Jump to content

Template talk:COVID-19 pandemic data/India medical cases chart

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

WikiProject COVID-19

[edit]

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) 18:06, 15 March 2020 (UTC)[reply]

Inconsistencies between case numbers in medical cases chart and actual numbers on the MoHFW website

[edit]
 – relates to the data here

Some of the case numbers in Template:2019–20 coronavirus pandemic data/India medical cases chart do not seem to match what appeared in the MoHFW website.

Here are the case numbers used in the template right now:

2020-03-10;0;4;50;;;50;+6;;+14%
2020-03-11;0;4;60;;;60;+10;;+20%
2020-03-12;1;4;73;;;73;+13;;+22%
2020-03-13;2;10;81;;;81;+8;;+11%
2020-03-14;2;10;97;;;97;+16;;+20%
2020-03-15;2;13;107;;;107;+10;;+10%
2020-03-16;2;13;118;;;118;+11;;+10%
2020-03-17;3;14;137;;;137;+19;;+16%
2020-03-18;3;14;151;;;151;+14;;+10%
2020-03-19;3;15;173;;;173;+22;;+15%
2020-03-20;4;23;223;;;223;+50;;+29%
2020-03-21;5;23;283;;;283;+60;;+27%
2020-03-22;7;24;360;;;360;+77;;+27%
2020-03-23;9;24;434;;;434;+74;;+21%
2020-03-24;9;40;519;;;519;+85;;+20%
2020-03-25;10;43;606;;;606;+87;;+17%
2020-03-26;16;45;694;;;694;+88;;+15%
2020-03-27;18;74;834;;;834;+140;;+20%
2020-03-28;20;80;918;;;918;+84;;+10%
2020-03-29;27;95;1024;;;1,024;+106;;+12%

Here is how these numbers would look if we were to use the updates that appeared on the MoHFW website:

2020-03-10;0;3;50;;;50;+6;;+14% <!-- fixed recoveries to 3 -->
2020-03-11;0;3;60;;;60;+10;;+20% <!-- fixed recoveries to 3 -->
2020-03-12;1;3;74;;;74;+14;;+23% <!-- fixed recovers to 3 and total to 74 -->
2020-03-13;2;10;81;;;81;+7;;+9% <!-- updated increment and growth percent due to fix in previous entry -->
2020-03-14;2;10;84;;;84;+3;;+4% <!-- fixed total to 84 -->
2020-03-15;2;13;110;;;110;+26;;+31% <!-- fixed total to 110 -->
2020-03-16;2;13;114;;;114;+4;;+4% <!-- fixed total to 114 -->
2020-03-17;3;14;137;;;137;+23;;+20% <!-- updated increment and growth percent due to fix in previous entry -->
2020-03-18;3;14;151;;;151;+14;;+10%
2020-03-19;4;20;173;;;173;+22;;+15% <!-- fixed deaths to 4 and recoveries to 20 -->
2020-03-20;4;23;223;;;223;+50;;+29%
2020-03-21;4;23;315;;;315;+92;;+41% <!-- fixed deaths to 4 and total to 315 -->
2020-03-22;7;24;360;;;360;+45;;+14% <!-- updated increment and growth percent due to fix in previous entry -->
2020-03-23;9;35;468;;;468;+108;;+30% <!-- fixed recovers to 35 and total to 468 -->
2020-03-24;10;40;519;;;519;+51;;+11% <!-- fixed deaths to 10 -->
2020-03-25;10;43;606;;;606;+87;;+17%
2020-03-26;16;45;694;;;694;+88;;+15%
2020-03-27;19;67;834;;;834;+140;;+20% <!-- fixed deaths to 19 and recoveries to 67 -->
2020-03-28;19;80;918;;;918;+84;;+10% <!-- fixed deaths to 19 and recoveries to 80 -->
2020-03-29;27;96;1024;;;1,024;+106;;+12% <!-- fixed recoveries to 96 -->

Here are the relevant reference links to support each entry:

By the way, I am maintaining an archive of all the official case numbers from MoHFW at https://indiacovid19.github.io/. Every entry on this page is supported by an archived snapshot of the actual source of the data. The data I have provided above is consistent with what is in this archive as well as what is in the archived snapshots of MoHFW. Please let me know if we can update the case numbers in this chart to make them consistent with MoHFW data.

