3 ms·
Here are the numbers: In 2019, the most recent year for which the FBI has data, the total US homicide rate was 4.3 deaths per 100,000 people [0] (possible unde
by neartheplain 5y ago
Here are the numbers:
In 2019, the most recent year for which the FBI has data, the total US homicide rate was 4.3 deaths per 100,000 people [0] (possible underestimate; incident reporting is voluntary, and 25% of police departments don't submit expanded homicide data). Human Rights Campaign recorded the deaths of 25 transgender and gender non-conforming people in 2019 [1]. There are an estimated 1 million people who identify as transgender in the US [2], and an estimated 1.2 million people who identify as nonbinary or gender non-conforming [3]. That produces a US trans and nonbinary homicide rate for 2019 of 1.1 per 100,000.
Of all homicides reported by HRC with transgender or gender non-confirming victims in 2019, just one was determined to involve a clear anti-LGBT motive. HRC's data may be an underestimate. However, it is consistent with their estimated homicide rates and rate of hate crimes reported in other years.
Any murder is a tragedy, hate crimes all the moreso. However, a single-digit number of annual anti-trans and nonbinary murders is thankfully not an epidemic. The overall homicide rate for trans people seems much lower than the national average.
Notably this data excludes assaults, harassment, stalking, or any other attacks which don't result in fatalities.
[0] https://crime-data-explorer.app.cloud.gov/pages/explorer/crime/shr https://crime-data-explorer.app.cloud.gov/pages/explorer/cri...
[1] https://www.hrc.org/resources/a-national-epidemic-fatal-anti-trans-violence-in-the-united-states-in-2019 https://www.hrc.org/resources/a-national-epidemic-fatal-anti...
[2] https://www.cdc.gov/hiv/group/gender/transgender/index.html https://www.cdc.gov/hiv/group/gender/transgender/index.html
[3] https://williamsinstitute.law.ucla.edu/publications/nonbinary-lgbtq-adults-us/ https://williamsinstitute.law.ucla.edu/publications/nonbinar...
- freemint 5y agoDon't role your own aggregate statistics when the thing you try to derive by mashing together things created for different purposes is measured or estimated by people. I don't role my own encryption for public use either.
- 9wzYQbTYsAIc 5y agoExcellent analogy, thank you.
- AnimalMuppet 5y agoYou're going to criticize and be condescending, but not supply better data? That's quite unhelpful...
- raxxorrax 5y agoCome on, that is just false. Don't believe a statistic you haven't doctored yourself is a common saying. But data evaluation is certainly something you should do yourself, because blindly trusting anyone else would be naive.
- neartheplain 5y agoIf higher-quality statistics are available, I'd certainly be curious to see them.