Template talk:COVID-19 pandemic data/Italy medical cases

Update tables

Perhaps the disaggregated cases should be added when the total number of cases are also added? — Preceding unsigned comment added by Jaxjaxlexie (talk • contribs) 18:16, 26 February 2020 (UTC)

Not all data are given at the same time. There's no problem in giving the latest total count in the cumulative data, and wait a few minutes for the full detail to appear. --Ritchie92 (talk) 18:45, 26 February 2020 (UTC)

Date format

Which date format should be used in the table? I think the options are either "dd Month yyyy" (as in 31 January 2020), or the European one "DD.MM.YYYY" (as in 31.01.2020) or "YYYY-MM-DD" (as in 2020-01-31), but currently there is some kind of mix between the last two. --Ritchie92 (talk) 19:35, 23 February 2020 (UTC)

DD.MM.YYYY is used in PL, but I wasn't aware of it being common elsewhere. For sorting (not switched this on for this table, at least for the moment), YYYY-MM-DD would be best. Otherwise DD Mon YYYY - with the three letter abbreviation to reduce space requirements - for "ordinary" readbility, would seem best to me. Boud (talk) 17:00, 24 February 2020 (UTC)

Any date format can be made sortable using {{dts}}. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:40, 24 February 2020 (UTC)

The "YYYY-MM-DD" format is best. Let's keep it. --Checco (talk) 16:22, 28 February 2020 (UTC)

Data consistency

There is an ongoing discussion at Talk:2020 coronavirus outbreak in Italy#data consistency in table. --Checco (talk) 16:20, 28 February 2020 (UTC)

Should we also keep track of the number of tests performed in this table

Italy government seems to release the total number of swab tests performed everyday. Right now it is only mentioned in the first few paragraphs of the article and is being overwritten every day. Should we add these numbers to the template so that we can keep track of them? Llull juny (talk) 05:57, 2 March 2020 (UTC)

This could be good to have. I haven't searched extensively for data but I only know of the two latest bulletins: 29 Feb and 1 Mar. Can we gather data for more days? --Ritchie92 (talk) 10:36, 2 March 2020 (UTC)

A posteriori negative cases

Hi all. This has been already mentioned in Talk:2020 coronavirus outbreak in Italy but I think it deserves its own discussion. Recently Facquis proposed a change on the calculations that I do not approve. I think that the two cases of Piedmont that have been detected negative should be striked as it's done now, because we know that exactly those two cases were actually negative. In the edit of Facquis instead one reports the wrong number, that has been corrected – not by newspapers – by the Protezione Civile itself! In the other cases like in Sicily or Liguria, we do not know which ones were wrong, so the only option is to insert a (bad-looking) "–" negative number in the table. However whenever possible I would use the strike, because it's easy to signal and understand as an error that has been corrected (i.e. if one clicks on the given source and finds a different number they don't have to worry about Wikipedia's errors, but will understand that it's been corrected later). However regarding the striked-through totals on the subsequent days (24...26) I am neutral about whether to remove the strikes and just give the real number or not. --Ritchie92 (talk) 11:33, 3 March 2020 (UTC)

I think not to use only the data of the civil protection bulletin released at 18:00 CET since 23 February 2020 is senseless. Also i think it would be much better to show the total data for each region since the source reports these and not the daily increase. I also specify that it is easier to obtain the daily increase by having the totals (b-a), rather than obtaining the totals by having the various daily increments (a + b + c ...)--Facquis (talk) 12:29, 3 March 2020 (UTC)

I think not to use only the data of the civil protection bulletin released at 18:00 CET since 23 February 2020 is senseless well, this already happens for the data on the 26 February, per talk discussion, where it was decided to use the regional data for Lombardy that was not reported in the PC bulletin. If you disagree with that you could contribute to the discussion. On the contrary, in this case the PC bulletin itself corrected the data a posteriori, the correction does not come from a non-PC source; so I don't see any discrepancy if we remove the two Piedmont cases from the count, instead it would be dishonest to just report the number for that day that we know for certain (and the PC confirms) it was wrong. Regarding the other issue, I moderately disagree with showing the cumulative number instead of the daily number per region: I know that it is more complicated to keep track of the daily increments, but that's what's important in the statistics for epidemics. Also, such a change would require a new consensus on this talk page, since until now I thik this version with the increments and the full total only in the last row is quite an established version. --Ritchie92 (talk) 13:09, 3 March 2020 (UTC)

WikiProject COVID-19

I've created WikiProject COVID-19 as a temporary or permanent WikiProject and invite editors to use this space for discussing ways to improve coverage of the ongoing 2019–20 coronavirus pandemic. Please bring your ideas to the project/talk page. Stay safe, --Another Believer _(Talk) 18:14, 15 March 2020 (UTC)

