There is a lot of criticism made of journalists but this by Ed Conway of Sky News is a good explanation of how to understand Covid 19 data and is worth watching, and worrying about: Some government ministers should watch it.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
Thanks for this Richard, really good review. I’ve been waiting for some good log scale graphs! What is interesting in this video is he runs through comparisons with other countries that are doing not too well, but ignored the included South Korea data – why? Why not mention that and ask what they are doing differently that there is such a huge difference? As far as I’m aware that difference is extensive CONTACT TRACING – why are governments not paying attention? Why aren’t they looking at positive results and acting accordingly? Why blank this and just yabber on how we are trying to do better than Spain? Why are they focusing on lock down and trying to pretend that’s the only way to resolve this?
All questions are rhetorical.
Log scales exist to get coronavirus on a comprehensible graph
They didn’t teach you that at school
They will now…
I guess you don’t get too many exponentials in accountancy, I’m not even sure when I first came across them, but I’m pretty sure we had to use log tables in school – I didn’t fully understand it at the time but later mathematics made me realise how fantastically good they were for big-number calculations pre-calculator.
Here is a web page of graphs:
https://www.travellingtabby.com/scotland-coronavirus-tracker/
The two bottom ones are the best indicators – the number of days for a doubling of deaths – 3 days is your red line, below that and you have a runaway pandemic, above it you are in control. Scotland is only just above the red line recently (about 4 days for deaths to double now). The graph below that is the same as in the clip above – death rate. It has dashed lines indicating ‘days to double’: England is hovering around the 3-day mark (runaway pandemic territory), Scotland just below,,, and South Korea wayyyy below (so more than a week for deaths to double. Doubling is exponential growth, and shows a straight line on your log graph) (sorry, your log-LINEAR graph). But, of course, it all depends on how accurate your data is.
I did log graphs at school
And one-eighth of my degree is in stats…
I do get them
I am still slightly reeling from your flat dismissal of GCSE maths as “utterly and completely useless” a couple of weeks ago, and now you want everyone to learn about log-linear graphs (aka semi-log graphs) at school? They might need to learn about logarithms first. Playing with a slide rule could help. Logarithms and graphs with logarithmic scales (log-log and log-linear) are on the Edexcel maths A-level syllabus, but not on the GCSE syllabus as far as I can see.
Perhaps you could give us an article on a redesigned GCSE maths syllabus, explaining what you mean “practical maths”, beyond budgeting, percentages, tax calculations, and (it seems) logarithms? But not “useless” subjects such as trigonometry and calculus?
God, 563 deaths yesterday, and 381 yesterday. After a brief hopeful blip on Sunday and Monday (reduced hospital admissions over the weekend?) we are back to 33% exponential growth with a vengeance.
And 33% growth is 10,000 a day deaths in a couple of weeks…
And, incidentally, you don’t need to know about logs to understand a log scale graph. You need to be able to follow logical explanation, which is not the same
Sort of ok as a primer to graphs. A brief mention of the measuring methodolgy – needs more on that and adjust accordingly. The final graph is a zinger. Showing that we are currently UNDER average – if that is including ALL deaths recently it seems to be setting up a narrative.
Ideally all countries ought to be doing the same tests, the best most accurate ones. But we are not being told if we are. Infact it seems bozo lied yet again at last weeks pmq’s saying we were doing 10,000 and would be doing 20,000, 25,000 within days!
We still aren’t meeting the first even older claim. Because we are not able to – i just a radio news item of a hospital doctor who wanted to 100 tests but could only manage THREE, because they haven’t got the necessary swabs never mind other items for the tests that are being lied about.
Sorry to elongate this post but this has just been reported in China – now that’s how to do science and medicine in a hurry with due care.
‘During a press conference held on Wednesday, Deng Ning, deputy dean of Tsinghua University’s Office of Scientific Research & Development said that a research team from the university, led by Zhang Linqi from Tsinghua’s medical school, has successfully isolated more than 200 strains of COVID-19 antibodies with high neutralization ability and their coding genes from recovered patients’ lymphocyte. ‘
Full article worth reading here
https://www.globaltimes.cn/content/1184451.shtml
Thanks Richard that was an interesting report. For those interested in lots of comparative data from different countries this website from an IT specialist at UCL should go some ways to satisfying your needs.
http://nrg.cs.ucl.ac.uk/mjh/covid19/#e