This new 232-page government-commissioned study, called “the Sullivan Report” after its leader, opens this way (click on the screenshot below to see the whole thing).
This independent review was commissioned in February 2024 by the Secretary of State for Science, Innovation and Technology.
The aims of the review are as follows:
i. Identifying obstacles to accurate data collection and research on sex and on gender identity in public bodies and in the research system
ii. Setting out good practice guidance for how to collect data on sex and gender identity
All public bodies, as defined by the Cabinet Office, are in scope of the review. The review also considers research institutions and organisations from outside the public sector, where relevant to the aims of the review.
The review is UK wide, respecting the devolved nature of areas of responsibility within the research and development landscape and the collection of relevant areas of data and statistics. This report concerns data and statistics. A further report will examine barriers to research.
The review is led by Professor Alice Sullivan, University College London, assisted by policy analysts Murray Blackburn Mackenzie, and Dr Kathryn Webb, University of Oxford.
I’ll confess that I’ve only skimmed it, and am relying on the summary given by the Times of London. (If you subscribe click the headline below, or find the piece archived here.) It does, however, appear to draw on many studies: over 500 of them.
The upshot is that biological sex (male or female, with a very, very tiny proportion of true intersexes) is not consistently identified in government documents, even though some of them mandate “gender identification”. While having both is usually okay by me, the absence of biological sex has, so the report (and the Times article) argue, caused the loss of useful data as well as harm to people. On pp. 5-6 the report makes ten recommendations, and here are just three:
2. Data on sex should be collected by default in all research and data collection commissioned by government and quasi-governmental organisations. By default, both sexes should be included in all research, including clinical trials, and sex should be considered as a factor in analysis and reporting. As a general rule (with some obvious exceptions), a 50/50 sex ratio is desirable in studies
3. The default target of any sex question should be sex (in other words, biological sex, natal sex, sex at birth). Questions which combine sex with gender identity, including gender identity as recognised by a Gender Recognition Certificate (GRC) have a mixed target. Sex as a biological category is constant across time and across jurisdictions, whereas the concept of ’legal sex’ subject to a GRC may be subject to change in the future and varies across jurisdictions. Using natal sex future-proofs data collection against any such change, ensuring consistency.
6. The word ‘gender’ should be avoided in question wording, as it has multiple distinct meanings, including: a synonym for sex; social structures and stereotypes associated with sex; and gender identity. If a question targeting gender identity is worded as a question on gender, this is likely to mislead many respondents. Questions on sex have also often been labelled as ‘gender’. Change in the use of the term ‘gender’ means that it is important that questions on sex are labelled explicitly as such.
Plus given that there are over a hundred genders on lists I’ve seen, what does it add to a survey to add “gender identity”? I don’t really object to it, but really, what does gender add for official forms if biological sex is already specified?
Here are some ways the data are erratic (I’m not sure that any study asked for “natal sex” or “biological sex” sex rather than just “sex,” which these days is ambiguous. (That information might be in the long report.) Quotes from the Times:
The study, led by Professor Alice Sullivan from University College London, investigated all public bodies and found “the meaning of sex is no longer stable in administrative or major survey data”.
Sullivan’s review found inconsistencies in the way sex and gender were recorded and conflated. Some official surveys were found to remove sex altogether and only collected information on gender identity.
This included a Royal Navy sexual harassment survey, which asked how respondents identified rather than asking for their sex “despite its obvious relevance to the subject matter”.
In another case, a children’s camping programme raised safeguarding concerns through collecting data on gender identity, with male, female and “other” response options.
. . .The Office for National Statistics (ONS) also previously caused confusion by proposing respondents to the 2021 England and Wales census could answer the sex question in terms of their subjective gender identity, rather than their sex. It was later forced to change this through judicial review.
This ambiguity also occurs in the National Health Service (NHS):
The review found that across the NHS “gender identity is consistently prioritised over or replaces sex”. She said that records that traditionally represented biological sex were “unreliable and can be altered on request by the patient” and that there had been a “gradual shift away from recording and analysing sex in NHS datasets”.
And the report describes some of the palpable harms deriving from this inconsistency. Bold headings are mine:
Medical Harm:
This meant there were “clear clinical risks”, such as patients not being called up for cervical smear tests or prostate exams, or the misinterpretation of lab results. Sullivan said: “This has potentially fatal consequences for trans people.”
In one case a paediatrician said that a child had been brought up in the preferred gender of the mother, which was different to their birth-assigned gender. “She [the mother] had gone to the GP and requested a change of gender/NHS number when the baby was a few weeks old and the GP had complied. Children’s social care did not perceive this as a child protection issue,” the doctor reported.
