Talk:List of countries by intentional homicide rate/Archive 4

Archive 1

Archive 2

Archive 3

Archive 4

Silly table, again: "Row"

I really (really!) hate to explain here the same points as I mentioned above because, firstly, I already explained them above. But maybe if I use baby-steps, somebody will listen and be able to understand my points this time.

I once again removed the single-column "Row" "table" under Intentional homicide victims per 100,000 inhabitants... with this edit, on the basis that, as ever, a pointless "table" is still pointless. There is no semantic justification for a "table" consisting of a single column (or a single row). Never. Not ever. There is no tabular correlation possible with a single column. The numbers from 1 to 230 have absolutely no connection to any other data anywhere on Wikipedia. Such a table violates the spirit of WP:ACCESS, a guidleine meant to implement the WMF's policy of anti-discrimination. Users with screen readers will surely by stymied by the weirdness of this mark-up.

As I (also) said before, the single-column "table" provides no more usefulness to sighted visitors than it does for the unsighted, as there's no reliable alignment with or connection to the table of actual data. What is the meaning of the 7th cell, the one with the content "7"? What is that supposed to tell me?

My edit was reverted by enivid, with an admonishment to see the reply to you by User:Timeshifter on the Talk page. The row numbers table allows to easily rank countries by Count/Rate. Unfortunately, the row numbers "table" does not allow to easily rank anything at all, either for sighted nor non-sighted readers. And Timeshifter's reply, if you mean, Row numbers are helpful for people who want to see rank order by homicide count, and also rank order by homicide rate. It can't currently be put in the table, and be able to do both rank orders, is not responsive to the issues I've tried to present. I don't deny that "Row numbers are helpful", but I absolutely deny that a completely standalone "table" of 1 to 230 is semantically correct or in any way useful. — JohnFromPinckney (talk) 04:16, 6 October 2020 (UTC)

Please let me show you how it can be used right now. When a visitor loads the page the main table is loaded sorted alphabetically by region - at this point, the Row table isn't useful. Now let's consider the visitor wants to know how Iraq ranks by total homicide count among other countries. By clicking the Count column twice, they get the table sorted by Count in descending order and can easily see that Iraq ranks 21st (please see the screenshot: https://ibb.co/n3MMnZW). Now, if, let's say, the visitor wants to know how Colombia ranks by homicide rate. Clicking twice on the Rate column sorts its in descending order, showing the visitor that Colombia is #17 by homicide rate (please see the screenshot: https://ibb.co/Ytv5bgr).

I understand that the Row table will not be useful to anyone who isn't interested in countries' rankings. However, that's a poor argument for its removal. Of course, if there is a better solution to the ranking problem, then you are free to implement it. Enivid (talk) 12:06, 6 October 2020 (UTC)

I agree with Enivid. And for those who are interested see previous discussion higher up:

#Sorry, still baffled

See also: phab:T42618 - jquery.tablesorter: Add support for a "fixed" column of row numbers.

JohnFromPinckney, you wrote: "Users with screen readers will surely by stymied by the weirdness of this mark-up."

What exactly are they getting from their screen reader? Maybe you should ask a blind person with a screen reader. It might not be as much of a problem as you are guessing it is.

I have asked people with screen readers to compare some tables, and found out that the assumptions people make have often been wrong. Do some tests and get back to us.

Let the developers at phab:T42618 know the results of these tests.

--Timeshifter (talk) 13:29, 6 October 2020 (UTC)

Hi, Timeshifter. Those who are interested to see the "previous discussion higher up", can follow the link to that very discussion which I included in my first sentence; you needn't have repeated it.

I frankly do not know what users of screen readers actually get, but I assume that, when (if) they get to that first nested table, they hear the row heading "Row", then "end of table" or "last column", then if they move down they get "one", then "two", then "three", and so on. If they somehow get to the second nested table, they will hear the row heading "Country ..." and can move around in the data table as they usually would, but experience no connection to that other inexplicable table, with the single column "Row".

Neither of these nested tables are captioned, only the outer containing table.

We are advised here, among other places, to avoid using nested tables, as they "are very confusing for screen reader users".

I (continue to) see no reason for us to produce inscrutable and non-semantic HTML, without clear captioning, just so that some people can have the feeling (on some browsers, sometimes) that they can visually see that Peru ranks 112 (or whatever), which, as I show below, doesn't even work.

As for phab:T42618, I do not think it appropriate for us to continue to serve semantically defective, poorly accessible and visually unusable code, under the fantastic notion that a phab issue from 2012 which nobody is working on (among many, many issues that are more highly prioritised and are themselves not getting worked on), will somehow, soon, get addressed.

The fact of the matter, IMO, is, that the desired feature will be, at best, a long time coming. I do not have a better approach to offer in terms of dynamic numbering (I wish I did!). So the best and most reasonable approach is to reinstate my edit (i.e., get rid of the Row "table", properly caption the data table, bite the bullet and do without a ranking column.). — JohnFromPinckney (talk) 02:21, 7 October 2020 (UTC)

Hello, Enivid. Thanks for your response. Setting aside for a moment the irrefutable fact that the Row "table" has absolutely no semantic connection to the country data (which I would have thought would be sufficient reason to abandon this silliness), I will try again to explain how wrong your assumptions about visual correlations between the separate tables are.

Nobody answered my question above, where I asked, What is the meaning of the 7th cell, the one with the content '7'? This is one of my first questions, even before I look at the wikicode, because here is what I see:

I'm no expert with uploading images, so I don't know how long or whether you can see my screenshot above. It shows what I see on my computer (in FF, but also IE): a steady alignment drift of the cells in the 1st "table" relative to the adjacent one. So: Is Malawi number 7? Number 8? Number 9? I can't easily tell. And of course it gets worse if I follow your instructions on "how it can be used right now", which I did. The results are depicted in the next two screenshots here.