-- Susam Pal (talk) 09:48, 30 March 2020 (UTC)[reply]

  • Comment - Hi, @Susam Pal: The numbers in the chart you see are used from first table in Stats section. The table you see in Statistics section was created by @Shanze1:, @Mr.Mani Raj Paul: and me around 17th March and to find the per day cases we used PIB press releases and various other news sources which you will see in references column in the table and between 11th to 17th march we had to use some other news sources due to unavailability of official data. And the thought that web archives of mohfw could be used didn't came through our mind. I am ok with updating it with mohfw data on whichever day we have archieves, but on some days web archives are not availble if i am right, then what should be done for these days?, others also give there opinions. @Brown Chocolate: @TheRichardPinto: @Kautilya3: @SaiP: - Mayankj429 (talk) 12:10, 30 March 2020 (UTC)[reply]
Thanks for the detailed comment. I went through all the PIB press releases myself after reading your comment and it helped me to make my own archive of the case numbers accurate too. I see that we have got a web archive for MoHFW for every date since 10 Mar 2020. For prior dates, we have got PIB archive for every date that has had any change in case numbers. -- Susam Pal (talk) 07:01, 6 April 2020 (UTC)[reply]
  • Comment - Susam Pal Saw your archive at https://indiacovid19.github.io/. It is using MoHFW data from only 11 March. It uses Worldometer for all entries prior to that. We kind of banished that source. Also, prior to 11 March, archive doesn't have captures (of MoHFW). We worked around this by using pib.gov.in press updates and a India Today source which cites MoHFW. So I'm are positive that the data here already is consistent with MoHFW data. So I don't agree on changing anything. Shanze1 (talk) 12:24, 30 March 2020 (UTC)[reply]
Susam Pal is not trying to update it with COVID data, but rather archived MoHFW data. The differences seem to be minor, possibly to do with the time of the day. I support taking the evening numbers uniformly. -- Kautilya3 (talk) 12:36, 30 March 2020 (UTC)[reply]
I am not suggesting any changes to the data in this template for dates prior to 11 Mar 2020. My suggestion is to update the data for 11 Mar 2020 and later for which we do have archived MoFHW data. -- Susam Pal (talk) 12:57, 30 March 2020 (UTC)[reply]
@Susam Pal: Sorry, my bad. Yes I think using the archived figures like you suggest can be done. Shanze1 (talk) 14:33, 30 March 2020 (UTC)[reply]
@Shanze1: Thanks for offering your feedback. I did go ahead and edit the template to use the archived figures three days ago. -- Susam Pal (talk) 13:41, 9 April 2020 (UTC)[reply]
Hi @Kautilya3: Do you mean 4 (on 2020-03-10), 4 (on 2020-03-11), 7 (on 2020-03-12), 7 (on 2020-03-13), and 10 (on 2020-03-14) cumulative recoveries? Can you point me to your source for these numbers? I checked the PIB press releases as well as the MoHFW archives and both show the following cumulative recovery numbers:
  • 2020-03-10: 3 recoveries: PIB (search for "discharged")
  • 2020-03-11: 3 recoveries: PIB
  • 2020-03-12: 3 recoveries: PIB
  • 2020-03-13: 3 recoveries: PIB MoHFW
  • 2020-03-14: 10 recoveries: PIB MoHFW

In particular, the data for 2020-03-13 indicates 3 recoveries both on the PIB website and the MoHFW website.