Cases by sex: percentages per age don't add up to 100%

In the tables of cases by sex the percentages per age don't add up to 100% -- ChaTo (talk) 09:12, 25 March 2020 (UTC)

Cases by gender

@Ritchie92: This template is modelled after the South Korean one, and the rationale there to include the tables included there was that those are all the tables reported by the KCDC. There's no consideration or implication of whether the inclusion of more information would be too much -- if KCDC had reported the gender by-age break down it would have appeared there too. I've noticed that this template seems to assume that the presentation in the South Korean article is some sort of a standard, from the arrangement of age to the inclusion of parentheses for the percentages -- it is not. The whole reason why the age is arranged as old to young and parentheses enclose the percentages is that that is the official layout by KCDC, and us editors merely copy-pasted the tables and didn't bother changing anything. Considering how the South Korean article includes much more information than is presented here (e.g. epidemiological info of clusters), I disagree that the information is too much in any sense. We have also not reproduced the entirety of the study; far from it, it's just one of the nine pages. Rethliopuks (talk) 09:31, 25 March 2020 (UTC)

I disagree that this template should reflect the South Korean one. Why should it? Also, we don't need to "reproduce" the study, and I think a table listing the distribution of cases by age or by sex is enough to communicate what is interesting. We don't need to clog the page with two more tables with the details of the age distribution for each sex. I am arguing that this is too much and it's actually not an extremely interesting piece of data: it's interesting to see that more women than men overall get infected and die, and also that more older people die than younger people. But that's it, I don't see the need of confirming the "older die more than younger" also for each sex. For example, if the distribution of cases and deaths by age was totally different for women and men, that would have been interesting to show. But adding two similarly boring tables just for the sake of copying from the source is really too much. --Ritchie92 (talk) 09:39, 25 March 2020 (UTC)

It's good you don't think it should reflect it; I was worried that you might. Well, sorry if my diction wasn't exactly right -- I meant it in the sense of WP:Verifiability. On that note, I am not sure if "interesting" is a valid criterion for Wikipedia's inclusion; there's Wikipedia:Notability#Stand-alone_lists for sure, among others.

I would argue that the by-gender table does exactly what you want to see: it reveals that the mortality rates of men vs. women across age brackets are far from uniform. You'd naïvely expect that the mortality ratio (M:F) of all ages fall around 1.6:1 as is shown with the total figure; it is not. The mortality ratio is only around 1.6:1 for those 70 years and older. Those between 30-69 have ratios of around 2.7:1, which I disagree that you could have expected from the data. I also disagree that you could have expected from the undifferentiated data that the mortality ratio peaks at around 50-59 or that those above 90 have the lowest mortality ratio of all groups. Hence the data is not boring either. Rethliopuks (talk) 09:57, 25 March 2020 (UTC)

Ok but it's still not a major, visible, macroscopic difference, that justifies clogging the article about the pandemic with two additional big tables. If you manage to condense all these "interesting" data in a single table then it's ok for me, but creating two more identical tables for this kind of information is superfluous. Also, the stand-alone lists is not the case here. These tables go directly into 2020 coronavirus pandemic in Italy and nowhere else. So either one creates a separate article for "statistics about the 2020 ..." and then there you are free to add all possible tables. But in a general-purpose article I think it's really superfluous and disruptive to add more and more tables indefinitely. As a matter of fact, I am of the opinion that there are already too many graphs and data in the section 2020 coronavirus pandemic in Italy#Statistics. --Ritchie92 (talk) 10:11, 25 March 2020 (UTC)

I respect that. It may be a good idea to plot a graph of some sort, although I'm not sure how best to implement that. Rethliopuks (talk) 10:13, 25 March 2020 (UTC)

I added charts and stacked charts to have a better visualization of the data in the source. --Ritchie92 (talk) 18:03, 25 March 2020 (UTC)

previus table (by region/day)

Coronagr (talk) 11:38, 29 March 2020 (UTC)Hi from Greece Please, could I have the previous version of the table? It is very useful to have all the numbers by day for every region, because I use the data for prediction models. Thank you in advance and I wish you the best!Coronagr (talk) 11:38, 29 March 2020 (UTC)

readability

@Ritchie92: The added region row is for readability -- most people's screen cannot accommodate the entire table, so without the row it will become confusing which cells are which to the human reader, especially when each row of this table is two lines tall on screens horizontally <=1790 pixels or around 1920 pixels. Rethliopuks (talk) 18:32, 19 April 2020 (UTC)

We should adopt a criterion to insert them. Every two months should probably be enough. --Ritchie92 (talk) 15:08, 20 April 2020 (UTC)

updates?

Are we only updating once a week now? — Preceding unsigned comment added by 67.243.220.61 (talk) 20:20, 4 March 2021 (UTC)

Apparently we aren't updating at all anymore? — Preceding unsigned comment added by 172.254.103.202 (talk) 17:22, 26 March 2021 (UTC)