• 
  Sorted by descending "Count", as per Enivid, showing area with "Iraq"
• 
  Sorted by descending "Rate", as per Enivid, showing area with "Columbia"
• 
  Unsorted data table, showing how much sooner the "Row" table ends

Contrary to your claims (and your screenshot, thanks!), I can not "easily see that Iraq ranks 21st". For me, it's sort of 22 or 23ish. When I sort by "Rate", as you suggest, I get Columbia at a very solid 18, which unfortunately varies from your 17.

The last screenshot here shows the last 3-1/2 cells of the Row "table", and (if I count correctly) the last 19 of the actual data table.

I do not find this juxtaposition of two completely disparate and unrelated sets of values to be at all useful; rather more confusing, misleading, and, well... the nicest thing I can call it is silly. There's no connection semantically, and no connection which is usable visually. Let's kill this thing and be reasonable. — JohnFromPinckney (talk) 02:21, 7 October 2020 (UTC)

Row numbers align with the table rows to the right if your screen is wide enough. I added a note above the table about that.

We still don't know what people with screen readers are getting. I created a sandbox (User:Timeshifter/Sandbox122) with a small part of the table with row numbers. And below that the same table without row numbers.

You can find people with screen readers to go to that page, and ask them what they are getting, and whether the row numbers are a significant problem for them.

In past tests comparing various tables a blind person told me that their screen reader had no problem at all with simple tables without scopes. And the current WCAG guidelines do not require scopes on simple tables. The most popular free screen reader has no problem with simple tables without scopes.

In another test comparing tables with and without a separate row for sorting icons the blind person said it was only a minor annoyance, and that the benefits for sighted readers outweighed his minor annoyance.

--Timeshifter (talk) 15:07, 7 October 2020 (UTC)

Okay, I'll make a few final snide remarks before (I hope) dropping this entirely.

Timeshifter, discussions with you have proven to be utterly exasperating. You seem unwilling or unable to read what others say, bent on restating and restating your same assertions which are often not even germane. I am NOT talking about your godammned scope attributes! Please try sticking to one idee fixe at a time.

Your sandbox is a pointless waste of time; your link to it here another pointless addition of useless text. We already have a test case in the article (after Enivid's revert). The current mark-up is inappropriate, but it's right there in the page. If somebody wants to see valid mark-up with an appropriate usage of tables and captions, they can go to my last edit, mentioned above.

You can say, "if your screen is wide enough" and leave all the notes you want (and, if I know you, you will), but I have two wide screens (together) with resizable windows, and at no point, and in no circumstances, do the row numbers align. I won't post another screenshot, so you can trust me or not, that a screen of 1600+1280=2880 pixels is not "wide enough". Your adamant claim is simply incorrect.

I do not expect that I "can find people with screen readers to go to that page". I can't even find people who understand what a table is.

I've had enough of these mindless arguments between people who can't or won't understand each other. So I guess you win. — JohnFromPinckney (talk) 10:22, 8 October 2020 (UTC)

The existence of some edge cases when it doesn't work doesn't cancel the usefulness of the Row table. It renders properly in all major browsers (including iOS Safari and Firefox) even with screen resolutions less than 600px wide. Removing it because it might not show properly in some browsers without replacing it with a similarly functioning facility would be a waste. Enivid (talk) 15:28, 7 October 2020 (UTC)

How could a "table" of only one single column EVER be useful? How can you insist on illegitimate mark-up based on the premise that (only) some browsers show an apparently useful alignment for (only) some users in (only) some circumstances? What kind of justification is that?

There is no similarly functioning facility, so we just have to do without. Sad, but true. Except for the fact that this is not a functioning facility, either. — JohnFromPinckney (talk) 10:22, 8 October 2020 (UTC)

A table of only one single column can be useful in such cases as described by me above. If this some is large enough, then I would insist for the mark-up to remain even if it is illegitimate until a better option is proposed. Enivid (talk) 11:53, 8 October 2020 (UTC)

JohnFromPinckney. There are many wide tables that don't show up well on mobile phones. Tables without row numbers. Are we supposed to delete all those tables for that reason? That is what we would do if we followed your logic.

What browser are you using? Enivid previously wrote: "It renders properly in all major browsers (including iOS Safari and Firefox) even with screen resolutions less than 600px wide."

So I don't understand why the row numbers do not align with the table rows on your wide screen.

And I pointed out the various examples of screen reader tests in order to illustrate that what MOS warriors say is often at odds with what blind people with screen readers experience. --Timeshifter (talk) 13:28, 8 October 2020 (UTC)

It doesn't make sense to compare statistics from different years

This list doesn't make sense. It doesn't make sense to compare, say, Mayotte in 2009, with Kenya in 2018. In the 9 years hence, how has Mayotte improved or degraded? This list can't tell you that. Is Mayotte still worse than Kenya? Who knows, looking at this list. — Preceding unsigned comment added by 70.184.23.23 (talk) 02:46, 9 November 2020 (UTC)

These are the latest years provided by the UNODC reference used for the main table. --Timeshifter (talk) 04:12, 10 November 2020 (UTC)

Where are the 2017 and 2018 numbers coming from?

