Talk:Casualties of the Israel–Hamas war

Inaccurate Description of UN Data

The pie chart depiction of the UN data has a slight inaccuracy in its description. The text currently reads: "UN analysis over killed Palestinian victims verified only from at least three independent sources in six months between Nov 2023 and April 2024, with the most frequent age among the victims being 5-9 years old." However, when you read the actual report, you can see that these numbers are only discussing people killed in residential structures and the like (see paragraph 12 in the Killing of Civilians section). In addition, OCHA data from the time has the ratio at 40% men, 32% children, 20% women, and 8% elderly. As such, I think the chart as it currently is should be removed. PotatoKugel (talk) 09:53, 29 November 2024 (UTC)

Agree, especially given the fact that the UN analysis only included about 8k fatalities out of about 34k reported at the time, and the report itself acknowledged that their sample is not necessarily representative. DancingOwl (talk) 10:52, 29 November 2024 (UTC)

This same mistake is in the opening paragraph, as well. PotatoKugel (talk) 05:58, 2 December 2024 (UTC)

If you think that there should be an edit, are you able to do it? I don't have enough contributions to edit this article. PotatoKugel (talk) 05:59, 2 December 2024 (UTC)

These are verified deaths per OHCA rather than the Gaza Health Ministry - which again is only a subset of reported deaths. The Gaza Health Ministy figures are acknowledged by OHCA as probbly accurate - but it does not say its own figures are worse than the health ministry's. And there has been enough people casting aspersions on the GHM that I can't see how changing it because of the GHM can be justified. I suppose you could have a separate chart for the GHM verified deaths figures. NadVolum (talk) 10:22, 2 December 2024 (UTC)

What about the fact that this ratio is only talking about deaths which occurred in residential structures? The article does not mention that this ratio only applies to deaths in residential structures.

Also what about the points brought up by DancingOwl? PotatoKugel (talk) 10:47, 2 December 2024 (UTC)

My reading is like that of the BBC - that about 80% of the people they were talking about were killed in residential or suchlike buildings and the figures refer to them all. And as to it being an 8k, that's perfectly fine statistically and suffers from exactly the same problem of today's 43,000 or whatever verified figure from GHM now and the BBC article about the report did not see anything like that as important enough to note. NadVolum (talk) 15:18, 2 December 2024 (UTC)

A few days ago Michael Spagat published an article, explaining why the sample is not representative of the whole:

https://aoav.org.uk/2024/the-ohchr-report-on-gaza-insights-flaws-and-the-wheres-daddy-programme/ DancingOwl (talk) 15:48, 2 December 2024 (UTC)

That's a good source, not the only thing it explains. Selfstudier (talk) 15:58, 2 December 2024 (UTC)

I have a lot to say about the "where's daddy part", but that would be wp:or, so I'll shut up :) DancingOwl (talk) 16:19, 2 December 2024 (UTC)

Agree that's a good source. I think Spagat should have twigged by now that the figure for minors is far too low - they should be about double that for women but it is only two thirds that. Oh well, that would also be WP:OR. NadVolum (talk) 01:01, 3 December 2024 (UTC).

I don't get his assumption that there are an equal number of male and female civilians. I think there is some truth in it, but it depends on the situation. For example, if Israel kills a Hamas militant, his wife, and three children on average for x amount of families, the ratio will be 20% men, 20% women, and 60% children. According to Dr. Spagat's estimate, we should assume that none of the men were members of Hamas. Yet, the scenario I am describing does not sound unreasonable to me at all, especially considering the fact that the report centers on deaths in residential housing (meaning, where it is perhaps more likely for there to be an equal number adult males and females, regardless of whether those males are part of Hamas of not). This is obviously wp:or, but I am curious to hear your thoughts. PotatoKugel (talk) 04:42, 3 December 2024 (UTC)

Sorry, to clarify, are you saying that 80% of the recorded deaths occurred in buildings and residential structures, but the percentage breakdowns (30% men, 70% combined women and children) were talking about all of the recorded deaths, not just the ones from inside buildings and residential structures? PotatoKugel (talk) 03:38, 3 December 2024 (UTC)

What the report says is not obvious to me but that is what the BBC seems to say. The assumption that there are equal numbers of male and female civilians amongst the dead is a fairly conservative way of estimating the number of militants - it assumes that besides militants everyone else is killed randomly. A count on the ground has indicated it is more like 65% of the dead men are civilians which would mean many more civilian men are killed than women. Your reasoning I think is that if the where's daddy programme is a major cause of death then there should be less civilian men killed than women. As far as I can make out though the where's daddy programme has only been able to track a fraction of militants - and often just Hamas civil servants and police rather than actual militants. In fact reading about how the Israeli's checked their Lavender system if I'm reading it right, and I very much hope they were not doing this but that's what it says, they wanted 90% accuracy in identifying a militant. That seems like a fundamental statistical mistake to me considering that the militants are such a small minority, the 10% error would identify maybe five or ten times as many people who were not militants as militants as it identified actual militants. NadVolum (talk) 11:53, 3 December 2024 (UTC)

I have been told by Selfstudier and Sean.hoyland that I need to be extended-confirmed to be allowed to have this discussion, so I will have to bow out here.

Thank you for your time and have a great day! PotatoKugel (talk) 15:07, 4 December 2024 (UTC)

Trying to give specific edit requests here.

