Statistical Authority, Sir David Norgrove:
Statistics on testing perhaps serve two main purposes [lovely use of mock diffidence in the ‘perhaps’]. The first is to help us understand the epidemic, alongside the ONS survey, showing us how many people are infected, or not, and their relevant characteristics. The second is to help manage the test programme… The way the data are analysed and presented currently gives them limited value for the first purpose. The aim seems to be to show the largest possible number of tests, even at the expense of understanding. It is also hard to believe the statistics work to support the testing programme itself. The statistics and analysis serve neither purpose well.
Hancock and his fellow ministers seem to have forgotten that in earlier moments of virtue, previous governments have set up a series of bodies to keep them numerically honest – the UK Statistical Authority, the Office for Budget Responsibility, the Office for National Statistics (ONS), amongst others. These are staffed by competent, principled, number-crunchers who appear at times to take a positive pleasure in pointing out the official misuse of data.
It is not that Norgrove himself is new to the game. He has been in office since 2017, and on 17 September 2019, he wrote to the then Foreign Secretary, one B. Johnson, about the Brexit Bus:
I am surprised and disappointed that you have chosen to repeat the figure of £350 million per week, in connection with the amount that might be available for extra public spending when we leave the European Union. This confuses gross and net contributions. It also assumes that payments currently made to the UK by the EU, including for example for the support of agriculture and scientific research, will not be paid by the UK government when we leave. It is a clear misuse of official statistics.
The explanation of these repeat offences is not innumeracy, but rather a varying approach to the function of figures. In the case of the bus, Cummings had correctly calculated that it did not matter if the numbers were challenged. The mere act of discussing the claim, up to and including Norgrove’s letter, anchored in the public mind that there was a substantial cost to EU membership.
Similarly, Hancock, desperately trying to defeat the coronavirus, seems to have calculated that the only way to mobilise action is to set and report huge targets, so as to create a boiling mass of activity amongst those charged with delivering outcomes. As anyone involved in running large organisations knows, there are more sober, disciplined, forms of project management, but Hancock seems entirely to lack the mental or practical resources to use these.
I came across this process when working on my book on solitude. As I reached the present, Theresa May published the world’s first strategy for tackling loneliness. When I examined the figures she was using, I found that her claim that 20% of the population was lonely was contradicted by data in the same document from the ONS, which had calculated a figure of 5% (the same figure as lately reported by the Nuffield / UCL study discussed in the diary entry for May 27). But it was the larger headline figure that featured in the press release accompanying the strategy, and in the subsequent public discussion. Statistical accuracy was subordinated to the need to dramatize a newly foregrounded social condition.
It was not difficult for a toiling researcher into the past to work this out. Historians can count when they need to.
Guess what is the subject Sir David Norgrove’s Oxford degree.
Look it up if you don’t believe me.