Enivid. I only see data up to 2016 here.:

• https://dataunodc.un.org/crime/intentional-homicide-victims

Where are the 2017 and 2018 numbers coming from? --Timeshifter (talk) 05:50, 16 November 2020 (UTC)

The numbers for 2017 and 2018 are coming from https://dataunodc.un.org/content/data/homicide/homicide-rate - the second reference link for the table. Enivid (talk) 12:30, 16 November 2020 (UTC)

Enivid. Could you explain exactly how you are getting all the numbers? What are you clicking? What software (if any) you are using. On table pages I like to add that to the references. It is very helpful to people verifying references. And it helps other editors do updates too. On this table page and others. --Timeshifter (talk) 22:47, 16 November 2020 (UTC)

There isn't much to explain. I just click the reference link and I get this view, which already has 2017 and 2018 listed (for Afghanistan by default) and you can change countries/regions and see the data. When I want to check lots of countries at the same time or with the UNODC's sources, I just click the XLS icon in the top left corner of that page (it reads "Bulk data download" on mouseover). This downloads an XLS spreadsheet with all the data. I use Firefox if that helps. Enivid (talk) 11:24, 17 November 2020 (UTC)

Thanks, Enivid. What spreadsheet software do you use? By the way, that bulk download link doesn't work for me when it is next to the social media links. Only until I scroll to get it away from those links does it work. And even then with difficulty. It's weird.

I use LibreOffice Calc. But I am a newb at spreadsheets. I figured out enough to incorporate some basic stuff into the Visual Editor section at the end of Help:Table.

I see a lot of by-country XLS tables (from OECD) etc. in the same format as the homicide rate XLS spreadsheet when opened in Calc. But I can't use most of them for Wikipedia tables because I can't figure out how to quickly remove all the data except for the most recent year for each country. If the data was in columns by year it would be easier. But even then it is difficult when the latest year is different for many nations.

Do you know some ways to do that? I want to incorporate some how-to info in Help:Table and pages like this:

Commons:Commons:Convert tables and charts to wiki code or image files

I wonder if there is some kind of regular expression I could use to clean up such data when converted to comma-separated values. To remove all entries except for the latest year.

--Timeshifter (talk) 23:30, 17 November 2020 (UTC)

It usually requires two clicks from me for the download to initiate with that XLS icon. I use Microsoft Excel 2016 with that table. I am not an expert in spreadsheets, but I believe it might be possible to do what you are looking for with PivotTable function in Excel. It allows you to transform a table by moving rows to columns and summarizing using set rules. Enivid (talk) 09:09, 18 November 2020 (UTC)

Row number note and class=nowrap

Row number note for the main table currently says:

• Row numbers are static. Other columns are sortable. This allows ranking of any column.

It previously said:

• Row numbers align with the table rows to the right if your screen is wide enough.

Since class=nowrap is in the top line of the table wikitext, then it does not matter how wide the screen is. The rows will always align, since the table never wraps.

Narrow your browser width to see. --Timeshifter (talk) 15:30, 26 December 2020 (UTC)

The rows will always align except when they don't, as I pointed out in § Silly table, again: "Row" above. I'm using the Modern skin on Firefox (still) and no amount of changing viewer width makes things better or worse, despite your repeated exhortations to care about this. "The rows will sometimes align" is the more accurate statement. — JohnFromPinckney (talk) 09:08, 27 December 2020 (UTC)

Ok JohnFromPinckney. I will take your word for it. Fortunately, the word "align" is not found anywhere in the article. So we are not making the claim in the article itself that "the rows will always align".

There are almost always some cases where tables don't work consistently across all browsers, skins, operating systems, mobile, desktop, laptop, tablet, Android, iOS, etc..

The table rows align in my Firefox browser (default skin) on my 2 Windows 10 desktops, 1 Windows 10 laptop, and my Firefox browser in latest iOS on my iphone X.

See another example: List of U.S. states and territories by incarceration and correctional supervision rate. Nearly all the tables have row numbers and class=nowrap.

All the rows in the tables with row numbers on that page stay aligned on all my devices. Even the table that is widest when viewed on the iphone X. It extends past the screen, but the rows stay aligned. I can drag the screen right and left on my iphone to see the whole table.

What device are you using, and what operating system? And what is the screen resolution? --Timeshifter (talk) 18:05, 29 December 2020 (UTC)

PDF page numbers in Adobe Acrobat Reader

Concerning page numbers in the first note in the Notes section of the article; one has to look at the page number in the bottom right corner of pages in the PDF document. Ignore the page numbers provided by Adobe Acrobat Reader.

In case others are confused (as I was), see this discussion thread:

• https://community.adobe.com/t5/acrobat-reader/incorrect-page-number/td-p/8798059?page=1

From that thread: "For instance, if the document/publication has a cover and table of contents those would be pages 1 and 2 since Reader is showing you the literal position you are at in the document. That would make page 1 into page 3." --Timeshifter (talk) 14:41, 28 January 2021 (UTC)

Australia table removed from the article

The other country tables were moved to dedicated country pages for homicide rates. This Australia table was the last one to be removed from the article.

I did not see a dedicated Australia page for homicide rates. So I removed the Australia table from the article and put it here on this talk page. I also put it on this page:

• Wikipedia:Australian Wikipedians' notice board. See diff. It will eventually end up in the archives there.

It was removed like the other individual country tables in order to shorten the article length here, and to hopefully have more concentrated editing on the individual country tables elsewhere. They tend to get forgotten on this global page.

Feel free to use this table on an existing Australia page. Maybe integrate it into another table. Or create a new page just for Australian homicide rates.