I would say that the caption on the pie chart which currently reads: "UN analysis over killed Palestinian victims verified only from at least three independent sources in six months between Nov 2023 and April 2024, with the most frequent age among the victims being 5-9 years old." Should be changed to "UN analysis over killed Palestinian victims in residential buildings or homes, verified only from at least three independent sources in six months between Nov 2023 and April 2024, with the most frequent age among the victims being 5-9 years old."

I would also like to change the opening paragraph from "In Nov 2024, the UN published its analysis covering only victims verified from at least three independent sources over 6 months span between Nov 2023 and April 2024 found that 70% of Palestinian deaths in Gaza are women and children." to "In Nov 2024, the UN published its analysis covering only victims verified from at least three independent sources over 6 months span between Nov 2023 and April 2024 found that 70% of Palestinian deaths in residential buildings or homes in Gaza are women and children." PotatoKugel (talk) 03:33, 5 December 2024 (UTC)

I checked with the figures before the report talked about residential buildings and the only change was 25% women rather than 26% women so nothing appreciable that I can see. Both the current references talk about the figures as they are - they may be mistaken in how they got there but it doesn't seem to make any difference so I'll leave it. I'll add a reference to the report as a primary source so peple can go to it if they want. NadVolum (talk) 13:19, 5 December 2024 (UTC)

New edit request: Perhaps the Gaza Health Ministry data from the time should be added to the first paragraph as well, which says that 52% of those killed were women and children. Also, perhaps, a second pie chart should be added reflecting the GMH data, or perhaps it should be removed entirely. PotatoKugel (talk) 14:30, 5 December 2024 (UTC)

Reactions and analysis section - suggested edit

In paragraph three of the reactions and analysis subheading, I think the last sentence should be amended due to stylistic and explanatory inconsistency. It currently reads "The Lancet correspondence has been criticized by the Chair of Every Casualty Counts network Prof. Michael Spagat, who wrote that it "lacks a solid foundation and is implausible"." This is an important line, but it lacks context as to his professional background and the mathematical and statistical nature of the organisation. The 'Every Casualty Counts' organisation has no immediate link for the reader, and therefore does not immediately explain that his role as chair is as a mathematics and economics of war researcher, that unlike the previously mentioned doctors does not have his professional credentials mentioned, only his titles. Examples of this type of useful further clarification in the same paragraph that have already been added, include the "British medical journal The Lancet" in sentence one, and "Jean-François Corty, a humanitarian doctor and president of the NGO Doctors of the World," in sentence two.

To amend this styilstic and explanatory inconsistency I suggest the final sentence be amended to the following:

"The Lancet correspondence has been criticized by the Chair of the casualty recording organisation Every Casualty Counts Prof. Michael Spagat, a war statistics researcher who has argued that it "lacks a solid foundation and is implausible"."

I believe it more accurately reflects the fact that the Lancet is being criticised from a statistical analysis point of view, not a medical negligence point of view (which the Lancet is stating the GHM missed)- with it currently (I believe) reading more as a medical criticism of the Lancet (a medical journal) instead of a statistical one.

Please feel free to leave a comment to discuss further and I will reply as soon as able, many thanks. Flarehayr (talk) 22:18, 10 December 2024 (UTC)

Starvation deaths in lead

Has this already been discussed here? If so, apologies. If not, I think it’s wrong to include as the third para of the lead a tendentious estimate from a letter by an ill-defined group that has only two secondary sources, Mother Jones and an inaccurate summary in a piece by an anthropology assistant professor from the dubious Costs of War project. This should go in the body, but isn’t due in the lead unless it gets other more solid reporting. BobFromBrockley (talk) 17:30, 15 December 2024 (UTC)

See Talk:Gaza genocide#RfC about starvation estimate Note that the estimate is not "tendentious", the group is not "ill-defined" and the Costs of War project is not "dubious", that's just made up.

Having said that, probably best not to include for time being. Selfstudier (talk) 17:44, 15 December 2024 (UTC)

So an overwhelming no to including in infobox there. I think same arguments apply to lead here. No objection at all to inclusion in the body. Happy to argue over my choice of words but don't think it's helpful. BobFromBrockley (talk) 11:37, 16 December 2024 (UTC)

Extended-confirmed-protected edit request on 18 December 2024

Number of deaths in Gaza are not reliable as they are provided by Hamas Sofia Cohen (talk) 20:04, 18 December 2024 (UTC)

Not done. The UN and most other reliable sources think they are reliable see BBC News: Israeli strikes kill 50 in Gaza, Hamas-run health ministry says. And as to the Israeli estimate in that of militants killed see the section in the article on Civilian to combatant ratio. NadVolum (talk) 23:21, 18 December 2024 (UTC)

Gaza Ministry oh Health Data

Perhaps the Gaza Health Ministry data from the beginning of the war through April (the same period as the UN report), which says that only 52% of those killed were women and children, should be added to the first paragraph. Also, perhaps, a second pie chart should be added reflecting the GMH data. PotatoKugel (talk) 06:41, 20 December 2024 (UTC)

That is when they started reporting the number of killed people whose identies had been verified. There were also very many people killed who had not been identified. A much higher proportion of people who have been killed have now been identified. Why is that particular date important? NadVolum (talk) 22:13, 20 December 2024 (UTC)