-- Susam Pal (talk) 05:17, 31 March 2020 (UTC)[reply]
Hi Susam Pal, I was only talking about the internal consistency of the numbers. (Going down from 4 to 0 doesn't make sense.) Your new verified numbers look fine, in particular changing the initial recoveries to "3". Cheers, Kautilya3 (talk) 07:20, 31 March 2020 (UTC)[reply]
Hi Kautilya3, That was indeed an error. I've fixed the table now. By the way, there was another incorrect information in my previous comment. The number of recoveries on 2020-03-13 appeared as 3 in my previous comment. It is indeed 10 as found in the referenced PIB press release. Quoting from the release: "Of these, 3 from Kerala were treated and discharged. In addition, 7 more from Safdurjung have also recovered." The table in my earlier comment has been updated as per the updated template format. The table I have updated above is now fixed and generated using this script. References for all case numbers used in the table are archived here as well as here. -- Susam Pal (talk) 11:02, 6 April 2020 (UTC)[reply]

It is mentioned as Active cases in the top, it should be Confirmed cases Goswami666 (talk) 13:42, 26 April 2020 (UTC)[reply]

@Goswami666: If you are talking about the legend for the orange bar that says "Active cases", then it should indeed be "Active cases". The orange bars indeed represent the number of active cases. The sum of the numbers repesented by the red (deaths), blue (recoveries), and orange (active cases) represents the total number of confirmed cases. To verify, let us take the entry for 2020-04-24 for example. The markup code for it contains 2020-04-24;723;4814;23452 which indicates 723 deaths, 4814 recoveries, and 23452 confirmed cases. The active cases for the orange bar is then automatically calculated to be 23452 - 4814 - 723 = 17915. Now if we hover on the orange bar for 2020-04-24 for a while, a tooltip pops up that says "17915", thus confirming that the orange bar represents the number of active cases -- Susam Pal (talk) 18:28, 29 April 2020 (UTC)[reply]

Percentage increases

[edit]

I propose that we remove this column. It is redundant to the graph and the actual figures, which are showing the same data, after all; furthermore, it is making the table so wide that it makes the text next to it hard to read on most screens (ie the average laptop screen). Vanamonde (Talk) 15:52, 30 March 2020 (UTC)[reply]

If we are going to fiddle with the format, I would rather switch the US chart [1], which seems compact as well as containing the most useful information. The numbers of deaths are useful because they are a direct indicator of the number of infections. -- Kautilya3 (talk) 21:22, 30 March 2020 (UTC)[reply]
No objections to switching out percentage increase in number of cases for number of deaths, if we can find a good source for that. Vanamonde (Talk) 16:07, 31 March 2020 (UTC)[reply]
Agree, US charts seems more useful - with death count also shown. Harsh182 (talk) 14:22, 2 April 2020 (UTC)[reply]

Miraculous week leading up to 12 June

[edit]

@Susam Pal and Mayankj429: In case anyone is interested, India turns out to have had a "miraculous" week of very low Poisson noise in the official national statistics just before reaching 10000 confirmed SARS-CoV-2 infections per day on 12 June 2020. This is presently one of the results in this preprint arXiv:2007.11779 Zenodo3990666 using the WP C19 Case Count Task Force data - for India that means data from this template - as the primary curated data collection. The JHU CSSE data gives almost identical results, except for India, for which it seems to have added up state-level India data instead of using official national data. The low-noise miracle is absent in the JHU data. The daily cases graph at https://indiacovid19.github.io appears to have the "low-noise miracle" present. Here is a pop-science explanation of why low noise is statistically unlikely to be natural. Boud (talk) 01:10, 1 September 2020 (UTC)[reply]

Semi-protected edit request on 6 November 2020

[edit]

Hello, how about adding the buttons to hide specific months in this template, as done for example here? https://en.wikipedia.org/wiki/Template:COVID-19_pandemic_data/United_States_medical_cases_chart 2A00:23C6:54A4:E201:F81B:87E6:F769:FC3D (talk) 13:36, 6 November 2020 (UTC)[reply]

 Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. I am not sure how this would be done. P,TO 19104 (talk) (contribs) 19:13, 21 November 2020 (UTC)[reply]
The United States medical cases chart template, and the templates for many other countries (e.g. Brazil_medical_cases_chart, France_medical_cases_chart) have a convenient and user-friendly feature to hide the values for cases more than 15 days before the current date. They also include buttons to show the case by month. This way the information arguably most important to the visitor, the current values for cases and those of the near past are immediately visible, without the need to scroll to the bottom of the page. This feature must be generally useful, since more and more country templates adopted it, and it should be easy to implement, by simply copying the existing code in one of these page. (I would do it myself but the page is locked.) Hopefully someone can help with this! — Preceding unsigned comment added by 2A00:23C6:54A4:E201:35D0:CBB0:C8E9:96C (talk) 21:16, 22 November 2020 (UTC)[reply]
Tagging Susam Pal, Please have a look at it. Mayankj429 (talk) 06:14, 24 November 2020 (UTC)[reply]
Thank you for tagging me Mayankj429. The proposed change sounds like a good idea to me. Hopefully someone will make the required edits. :-) I will not do it myself because I am quite busy at this time but if someone does the work, I will welcome it. -- Susam Pal (talk) 19:21, 24 November 2020 (UTC)[reply]
Susam Pal, Ok then I will look into it and do the changes soon. - Mayankj429 (talk) 05:34, 25 November 2020 (UTC)[reply]
 Done This feature was available before don't know who removed it. Anyway now Monthwise Toggle button is there. Mayankj429 (talk) 05:52, 25 November 2020 (UTC)[reply]
Thank you, that looks good! A small improvement that could be done is setting the "duration" value to 15, to mirror the templates of the other countries. It is now set to 21 ("duration=21"). — Preceding unsigned comment added by 2A00:23C6:54A4:E201:1DC5:CCC4:1B02:5FCD (talk) 23:32, 28 November 2020 (UTC)[reply]
I'm setting the |answered= or |ans= parameter to no for this edit request in order for more editors to see it. As mentioned in the previous comment, setting the "duration" value to 15 will bring this template in the same formatting as the templates of the other countries, very useful to quickly scan the latest values.
 Done, changed default to 15 days as requested. ‑‑ElHef (Meep?) 21:08, 5 December 2020 (UTC)[reply]
This is perfect, thanks a lot! — Preceding unsigned comment added by 2A00:23C6:54A4:E201:9854:1F26:5131:601D (talk) 12:09, 6 December 2020 (UTC)[reply]

Percentage or absolute option for new infections?

[edit]

I started a discussion on that topic here Project COVID-19, Medical cases charts - change type. My personal opinion is that absolute change display suits much better in an epidemic's second wave. But also the status quo ante (percentage) should be kept here until some agreement is reached (locally or possibly a consensus in linked discussion). -- Kohraa Mondel (talk) 15:19, 3 December 2020 (UTC)[reply]

@Kohraa Mondel: you are right, I came here to leave the same comment now months later. The two most relevant data points are the % change in active infections day over day -- which we seem to have the data to calculate -- and the absolute change in confirmed infections. I am not sure how to implement this. – SJ + 00:55, 27 April 2021 (UTC)[reply]
@Sj: Absolute changes is quite easy, you just set the changetype argument to a instead of p (or individually for the two columns by changetype1 and changetype1). In Module:Medical_cases_chart/sandbox2 I implemented changetype = c for active cases among others like 7-day incidence and rolling 7-day average prototypically, but Alexis and I wanted to tidy up the parameter processing before transfering this to the main module, and it got forgotten inbetween. So changetype = a could be used right away, about the rest I can make no promises. -- Kohraa Mondel (talk) 09:26, 27 April 2021 (UTC)[reply]
U can compare this table e.g. with the one I'm filling in in the etwi, besides that, it depends on specific units (countries), and specific situation, and on many more factors,... one has simply to try (experiment), besides, there have been rumours, for some decades, that Wikipedia is meant for auditorium, not for the ones who are filling this space. — Pietadè (talk) 09:40, 27 April 2021 (UTC)[reply]