---

Intentional homicide rate (per 100,000)

< 1

1–5

5–10

10–21.1

> 21.1

Australia	Sources	2002	2003	2004	2005	2006	2007	2008	2009	2010	2011
Northern Territory		1.3	1.2	1.1	1.0	1.6	1.0	1.1	1.2	1.4	1.0
Tasmania		1.9	1.2	0.4	0.6	0.9	0.6	1.4	0.9	1.0	0.8
New South Wales		1.8	1.7	2.1	1.8	1.7	2.4	1.6	1.3	1.4	1.6
Western Australia		1.5	1.4	1.6	1.2	1.5	1.6	1.3	1.1	1.4	1.6
Queensland		1.9	2.2	1.3	1.6	1.4	1.5	1.3	1.3	1.2	1.2
Australian Capital Territory		0.0	0.6	0.9	0.9	0.6	0.4	1.2	0.9	0.9	1.1
Victoria		1.8	1.7	1.6	2.2	1.4	1.7	2.4	2.1	1.1	1.4
South Australia		1.2	0.4	0.7	1.0	0.9	0.7	0.8	1.1	0.8	1.0

References

--Timeshifter (talk) 20:07, 26 February 2021 (UTC)

What a bunch of Bologni.

There is no chance in HELL, that Brazil is safer than the United States. Brazil is a cesspull of crime and murder. Allanana79 (talk) 17:21, 26 April 2021 (UTC)

The article doesn't claim anything like that. If you have any suggestions to improve the article, you should be clearer. Enivid (talk) 14:22, 27 April 2021 (UTC)

Line height in "Static column begin" template

It looks like the current use of Template:Static column begin results in line numbers column appearing with an increased line height, which doesn't align with the main table. The interesting part is that it looks normal when viewing a preview while editing the page but breaks down when viewing after publishing. Perhaps someone could look into what's wrong with it? Because I've tried and failed to make sense of it. Enivid (talk) 14:26, 27 April 2021 (UTC)

@Enivid: this is currently affecting multiple articles, see: Talk:List of countries by incarceration rate#Tables have messed up index columns. – Finnusertop (talk ⋅ contribs) 14:31, 27 April 2021 (UTC)

Add column with WHO estimated homicide rate?

The WHO published a 2014 report which includes their best estimates for homicide rate per country in 2012. For countries having a lack of reliable data, the estimates are based on a regression model. You can read about their methodology starting on page 62 of the report, and the table of estimated homicide rates begins on page 231. (These estimates are briefly mentioned in this article, as they're the numbers used by UNODC for the 70 countries for which no official numbers are available.)

I'm wondering whether it would be useful to include these numbers as a separate column. As we mention in the intro (and as has been discussed on this talk page before) "the reliability of underlying national murder rate data may vary." I think the addition of a column with the WHO estimated homicide rate would help readers understand which countries' numbers are most uncertain. e.g. the homicide rate in Sierra Leone according to criminal justice statistics published by UNODC is 1.9/100,000, but the WHO estimate is 13.0/100,000 (95% CI 2.9-55.7). Colin M (talk) 15:10, 5 May 2021 (UTC)

Are those the "WHO" entries in the collapsed sources table?

Maybe you could put a note in front of that collapsed table explaining this. With a link to the WHO reference.

I don't want old 2012 WHO numbers except when nothing more recent is available from UNODC-vetted sources.

You could mention that the UNODC numbers may vary from the WHO numbers. And that readers can go to the WHO source for more info. --Timeshifter (talk) 15:21, 5 May 2021 (UTC)

Yes, same as the "WHO Estimate" value in the sources table. I can see how it would be non-ideal to put 2012 estimates next to more recent numbers, but I think it's important to give the reader context on the reliability of official numbers for individual countries. What about a separate section+table for just the WHO estimates? Colin M (talk) 15:31, 5 May 2021 (UTC)

I think it would make a difficult article even more difficult to edit. Many people have tried to add data from other sources than UNODC. It was a constant battle until multiple notes were placed in the article, the editing window, etc. to warn people that anything other than UNODC would be deleted. It still happens, though with less frequency. --Timeshifter (talk) 16:36, 5 May 2021 (UTC)

Where do you see the difficulty coming in? It would be one static data source which would apply to all countries. It seems like most of the difficulty in the past has arisen from people wanting to add certain numbers for certain countries from different authorities on an ad-hoc basis. Colin M (talk) 16:51, 5 May 2021 (UTC)

Besides the explanation in the note above the sources table, you could add it as an external link. You could point to the external link section from above the sources table. Explain further in the external links section. But I really don't see the need for another table in this article. --Timeshifter (talk) 19:28, 5 May 2021 (UTC)

"The time allocated for running scripts has expired"

Pppery. See diff.

I don't know if you noticed this, but your changes ended up with this at the end of the main table on the last 6 rows:

"The time allocated for running scripts has expired" --Timeshifter (talk) 02:00, 18 September 2021 (UTC)

Thanks for pointing that out. I did in fact not notice that (it's easy to miss things like that when making giant repetitive edits).  Fixed. Given that you've started a discussion here, I might as well ask for your opinion on the merit of the edit (is that the correct way of replacing {{flaglist+link}} given the delete consensus at TfD)? * Pppery * _{it has begun...} 02:04, 18 September 2021 (UTC)

Pppery. Please see your talk page. --Timeshifter (talk) 03:25, 18 September 2021 (UTC)

Wikilinks to continents on every line?

@Timeshifter: Is there really a point in adding those wikilinks to the continents in the homicide rates table? — Preceding unsigned comment added by Vgbyp (talk • contribs) 18 November 2021 (UTC)

See JohnFromPinckney posts in the previous talk section. Regions and subregions have long been linked on this country table and others.

