21

This image claims the following:

2016 was recorded to be the deadliest year for the LGBTQ community.

27 transgender people were murdered in 2016, overtaking previous records in 2015 (21 deaths), 2014 (20) and 2013 (18), which shows that transphobic violence is at an all-time high.

I want to know two things. First, is the first statement accurate, that 2016 was the deadliest year for the LBGTQ community?

If it wasn't, then is the implication from the second sentence, that 2016 was the most deadly date for the transgender community only, not the entire LGBTQ, accurate?

To address the usual concerns I've seen mentioned, this image is from cracked.com. However, the site in general claims to provide accurate information, in a comedic fashion, not false information. In addition the 'pictofacts' this image is from are reputed to be 100% accurate and non-comedic facts. As such I would consider this noteworthy, as it will be seen by a large number of individuals and is depicted as being an accurate claim.

jwodder
  • 494
  • 2
  • 9
  • 13
dsollen
  • 10,062
  • 12
  • 43
  • 71
  • 3
    Do you have any additional information of the original claim? Maybe a webpage that hosted the image with additional information on it? That will make it much easier to track down the source of these numbers for verification. If not, I'm sure someone here will find it anyways. – A Bailey Jul 11 '17 at 17:43
  • 1
    I'll go ahead and flatly assume this is about the US. BTW is this about LGBTQ+ or about transgendered people in particular? – Sklivvz Jul 11 '17 at 18:09
  • 3
    @Sklivvz reading the claim as written it supposedly is about LGBTQ+ community, but then they only quote transgendered statistics. I suspect they *meant* it was only deadliest for transgendered and not LGBTQ+ and just tossed in LGBTQ header without ralizing LGBTQ implies far more then trans. Still this is why I asked about both if it was deadliest for LGBTQ and if it was deadliest only for transgendered, to figure out if either interpretation is accurate. – dsollen Jul 11 '17 at 18:20
  • 7
    It is perhaps worth to notice that if the numbers match, are from the US and the estimates on the number of transgender people in the US (1.4 million) are correct, transgender victims are significantly underrepresented in the murder statistics. In the case descriptions in the HRC document BobTheAverage is linking to in his answer, there are also only very few indicators that the cases can be categorized as hate crime or that the sexual identity of the victims had anything to do with the murder. – Tor-Einar Jarnbjo Jul 11 '17 at 19:39
  • 35
    I'd think the mass-shooting event at the Pulse night club in Orlando would badly skew statistics in that regard. – PoloHoleSet Jul 11 '17 at 21:01
  • 18
    27 isn't even a statistically significant fluctuation. If we assume the base rate is 20/year (as suggested by the previous three years), 27 in a single year has a p value 0.08. – eyeballfrog Jul 11 '17 at 21:39
  • 3
    @dsollen please note, we are *transgender* not *transgendered* :) – cat Jul 12 '17 at 01:40
  • 3
    Only 27? Are these statistics just from one country, or something? That number's very low. Please remember that there is a whole world out here; we are not all Americans. – Lightness Races in Orbit Jul 12 '17 at 11:12
  • 1
    @LightnessRacesinOrbit - I believed the referenced claim is probably from a more localized perspective, with the aim of changing our national behavior or viewpoint. It would be difficult for the person to get information from more closed societies, for instance, or to hope to affect change there. It's not the OP who is choosing the sample location, OP just saw a claim and wondered about it, though perhaps including those specifics would improve the question and ability to assess the claim. – PoloHoleSet Jul 12 '17 at 13:57
  • @PoloHoleSet: And, conversely, there are many far more open societies in the world than the USA. – Lightness Races in Orbit Jul 12 '17 at 14:01
  • 2
    @LightnessRacesinOrbit - I'm not claiming otherwise. I'm pointing out the difficulty for someone hoping to enact change in even attempting to get their arms around an accurate "worldwide" number to assess. If someone has a very specific number, that pretty much tells us they aren't talking about the worldwide picture. Having more open societies than the US would not present a similar obstacle. – PoloHoleSet Jul 12 '17 at 14:05
  • @PoloHoleSet: I'm not claiming that you claimed otherwise :) – Lightness Races in Orbit Jul 12 '17 at 14:25
  • 1
    @LightnessRacesinOrbit - we are often delusional about our level of awesomeness here and ignorant about the rest of the world, so if you thought my comment was making certain assumptions about that, it would be understandable. I wanted to clarify because of that, not because I was feeling accused. – PoloHoleSet Jul 12 '17 at 14:27
  • @PoloHoleSet: Good :) – Lightness Races in Orbit Jul 12 '17 at 15:00
  • Disclaimer: quick data got from a google search, and most of it is related to 2015. Write a new article writing "in 2016 27 transgender were murdered, over an estimation of 1.4M transgender in the USA, for a rate of 19 murders per million, while the 2015 rate was around 50 murders/M, which shows that eterophobic violence is at an all-time high." (note: this is provocative only)... – frarugi87 Jul 13 '17 at 07:54
  • 3
    Personally, I find the claim dubious just based on social attitudes a few thousand years ago... – jpmc26 Jul 13 '17 at 08:44
  • 1
    "Deadliest" must surely include deaths from disease as well as murder. In which case the deadliest year was the peak year for HIV deaths -- the number of murders is tiny by comparison. – Mike Scott Jul 13 '17 at 10:25
  • 2
    I would suspect more than 27 homosexuals were killed in Nazi Germany during WWII. They weren't eliminated like the jews were (unless they were jewish homosexuals) but conditions in concentration camps were bad enough that some people would just die for one reason or another (disease, old age), or get shot for trying to escape or for disobeying. – Pharap Jul 13 '17 at 16:14

