 |
 |
I was not talking about or defending the fact that it is difficult to find information about the data sources and what they do and don't contain. By now that should be properly documented. It is possible to find out fuller information by digging around, but even then some knowledge of how routine data are compiled is needed to know where to look. I have been meaning to put my questions to Public Health England, but haven't yet made time to do this as I have other things to do. ONS has now published some comparisons of data sources and explanations of why they differ, but questions still remain. What I was commenting on was behind the back criticisms of a graph which appears to have been hastily prepared on a Bank Holiday as being 'boilerplate Excel'. What we see is on the one hand, an overstretched government and NHS statisticians responding to an occasion for which they have not been properly resourced and on the other hand, other statisticians who are whiling away the boredom of the lockdown by firing off criticisms of it to the UKSA and jiscmail lists. Alison Macfarlane ________________________________ From: simon briscoe <[log in to unmask]> on behalf of Simon Briscoe <[log in to unmask]> Sent: 14 April 2020 21:16 To: Macfarlane, Alison <[log in to unmask]> Cc: [log in to unmask] <[log in to unmask]> Subject: Re: Reasons for asymptote CAUTION: This email originated from outside of the organisation. Do not click links or open attachments unless you recognise the sender and believe the content to be safe. Alison, I do think that if decisions are being made on data shown in those charts, which is the impression we are given, then the GSS/ONS Statistics code of practice applies and the sources and data should be published. I have written to UKSA - see this blog for my thinking. https://simonbriscoeblog.wordpress.com/2020/04/01/new-and-open-data-can-help-tackle-covid-19/ I am not slagging the people concerned off behind their backs I’d ask them to their faces if I knew who they were. And it would be nice if UKSA replied. Simon On 14 Apr 2020, at 21:07, Macfarlane, Alison <[log in to unmask]<mailto:[log in to unmask]>> wrote: Allan is not the only person to notice that the statistical presentation in some of the coronavirus press briefings are not up to the standards we should like to expect. This is hardly surprising, given the pressure that the people producing them are likely to be working under, with high profile daily press conferences, every day, including bank holidays. Rather than spend time slagging the people concerned off behind their backs on allstat, perhaps an email to the organisation concerned, commenting on ways of improving the presentations might be more constructive. As a long standing subscriber to Radical Statistics, which has a long tradition of aiming for constructive criticism of government statistics, I am concerned that we are soliciting destructive criticism of this sort. Of course, if a group of government statisticians found itself at a loose end and decided to write an article giving examples of poor presentation of statistics by academics, it would have a field day! As allstat is not a discussion list, I will say no more, except that I think there are better ways of improving the standard of presentation of statistics across the board. Alison Macfarlane ________________________________ From: A UK-based worldwide e-mail broadcast system mailing list <[log in to unmask]<mailto:[log in to unmask]>> on behalf of John Bibby <[log in to unmask]<mailto:[log in to unmask]>> Sent: 13 April 2020 17:58 To: [log in to unmask]<mailto:[log in to unmask]> <[log in to unmask]<mailto:[log in to unmask]>> Subject: Re: Reasons for asymptote CAUTION: This email originated from outside of the organisation. Do not click links or open attachments unless you recognise the sender and believe the content to be safe. Thanks Allan. Would you or anyone care to do a short article on stats in government presentations? This is for a special coronavirus issue of Radical Statistics. Other topics also welcome. Max 2000 words by 1st May please. Please forward to anyone who might be interested. John BIBBY On Mon, 13 Apr 2020, 18:26 Allan Reese, <[log in to unmask]<mailto:[log in to unmask]>> wrote: Today's Downing Street briefing on Covid19 included a line graph of numbers of patients in hospital with Covid19. The top line for London went over "4K" but appears to be levelling out in the last week. What would be helpful would be a reference line for the number of beds available in those hospitals - not obvious that the line demonstrates less demand rather than saturation. Also to note that the COBR (Cabinet Office Briefing Room) appear to pay little attention to informative labelling. The Y axis is labelled "0K, 2K, 4K" which is insultingly nerdy - do they mean 4000 or 4096? I doubt such "illiteracy" would be tolerated in the Civil Service for the text in a news release. It's boilerplate Excel, exported without revision. Allan -- R Allan Reese, Dorchester Dorset You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask]<mailto:[log in to unmask]>, leaving the subject line blank. You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask]<mailto:[log in to unmask]>, leaving the subject line blank. You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask]<mailto:[log in to unmask]>, leaving the subject line blank. You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.