Some people, myself included, like comparing homicide rates within subregions or regions. I like looking up info on those countries to try to figure out why one particular subregion or region has higher or lower rates than a different region or subregion.

To see how the links are added all at once to a column see:

User talk:PrimeHunter#Is there a regular expression that will add link brackets to text in each cell in a column?

--Timeshifter (talk) 16:53, 18 November 2021 (UTC)

I get why we need the country and subregion wikilinks, but regions? They seem rather irrelevant here. Is there a big enough chance that someone who doesn't know what Africa or Asia is browses this table? Countries obviously need wikilinks; subregions need wikilinks because the constant UN geoscheme / non-UN geoschemes confusion. Anyway, linked region names isn't a big deal, but I thought they were utterly redundant here. Vgbyp (talk) 13:56, 19 November 2021 (UTC)

It can be confusing along the edges of some regions. And the Wikipedia pages for regions explain the various different schemes used over time. --Timeshifter (talk) 14:05, 19 November 2021 (UTC)

Main table completely updated from UNODC source

See User:Timeshifter/Sandbox165 for info and links.

The table is now in alphabetical order by country.

Random editors, and regular editors, sometimes find updates at the source, and enter the new data here for one country, or several countries.

That can mess up a table that is in rank order. That is why I put the table in alphabetical order by country.

I did the update because it seems people were finding a few errors. It was easier to do a full update rather than to look for errors. The table data may not actually be that different from before.

I prefer the single decimal place in the rates. The UNODC sources uses the single decimal place in the more easily accessible info. The bulk data download file has many decimal places for the rates. That is overkill for Wikipedia tables. --Timeshifter (talk) 23:11, 2 November 2021 (UTC)

Cool. Was it an oversight that you removed the style="text-align:center" for the Region and Subregion columns? That aspect is, well, not an improvement. — JohnFromPinckney (talk / edits) 00:01, 3 November 2021 (UTC)

When you start from the data file there is no CSS styling. The styling I usually do comes from adding Template:Static row numbers and its associated sub-templates. Also, some flag templates add left alignment of the country column links. As with the flag template used in this table.

I only really care about styling the country names (left aligned) and the data columns (right aligned). Others can work on additional styling. --Timeshifter (talk) 02:57, 3 November 2021 (UTC)

Not if you're going to dump another data file in a few weeks and overwrite the styling. Isn't that how you do it? — JohnFromPinckney (talk / edits) 14:50, 3 November 2021 (UTC)

See: User:Timeshifter/Sandbox165. Data starts as plain text. I don't intend to update again soon. I may never update the table fully again. For one thing I don't know how to get that plain text table. I copied that from here:

Wikipedia:Reference desk/Archives/Computing/2021 October 28#How can I see only the latest year for each country in a spreadsheet?

If someone provides another plain text table with later data, then I may update fully again.

Of course, some styling is lost. But what is more important, the latest data, or styling?

--Timeshifter (talk) 00:15, 4 November 2021 (UTC)

Here is a way I understand for getting the latest data available for each country:

• LibreOffice Calc (freeware) method:
• How can I see only the latest year for each country in a spreadsheet?

It needs to be summarized, and rewritten more clearly, and with more illustrations. On a separate page, or in a sandbox. I may or may not get around to it. --Timeshifter (talk) 06:48, 29 November 2021 (UTC)

Kenya?

In what world is Kenya more safe than Tanzania? — Preceding unsigned comment added by Nlivataye (talk • contribs) 15:06, 29 May 2022 (UTC)

Could the figure be deleted, or replaced with an accurate one?

Fustbariclation (talk) 16:18, 2 July 2022 (UTC)

Contact the map and chart creators. Check the categories of the existing maps and charts. Find better maps and charts. Create your own. Check other categories on the Commons. Find maps and charts elsewhere online. Many are in the public domain. See:

commons: Commons:Threshold of originality#Maps

commons: Commons: Category:PD map

commons: Commons:Threshold of originality#Charts

commons: Commons: Category:PD chart

--Timeshifter (talk) 17:08, 2 July 2022 (UTC)

UNODC link not working

Hey, if I go to the main/primary link for the data at dataunodc.un.org, it just says the content isn't available. They may have done a switcheroo on the link path. I don't know the exact specifications for the data search, so I don't want to muck with it - so, if someone who's familiar with it can check and fix as needed, that would be great. cheers. anastrophe, an editor he is. 23:50, 9 October 2022 (UTC)

They've moved their data page. I've now updated the reference link. Vgbyp (talk) 07:01, 10 October 2022 (UTC)

Excellent, thank you very much. cheers. anastrophe, an editor he is. 20:36, 10 October 2022 (UTC)

Need table fully updated

I don't currently have the health, time, or energy to do this.

See:

• User:Timeshifter/Sandbox165
• See: Help:Table#Picking selected dates from massive .csv files. Scroll down to the info on List of countries by intentional homicide rate. From that help page currently:

List of countries by intentional homicide rate has a country table that uses a source with multiple years. But the table only shows data for the latest available year for each country. Some more complex transposing needs to be done with the source spreadsheet. To do it quickly see these threads:

• How can I see only the latest year for each country in a spreadsheet? - for Calc.
• Wikipedia:Reference desk/Archives/Computing/2021 October 28#How can I see only the latest year for each country in a spreadsheet? - for Excel.

Those threads need to be rewritten and simplified onto new pages. See also this section: Help:Table#Converting rows to columns and columns to rows. --Timeshifter (talk) 18:04, 22 December 2022 (UTC)

I've just updated the table using a much simpler approach:

1. Use Filters to get only "Total" values for everything and only the "Victims of intentional homicide" data. The result is saved as a separate table.