4 Answers4

35

The data this claim seems to be based on is severely limited. I applaud their efforts to bring attention to this issue, but they do not appear to be using scientific counting methods, and their data only goes back a few years.

Although not explicitly stated in the claim, the claim appears to be specific to the United States.

This document published by the Human Rights Campaign (HRC) appears to be the source of the 2015 murder numbers in the image. It gives names, pictures and details of 21 trans people murdered in 2015. They do not give details on how they found or verified these names. On page 27 it discusses how the FBI's data collection methods undercount anti-trans violence. The FBI tables for 2015 list 1 trans person murdered.

The HRC report says of the FBI's numbers.

The FBI has taken important steps toward improving data collection for bias-motivated crimes based on gender identity and expression, but the most recent data reported (2013) suggests that local jurisdictions fail to report many of these crimes, including homicides, as bias motivated.

Note: This report was published in 2015, when the FBI numbers I just linked were not available.

Based on the HRC's discussion of the challenges of collecting data, I believe their numbers are also an undercount. I applaud both their effort, and their openness about the weaknesses in their data.

The HRC states that "in 2009, the FBI began tracking bias-motivated crimes based on the victim’s actual or perceived gender identity." So not only are the FBI numbers an undercount, they don't go back very far in time.

This article, published by an advocacy group for LGBTQ issues, reports the number of murdered transgender people in 2016, but does not say where there data comes from. It also lists the names, and a few details for each murder. It repeats the claim, that 2016 was the deadliest year for trans people, but does not attempt to prove it by presenting any historical data.

In summary, I found what appears to be semi-quality data going back a few years that comes from independent organizations. The FBI, which should compile authoritative crime statistics, appears to be severely deficient on this count. The claim that 2016 had more anti-trans murders than any other year ever is not based on historical data.

If the claim were revised to say that 2016 had more trans murders than any year since 2013, that claim is grounded in some evidence. In recent years, trans people have gotten more organized, and gotten more media attention. It is entirely possible that the rate of trans murders is flat, but more of those murders are being recognized for what they truly are. The claim could be true, or it could be a case of reporting bias.