It’s even worse because in the UK (and many states in the U.S.) people can change their official records to a non-natal sex, which can also be harmful:
Sullivan’s review said the patient’s ability to change their records “puts transgender individuals at a particular disadvantage and as such is potentially discriminatory”. She said that in some cases samples such as blood tests could be rejected by laboratories or sex-specific cancer referrals could be missed.
Legal Harm:
In the justice system Sullivan found that definitions of sex and gender were “highly inconsistent”.
Sex can be recorded differently across the prison service, while many police forces record sex as the gender given in a gender recognition certificate.
The review said it meant that data across police forces was not reliable, particularly in patterns of female offending and “the classification of a small number of males within the female category may result in artificial significant increase in female offending rates”.
She said: “Many police forces record crimes by male suspects as though they were committed by women at the request of the perpetrator or based on how a person ‘presents’.”
Guidance notes for officers on the Police National Computer (PNC) state that it is “quite possible” that an arrested person who has acquired a gender recognition certificate and not informed the police “could be released or otherwise dealt with before any link to their previous offending history is known (through confirmation by fingerprints)”. The review found that this was also likely to be true of those who self-declared a different sex and name.
Distortion of educational data:
In schools and universities, the review found that children and young adults were more likely to report being transgender but that without biological sex being recorded data that showed the different life outcomes, including earnings, could not be relied upon.
The review said: “Significant sex-based effects could either be missed, because they are wrongly assumed to be due to changing practices in self-identification, or conversely wrongly inferred, as the data has become impossible to read reliably for sex-based effects.”
Pay differentials:
The lack of reliable data was also found to have an impact on pay gap reporting.
UK public authorities and private sector employers with headcounts of 250 or more have been required by law to report annually on their gender pay gap — which records how much less women earn than men.
However, those who identify as non-binary are excluded from the data and gender identity is recorded rather than sex.
Sullivan’s recommendations. These are given in extenso following p. 6 in the report, but Sullivan is opposed to reporting gender identity (and I think this means non-natal sex for transsexuals):
It has been argued that recording biological sex alongside gender identity could interfere with a person’s human rights.
However, Sullivan found — and presented legal advice to back it up — that recording sex as gender identity was in itself likely to breach UK data laws and potentially human rights laws.
She said: “There are things that statistics cannot do. Statistics are not a means of personal self-expression. They can neither validate nor invalidate individual identities, and they cannot see into people’s souls.”
Remember that this is a report commissioned by the UK government, so take it up with them if you don’t like what it says. However, I do agree that on birth certificates and official documents, like driver’s licenses, what should appear is not gender identity but biological sex. If you want to add “gender self-assignment” to things like medical records, I don’t have a huge objection to that so long as biological sex is the primary thing identified, because I see no medical advantage, and palpable disadvantage, to recording only gender identity.
I’m still waiting to see how advocates of the “multidimensional, multivariate” concept of sex tell us how to DEFINE AND DETERMINE people’s sex. For the Sullivan Report, at least, trans women are not women, nor are trans men men—not in the official sense.
h/t: Richard
It seems to me that biological sex is the primary variable of interest in most cases, as medical research and treatments can be different for each sex. The difference can be life and death. That’s obvious (well, at least to me).
Gender is also OK as a separate variable, as this information may provide insight into the social aspects of the data sample. Gender as a term, however, might be less reliable than sex, since people often conflate sex and gender and because gender is a continuous variable. One option might be to ask people to check a box (male or female) for sex and to write in their answer for gender.
A problem arises if the questionnaire asks only about one of these variables. Since people conflate sex and gender, it would be difficult to know whether people are responding with their biological status or their social standing. The data are sullied by this ambiguity and lose their value.
The invention of gender as something distinct from sex is a major problem in itself. Gender as it is often used today is no more real and measurable than soul or spirit. It is whatever the user wishes it to be and is therefore meaningless for medical purposes.
In my opinion, birth certificates should be based on facts and not be changed unless genetic testing (for example, 5-ARD cases) shows that the birth certificate was/is wrong. However, the case of driver’s license is not so clear (in my opinion). If the driver’s license has a photo, then the sexual designation should be correct. However, in the absence of a photo I would be inclined to go with gender identity. I would apply the same rules to passports.
But if gender is on the driver’s licence it should be marked as such, not as sex. Anything else would be falsification of an official document.
I suggest, ‘G F’ and ‘G M’ should be used. I am presuming no photo. If a photo is provided, then ‘F’ and ‘M’ should be used (based on sex, not gender identity).
My goal is to make it easy for police/CBP agents to do their job in all cases
The problem is a lot of MTF think their photos actually reflect their ability to pass as the opposite sex. They don’t.