2. This table is is then Filtered to show only Counts.

3. It is then Sorted by Year (descending).

4. The resulting rows are copied to a new sheet.

5. On that new sheet, Remove Duplicates is applied based on the Country column. As a result, we now have here all the countries with latest year Count values.

6. Return to the table from step 2. Switch the Filter from Counts to Rates now.

7. Copy the rows to a new sheet.

8. On that new sheet, Remove Duplicates is applied based on the Country column. As a result, we now have here all the countries with latest year Rates values.

9. Copy all the rows to the table in step 5.

10. Sort by Country. It will now have all countries with two rows per each - the latest year's with Count and Rate.

It is then trivial to compare the existing Wikipedia table to the UNODC table. It cannot automatically generate the entire table, but when there are just a few changes to the data, it is a simple and rather fast approach.

Vgbyp (talk) 15:57, 23 December 2022 (UTC)

Vgbyp. Thanks! I count around 40+ countries you updated. What software did you use to do all this? And how much time did it all take? --Timeshifter (talk) 23:57, 23 December 2022 (UTC)

Excel 2016. About an hour total - 30 minutes to figure out how to be sort and filter the data and then 30 minutes to go through the table, comparing and editing what needed to be updated. Vgbyp (talk) 12:34, 24 December 2022 (UTC)

Vgbyp. It would be nice if you, and others, could list the exact steps for doing the same, or similar, in freeware LibreOffice Calc. It is what I use. But I am a newb using it, and I soon forget stuff I learn with it. Because I don't use it often enough. --Timeshifter (talk) 16:44, 24 December 2022 (UTC)

I am no expert in LibreOffice Calc, but I believe that if it can open the downloaded .xlsx file from UNODC and if it shows the sorting/filtering controls there, then the steps are pretty much the same because they don't even involve any formulas. Vgbyp (talk) 20:16, 24 December 2022 (UTC)

Vgbyp. I wrote all of this section: Help:Table#Picking selected dates from massive .csv files. I stopped updating that table it is explaining because I just don't have the health, time, and energy right now. And so I concentrate on other things on Wikipedia. Even with all those detailed instructions I would probably have problems. Each time I updated the table I noticed things needing updating in the instructions. So I did so.

Same is true for this section, and the one that follows it:

Help:Tables and locations#Add flags. Link countries, states, etc. in tables. That section is so clear now that lots of newbs use it.

Anyway if you feel inspired you could create subpages here with some instructions: Talk:List of countries by intentional homicide rate/LibreOffice Calc instructions and Talk:List of countries by intentional homicide rate/Excel instructions.

I started the Excel subpage. --Timeshifter (talk) 21:35, 24 December 2022 (UTC)

I started the Calc subpage: Talk:List of countries by intentional homicide rate/LibreOffice Calc instructions. --Timeshifter (talk) 14:11, 18 January 2023 (UTC)

UNODC-vetted

Note: First post was moved from a user talk page to here so that others can read and participate if they so choose.

Can you explain why you say only unodc is allowed yet the tables source column contains:

CTS SDG WHO DATASUS MD MOJ EXTERNAL NSP USMIL DPKO?

Rote1234 (talk) 13:11, 18 January 2023 (UTC)

Rote1234. We require that all the external data is examined and vetted by UNODC. Using a vetting source is common for many Wikipedia lists and tables. --Timeshifter (talk) 14:05, 18 January 2023 (UTC)

My question is then why does the source column for various countries say: CTS SDG WHO DATASUS MD MOJ EXTERNAL NSP USMIL DPKO?Rote1234 (talk) 14:07, 18 January 2023 (UTC)

Because they are the sources. UNODC is not the source. UNODC is just vetting the data. --Timeshifter (talk) 14:13, 18 January 2023 (UTC)

Ok then I will just add a sentence of this new source under the Other Studies section since other list articles on wiki also allow multiple studies.Rote1234 (talk) 14:19, 18 January 2023 (UTC)

The source you found is for a single country. It is not a study with data for multiple countries. --Timeshifter (talk) 14:46, 18 January 2023 (UTC)

That's fine it can go under the Other Studies section of the article that allows for newer studies on select countries. Rote1234 (talk) 15:07, 18 January 2023 (UTC)

No it can't. That would require a new consensus. This has been discussed before. See the talk page archives. --Timeshifter (talk) 15:21, 18 January 2023 (UTC)

The other studies section has a sentence about Jamaica and another about Asturias. Of course it can have a study about Brazil. Rote1234 (talk) 15:28, 18 January 2023 (UTC)

In general, there is nothing wrong in adding non-UNODC data in the 'Other studies' section, however, it needs to be notable for some reason for it to require a mention there - similar to how the current two entries there are notable. It wouldn't be a good idea to add just a simple statement on the number of intentional homicides in a single country. Vgbyp (talk) 18:36, 18 January 2023 (UTC)

You are correct and it's notable because the source, along with many other articles that celebrate the fact, says that it's the lowest amount in 15 years! Rote1234 (talk) 19:10, 18 January 2023 (UTC)

Then we could have little notes about dozens of countries having the highest or lowest amount in a certain number of years. And some of it would go out of date every year. Then we would be back to previous versions of this page with out-of-date individual country tables, and so on. That kind of info was moved to the "Crime in LOCATION" articles, and into other articles where it could be kept up to date. See the talk archives for a record of this.