BobTheAverage
  • 11,961
  • 6
  • 43
  • 54
  • great answer. But if you could also address rather it was deadliest for the LGBTQ community, per the first sentence of the claim, I could accept this answer immediately. It's actually probably hard to define how deadly it was for the Queer part of LGBTQ, too hard to define who qualifies as Queer and almost certintly no statistics on death rates for them, but something addressing how deadly it was for the LGB community would suffice. – dsollen Jul 11 '17 at 18:24
  • 13
    @Dsollen The data examined, here and in the claim, says nothing about the LGB or Q communities. I think your claimant was overstepping what little data they had in an effort to broaden the reach of their message. It would be another hour of research at least to find all of that. – BobTheAverage Jul 11 '17 at 18:29
  • 1
    One should also consider historic patterns; even if 2016 was the deadliest year in recent history, how does it compare to treatment of homosexuals/etc. in, say, the 1700s? – JAB Jul 11 '17 at 22:50
  • @Jab One should consider the things they have reliable data to consider. If you can find reliable data, I will consider it. – BobTheAverage Jul 12 '17 at 18:38
  • 1
    Pretty sure that things were a lot worse for LGBTQ in a lot of Europe between the late 30s and mid 40s .... – ivanivan Jul 13 '17 at 02:17
32

In addition to BobTheAverage's asnwer, the claim has other notable problems (hat/tip to commenters on the question for some of them):

  1. The claim tries to imply (not just the cracked.com image, but HRC's document it's based on) that these victims were specifically victimized because of their transgender status. The report is misleadingly called "Addressing anti-transgender violence", for example, implying every victim on the list was targeted due to being TG.

    The data presented does NOT bear this out. Going over each of 21 HRC cases listed for 2015:

    • 2 were "claimed" to be caused by assailant being angry over victim being TG. However, the claim is according to victim's friends, without independent corroboration presented.

    • 2 has extremely weak possible evidence it may have been TG related (there was a spike of TG related crimes in the area at the time and both victims were in same area).

    • 14 had absolutely no indication that they had anything to do with TG status

      At least 5 of them were clear cut domestic violence cases; one of these the HRC paper deliberately mischaracterized as "possible hate crime"). HRC's own analysis 2013-2015 acknowledged "At least 8, or 15 percent, were killed by intimate partners".

    • 3 definitely NOT based on TG status (since the state has hate crime law and the perp wasn't charged under it). 1 of them was a clear cut domestic violence.

    In other words, of 21 in 2015, ZERO were proven to be violence aimed at someone's TG status by HRC or by justice system; and only 4 had at least some possible evidence presented by HRC indicating it may have been due to TG status.

  1. They don't even account for other risk factors than being transgender, even though they honestly mention one:

    At least 18, or 34 percent, were or likely may have been engaged in survival sex work at the time of their deaths.

    As per Wikipedia, "In 2004 the homicide rate for female sex workers in the United States was estimated to be 204 per 100,000", yet "In 2004, there were 5.5 homicides for every 100,000 persons".

  1. Another implication is that somehow TG status makes someone more of a murder target, which doesn't fit the data.

    It is perhaps worth to notice that if the numbers match, are from the US and the estimates on the number of transgender people in the US (1.4 million) are correct, transgender victims are significantly underrepresented in the murder statistics. (comment by Tor-Einar Jarnbjo)

    US homicide rate was 4.88 per 100K (I think the data was 2004) and ~4-5 in 2012 as per the same Wiki page color map.

    Even ignoring the issues listed above; transgender homicide victim rate in US would be 2 in 200k, 2-2.5 less than general population; in 2016 and 1.4 per 100k in 2015.

  1. Leaving all those weaknesses aside, even if the numbers were accurate and statistics not misleading and the conclusions they want to draw actually followed from their statistics (none of which is the case), the numbers are too small to draw meaningful conclusion. As eyeballfrog's comment says:

    As 27 isn't even a statistically significant fluctuation. If we assume the base rate is 20/year (as suggested by the previous three years), 27 in a single year has a p value 0.08.

  2. This is somewhat contradictory to point #3 above, but may also explain the discrepancy (that TG rates quoted seem far below average US rate). Either way it's a problem with the claim's data, though in opposite direction.

