I stuck this in the Stats thread yesterday.
ANALYSIS OF THE UK STATISTICS FOR HOMICIDES OF TRANSGENDER PEOPLE IN THE UK
Transgender Homicides in Britain, 2000-2025: Victims and Perpetrators
by Michael Biggs & Ace North
Feb 2026
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6182901
Abstract
Transgender people are often portrayed as especially vulnerable to violence, but estimating victimization rates is difficult because reliable population denominators are lacking. This paper proposes an alternative approach, comparing the ratio of transgender homicide victims to perpetrators. It analyzes all homicides involving transgender people in Britain from 2000 to 2025. Victims were outnumbered by perpetrators, even excluding those who declared a transgender identity after imprisonment. Almost all cases involved natal males identifying as transwomen. The victim-perpetrator ratio among these individuals closely resembles that for males overall and differs markedly from that for females. BBC News published more than four times as many articles on transgender victims as on perpetrators, contributing to perceptions of exceptional vulnerability.
Just highlighting:
The victim-perpetrator ratio among these individuals closely resembles that for males overall and differs markedly from that for females. BBC News published more than four times as many articles on transgender victims as on perpetrators, contributing to perceptions of exceptional vulnerability.
However, here is the conclusion from the paper.
Conclusion
It should be acknowledged that the analysis depends on the enumeration of transgender victims and perpetrators in Britain from 2000 to 2025 being complete or nearly so. The possibility that some individuals are not counted—either because a murder went undetected or because the individual’s transgender status was not known—cannot be excluded. Nevertheless, the initial lists were compiled by activists who were predisposed to maximize the number of victims and perpetrators respectively, and these biases offset each other. We have verified the initial lists against news reports, and in every case found these to be accurate. Furthermore, we applied uniform criteria to victims and perpetrators, hence the exclusion of cross-dressers from the list of perpetrators.
We have introduced an alternative metric for comparing violence: the victim/perpetrator ratio. Using this ratio, the paper is the first to compare the numbers of transgender victims and of perpetrators—and to compare the ratio in media reports. There are three main findings. First, more transgender people committed homicide than were victims of homicide in Britain in the 21st century. The victim/perpetrator ratio was 0.7 excluding post-imprisonment transitioners.
Without reliable figures on the transgender population, it is unknown whether transgender people were at greater risk of homicide than the population as a whole. If they were at greater risk than the population, however, then we would also conclude—given the victim/ perpetrator ratio was less than one—that they were more likely to commit homicide. If the extent of fatal violence suffered by transgender people in Britain is considered to be an epidemic, then the same epithet applies to the fatal violence inflicted by transgender people.
The second finding is that transwomen followed the male rather than female pattern of homicide. The victim/perpetrator ratio for natal males identifying as transgender was 0.8, and this approximates the ratio for all males, 0.7. It is much smaller than the ratio for all females, 2.9; the difference is statistically significant despite the small numbers. This finding has obvious implications for policies in the sphere of criminal justice, for example in the placement of transwomen in women’s prisons.
The third finding is that the BBC published many more news articles mentioning transgender victims than perpetrators. The victim/perpetrator ratio in reports that mentioned the individual’s transgender identity was 4.5. The extraordinary coverage of one horrific killing accounts for some of this disparity, but not all. Unbalanced media coverage creates an exaggerated impression of transgender people as victims of homicide. The lack of balance has various causes, aside from editorial choices. One is the legal system: it discourages the disclosure of a suspect’s transgender status, but encourages the disclosure of a victim’s status with the category of transphobic hate crime (introduced in Scotland in 2009 and England and Wales in 2012). Another cause is the response of advocacy organizations.
Naturally organizations in the LGBT movement will publicize victims from the communities they represent, exemplified by the annual Trans Day of Remembrance. In recent years, the gender- critical movement has called attention to transgender perpetrators of violence, but this does not appear to have impacted the BBC’s reporting (though it has influenced right-wing media like the Daily Telegraph).
Can these findings be generalized beyond Britain? In the United States, the composition of transgender victims is quite different, with the majority being black. In addition, transgender people in America seem to experience a higher risk of murder—relative to the population— than in Britain. Therefore we might expect the victim/perpetrator ratio to be higher in the United States, though that is a question for future research. Can these findings be generalized to lesser forms of violence? The finding that transwomen are closer to the male than the female pattern of homicide echoes the result from the Swedish longitudinal study of violent crime (Dhejne et al. 2011). Unfortunately almost all studies of violence focus exclusively on transgender people as victims. The nearest is a survey of Finnish school students that asked respondents whether they bullied others as well as whether they experienced bullying (Heino, Ellonen, and Kaltiala 2021).
Transgender students reported being bullied more than their peers did, but they also admitting bullying others more. The study’s data enable victim/perpetrator ratios to be calculated. For all students in total, the ratio was 2.0; for transgender students, it was 1.4. Thus transgender students were relatively more likely to bully others (or at least to report it).
There is an important lesson here for academic research. There are many more studies of transgender people as victims of violence than as perpetrators of violence. Perpetrators are discussed, moreover, primarily as victims of the prison system. No individual study can be faulted for focusing on a single aspect of a phenomenon, of course, but in aggregate they can nevertheless provide a misleading portrayal of the phenomenon as a whole. We hope the victim/ perpetrator ratio will provide a useful metric for empirical research, while also serving as a reminder of potential epistemic biases in social science (Burt 2026).