Your individual country info could be put in Crime in Brazil. --Timeshifter (talk) 21:56, 18 January 2023 (UTC)

Moved to Talk:Crime in Spain

This below was moved from List of countries by intentional homicide rate to Talk:Crime in Spain.

On the other side, Asturias region, in Northern Spain, has one of world's lowest homicide rates. With a population of 1 million people, it registered only 1 homicide during full year 2021. Its biggest city, Gijón, with more than 250,000 inhabitants, has not registered a homicide since February 2020 (as of June 2022), before COVID-19 lockdown.

References

See talk archives here for many other examples of individual country info and tables being moved elsewhere where they can be put in context, and kept up to date. --Timeshifter (talk) 05:27, 19 January 2023 (UTC)

Botswana and Namibia?😳

I think we need evidence that Botswana and Namibia are way above on the homicide rate despite being the safest and less populous stable democracies in Africa Nlivataye (talk) 08:07, 29 January 2023 (UTC)

We use UNODC as a reliable source for the data in this table. The latest year for each country varies, so it might explain the difference in rates you see. Vgbyp (talk) 10:26, 29 January 2023 (UTC)

World Bank page with great table of latest data

If you scroll down a little on this World Bank page you will see a table of the latest homicide rates by country. With the year of the data for each country.

• https://data.worldbank.org/indicator/VC.IHR.PSRC.P5?end=2020&locations=PR&start=1990&view=chart

It would be a good secondary table to add to the article.

Unfortunately I haven't figured out how to copy that table off the page and into a sandbox in table form. Any ideas? --Timeshifter (talk) 01:03, 5 July 2023 (UTC)

I could probably generate a table using their CSV. Which columns do you think should be included in the table and should it get colored like the UNODC table? Vgbyp (talk) 10:13, 5 July 2023 (UTC)

I don't see any coloring on the World Bank page, nor in the dataset.

The dataset opens correctly. No pivot needed. Years back to 1990.

The problem is that some countries have no recent data. So one solution is to break up the dataset into several groups of years. I don't have a problem with that. It is a simple solution.

But I was hoping to also have a World Bank table with just the most recent year available. I believe you figured this out for the UNODC table:

Talk:List of countries by intentional homicide rate/Excel instructions

Other instructions:

Talk:List of countries by intentional homicide rate/LibreOffice Calc instructions

So maybe you can figure it out for the World Bank table. --Timeshifter (talk) 16:53, 5 July 2023 (UTC)

OK, I've got the latest rate/year values for each country using Excel's VLOOKUP function. Now what do I do with it? Where do I paste it? Vgbyp (talk) 19:33, 6 July 2023 (UTC)

I suggest creating a "World Bank table" section (or similar section heading) just before this section: "Other multi-country studies". --Timeshifter (talk) 03:47, 7 July 2023 (UTC)

You know, I've finally converted it all to a Wiki table format and was about to insert a new section in the article when I noticed that the World Bank's website marks its intentional homicide statistics as based on 'UN Office on Drugs and Crime's International Homicide Statistics database.' Doesn't this mean that they are taking it from the same UNODC table (just at some unspecified frequency)? Vgbyp (talk) 08:27, 7 July 2023 (UTC)

Can you paste what you have so far in a permanent sandbox? It's only permanent if it is named or numbered. I created this one for you if you don't want to set up your own:

• User:Timeshifter/Sandbox213

I want to see exactly what you have. For a variety of reasons. --Timeshifter (talk) 17:05, 7 July 2023 (UTC)

I see the table of latest rates in the sandbox. I think it should go in the article. It can be described as a World Bank table based on UNODC data. I don't think the rates for groups of countries are needed.

I think the World Bank timelines should be in the article also. World Bank tables based on UNODC data. For all years. That means those timelines can be broken up into around 3 parts. I know how to do that fairly fast.

I think the old UNODC chart should be deleted.

I don't think counts should be in the tables. Counts are sometimes partial counts. Then UNODC estimates counts and rates for the rest of the country. It's the rates that matter when comparing countries.

Plus doing without the counts makes updating the tables regularly a lot easier. --Timeshifter (talk) 00:38, 8 July 2023 (UTC)

My main concern is that the data is the same (or largely the same) as in the UNODC table. Vgbyp (talk) 08:01, 8 July 2023 (UTC)

How much time does it take to create the latest year available table from the World Bank source file versus the UNODC source file? I don't think both should be in the article either. I prefer that the easier one is used. So that others can help out. Like me. Once I figure out your method, but with LibreOffice Calc. --Timeshifter (talk) 17:24, 8 July 2023 (UTC)

It took me about 5 minutes to find out that the LOOKUP function can help me do it and then 5 more minutes to actually do it. However, I didn't clean the data of regions (they are listed along with countries) and N/A cases. Vgbyp (talk) 10:47, 9 July 2023 (UTC)

Thanks. I updated this instruction page: Talk:List of countries by intentional homicide rate/Excel instructions. Feel free to add more info. --Timeshifter (talk) 13:42, 9 July 2023 (UTC)

I agree, it does not make sense at all to add a second table which is based on the same data.