    There's no indication that every victim in US is known for sure as to whether they are or aren't transgender (the data in HRC report seems to use facts provided by victim's family/friends, not some official accurate registry - which thankfully does NOT exist - of who is transgender or not).

    Therefore, it's eminently plausible that there are more TG victims not listed in HRC/cracked statistics, simply because there's no knowledge of their gender status. I wouldn't be surprised if the actual numbers are far higher (by the same 2-2.5x factor, at least, that was the discrepancy between HRC data and overall US rate, bringing the rates to parity).

user5341
  • 31,075
  • 8
  • 130
  • 178
  • [Trans Murder Monitoring (TMM) (US data: top of page 3)](http://transrespect.org/wp-content/uploads/2017/03/TvT_TMM_TDoV2017_Tables_EN.pdf). I may ask the author(s) of the report personally how the data is derived from the various sources and interpreted if someone is interested. – klanomath Jul 12 '17 at 16:29
  • @klanomath - wouldn't hurt; but the link's figures for 2015/2016 match closely with the ones already in this post; so I doubt either their sources or their methodology is wildly different (or better). – user5341 Jul 12 '17 at 16:32
  • 2
    So most likely more than the 27 transgender people have been murdered, but neither murderer nor police were / are aware that they are transgender, just according to the murder statistics. – gnasher729 Jul 12 '17 at 18:03
  • 1
    @gnasher729 - to be more precise, it's **likely** (though not guaranteed) more than 27 people **who happened to be transgender** were victims based on overall crime statistics. However, **there's basically no evidence** that many of them were victimized specifically on the basis of them being TG outside of 4 for whom at least some evidence was presented - having said that, in all fairness, there's also no evidence that this wasn't the basis, either in some of the cases where there's just no info at all. – user5341 Jul 13 '17 at 14:10
4

Not for the LGBTQ community as a whole. The LGBTQ community was devastated by the AIDS crisis of the 80s on a scale that's difficult to imagine today.

In 1995 there were 48,371 known AIDS related deaths in the USA. I don't know precisely how many of those were LGBTQ people, but it does seem likely that the deadliest year would be in the early to mid 90s - when many people who caught HIV during the 80s died.

Robyn
  • 173
  • 3
  • 8
    This answer needs references –  Jul 12 '17 at 07:58
  • 5
    While the answer addresses the exact technical wording of the claim's **title**, the actual claim text makes it clear the context is specific to homicides and **not** medical causes. – user5341 Jul 12 '17 at 14:10
  • Agreed on the needs for references. [Here's a reference](https://web.archive.org/web/20040213152152/http://www.fda.gov/fdac/features/1999/499_aids.html) for 41,699 deaths in 1995 in the U.S. The source is the FDA citing the National Center for Health Statistics, though it's an old article, so perhaps newer data supports the higher number. – reirab Jul 12 '17 at 16:36
  • 1
    [Welcome to Skeptics!](http://meta.skeptics.stackexchange.com/questions/1505/welcome-to-new-users) Please [provide some references](http://meta.skeptics.stackexchange.com/q/5) to support your claims. – Oddthinking Jul 13 '17 at 04:18
1

As far as deadly attacks go, the Upstairs Lounge arson attack killed more people than that in a single incident. That was absolutely a hate crime against the gay community.

I would agree with your second statement - whilst not the deadliest year for the LGBT* community, it appears to be the worst year for the trans* community. However this is based on extremely limited data, so it will certainly have the same kind of reporting bias that you find with all crimes of gender-related/sexuality-related/sexual violence.

Graham
  • 1,592
  • 12
  • 13
  • @JanDoggen: It's not a great answer, but it does completely refute the claims made by the image. (The image claims that 2013 and so on were "previous record highs", not that they only started counting in 2013.) The fact that an answer this simple is a counter-example to everything in the image shows how wrong it is (but mostly because it says LGBTQ instead of just T). – Peter Cordes Jul 13 '17 at 13:47