By the way: count is necessary to give an idea about the "size" of the country (amount of inhabitants). -- ZH8000 (talk) 12:11, 8 July 2023 (UTC)

It can be a lot of work to sync 2 columns. 1 for counts and 1 for rates. So many ways to mess it up. And it can be much more work. It depends on how the source data is organized on the spreadsheet. And the article is titled "homicide rate". Syncing a column for population is even more difficult and time consuming. --Timeshifter (talk) 17:24, 8 July 2023 (UTC)

New Zealand homicide rate distorted by Christchurch Mosque Attack

A note needs to be placed explaining that New Zealand's homicide rate for 2019 was distorted from its long running average by 51 deaths in a single mass shooting event on 15 March 2019 targetting mosques in Christchurch, New Zealand. Brokenhearted, not broken: The scale of loss in the Christchurch mosque shootings (stuff.co.nz) 2405:DA40:3758:E000:2D68:BCD3:A28D:C966 (talk) 13:41, 13 August 2023 (UTC)

This article is just a list article. Detailed info should go in the "Crime in LOCATION" articles. I am surprised that longer-term homicide rates being effected by the Christchurch mosque shootings are not mentioned in the Crime in New Zealand article. Maybe you can fix that with a little timeline homicide rate chart there. --Timeshifter (talk) 14:42, 13 August 2023 (UTC)

Update table with latest rates and counts

Vgbyp. See:

• User:Timeshifter/Sandbox213
• User:Timeshifter/Sandbox221

You said it took you 5 minutes to create the homicide rates table in sandbox 213. Could you update it to today's numbers?

And could you do the same for the homicide count numbers in sandbox 221?

I can take it from there. If the number of rows are exactly the same in both sandboxes (as they should be since rates have to be calculated from counts), then I can easily paste the 2 tables together and post it.

I will then replace the current country table in the article. Since the new table would also come from UNODC-vetted data, then little would have to be changed in the article except to add the World Bank reference.

I see now that I didn't state the obvious in the last thread. That if the number of rows was the same for the counts and rates, then creating a table with both was not a problem.

Leave in all the data of regions in both tables. Just to make it easier to have the same number of rows. And it is easier for you. I may or may not keep some or all of the region data in the final data. But first I will create a table in a new sandbox with all the data:

• User:Timeshifter/Sandbox222

I've updated the table in User:Timeshifter/Sandbox213, but the data didn't change since the last time. I couldn't find the homicide count (not rate) data on the World Bank website. Could you please direct me? Vgbyp (talk) 06:19, 9 August 2023 (UTC)

Vgbyp. Thanks for the updated table. I assumed (incorrectly) that the World Bank data downloads would also have the counts.

See: Talk:List of countries by intentional homicide rate/Excel instructions. At one point you "have all countries with two rows per each - the latest year's with Count and Rate." How long does it take you to get to that point?

Could you paste that into User:Timeshifter/Sandbox221?

I want to see if I can use LibreOffice Calc, and the pivot function, etc. to create a column for rates, and a column for counts. In the same table.

I will paste that table into User:Timeshifter/Sandbox222. Feel free to beat me to it. :) --Timeshifter (talk) 13:11, 9 August 2023 (UTC)

I've inserted two versions of the same table into User:Timeshifter/Sandbox221 - one raw and one with rounding, added headers, and removed extra columns. Vgbyp (talk) 07:12, 10 August 2023 (UTC)

Thanks Vgbyp! That made things much easier. I figured out a way to separate the rates and counts into 2 columns. Article table has been fully updated except for some flags and country links. I leave those for others. I may not have the time, health, or energy to get around to it.

See: User:Timeshifter/Sandbox223 and User:Timeshifter/Sandbox224. I left User:Timeshifter/Sandbox222 open if you need a place to work further on the table.

See: Help:Tables and locations#Separate counts and rates to 2 columns

Last version of table is here (to see some of the formatting and names of countries missing flags and links in latest table):

https://en.wikipedia.org/w/index.php?title=List_of_countries_by_intentional_homicide_rate&oldid=1169362282

--Timeshifter (talk) 21:55, 10 August 2023 (UTC)

Vgbyp and others. I got rid of all the red links in all sections of User:Timeshifter/Sandbox224. This makes future updating much simpler since the redirects are permanent. No customization of the {{flagg}} template is needed for individual countries after future updates. I copied that sandbox table to here.

See redirects info in this section here: Help:Table#Adding flags and linking countries, states, etc. in lists.

The only thing needed after future updates is some alphabetization, and some text added for a couple entries:

Location	Region	Subregion	Year	Rates	Counts
Iraq. Central Iraq	Asia	Western Asia	2021	15.4	5,459
United Kingdom *. England and Wales	Europe	Northern Europe	2021	1.2	696

"England and Wales" can be made into a link: England and Wales. --Timeshifter (talk) 17:16, 12 August 2023 (UTC)

I struck out sandbox 222 above since neither of us used it for this purpose. I am going to use it for other purposes now since it is free. --Timeshifter (talk) 14:52, 13 August 2023 (UTC)

Semi-protected edit request on 26 August 2023

There are 2022 updates for many of these countries, as they are currently only showing 2021 and below. An example is El Salvador, going from 18.1 (2021) to 7.8 (2022). Many are outdated. PervazG (talk) 20:29, 26 August 2023 (UTC)

 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. M.Bitton (talk) 20:59, 26 August 2023 (UTC)

PervazG and M.Bitton. Please look at the UNODC reference used for this table:

https://dataunodc.un.org/dp-intentional-homicide-victims

The only numbers for 2022 are for Haiti, and they are incomplete: 2021 for rate. 2022 for total homicides. Probably difficult to get an accurate population in the middle of the 2022 chaos in Haiti. Population is needed to calculate the rate. So 2021 numbers are used in the table for Haiti.

As noted in multiple places only UNODC-vetted numbers are used in this table. --Timeshifter (talk) 00:39, 27 August 2023 (UTC)

Understood. I personally did not know about the UNODC rule for this table. Disregard my comment. Podlerman (talk) 01:35, 28 August 2023 (UTC)