How is their gender identity with the lack of a photo going to reflect the truth of the actual sex. Down the rabbit hole of police are looking for a 6.5 foot bearded lady with blue eyes accused of sexually assaulting a pensioner in the Dorset area at 4 pm this Thursday. The data should be of real value and not pretend ideas.
Social and cultural roles and behaviors that are accepted and expected for male and female but which are not a direct consequence of biology are very real and need a name. People once used “gender”. Now that gender has been used to help confuse the issue of biological sex vs personal identity, the word has lost any clarity it may once have had. It makes sense for people to state their sex regarding the issues listed. It is helpful too for someone to state their “identity” if it is not the same as their sex. However, since many trans people and supporters do not think there is a difference between their perceived sense of identity and their biological sex, (biological sex being a phrase they hate), there is no certainty that such forms will be filled out accurately.
I appreciate the findings here that the loss of statistical information pertaining to sex has negative consequences, including for trans people requiring health care. I also think it is quite wrong for men who commit rape and violence to claim they are women, making it seem that percentages of females committing rape and domestic violence are on the rise. Middle aged men who went into traditionally male fields and attained distinction, then changed their sex identity, can claim to be the “first woman” to achieve such distinction. An abilty to legally change sex has negative consequences.
The fact that there are ‘expected and accepted’ behaviours/roles for M & F is the problem. ‘Feminine’ behaviour does not make a M a F and vice versa. We are our sex. Everything else is personality. ‘Gender Identity’ is farcical labelling.
No doubt. When having discussions on this topic I used to ask Trans Rights Advocates what they meant by “gender” and usually received some variation of “this-is-so-basic-I-can’t-believe-you-asked-this Look It Up!” Tedious, because of course I have pages of collected definitions all falling roughly into these three categories — and then there are vague and often contradictory definitions within those. Which one do you mean? And will you please stick with it and not jump to one of the others in the middle of the discussion?
Who would have guessed that the general public would get confused about “gender” when answering survey questions?
Gender Critical feminists in the UK have been warning about the problems that recording male crimes as female ones will have for at least a decade, that I’ve seen. Who would have guessed that they’d be right? Besides the people who are paying attention and can make rational extrapolations into the future, I mean.
As for recording gender identity, it might be significant medically because it indicates that an individual might be taking drugs or hormones related to efforts to eradicate sex characteristics. That could be important for diagnosis or treatment. It might also help divert people who are prone to life-saving duplicity from taking the opportunity to deny their sex by deliberately marking the wrong one, for then they can feel seen and accommodated elsewhere on the form.
The hilarious thing is that the “This is so basic, look it up” crowd then tie themselves completely into knots if they actually have to explain their “oh so basic” concept.
Put three of them in a room together and their definitions will cancel each other out. 🤪
The BBC covered the report with an interview with its author, Professor Alice Sullivan – and for once did a pretty decent job. (The link to the 5-minute segment is here, but the Beeb is changing how people outside the UK can listen so I can’t guarantee that it will work: https://x.com/SexMattersOrg/status/1902631434277257379 )
Now the BBC just needs to acknowledge its own role in undermining the accurate reporting of sex! (Earlier this week a BBC presenter apologised for calling Eddie Izzard a man. The presenter who interviewed Alice Sullivan about her report was found guilty of doing something similar after an investigation last year.)
Edited to add: The rot is so deep that it took a judicial review to insist that the 2021 census question about sex could only be answered by giving the respondent’s sex at birth. This only applied to England and Wales and I believe that in Scotland the data was allowed to be corrupted.
I very much appreciate our Host expending time and energy examining and holding forth on, and making us aware of, such matters.
🎯 a basic common sense issue and that is precisely what is lacking in the trans politics.
RE the lack of basic common sense:
French gynaecologist suspended for refusing to treat trans woman. The Times (of London, England), Jan 31, 2025
Dr Victor Acharian, who sparked a media storm after telling a patient he only treated ‘real women’, has been barred from practising for a month
https://archive.ph/MvH1J
Excerpts:
Curing Trans Delusion: “Gynecologists Are Trained To Treat WOMEN’s Genitals.” The European Conservative, Feb 18, 2025 (freely accessible on web)
A French gynecologist has been suspended from practicing medicine for one month and penalized with an additional five months of probation by the French Medical Council for declining to treat a 26-year-old man who identifies as a woman.
I love how referring a patient to a doctor who might be more competent is now not only “violent” but “hyper-violent”.
In response to Prof. Sullivan’s report, the Secretary of State for Health, Wes Streeting, has already banned the NHS in England from changing the sex of anyone under the age of 18 in its records. He has also committed the UK government to acting on the broader findings of the Sullivan Report.
Biological sex is the key variable in most medical and scientific contexts since it directly impacts research, treatments, and health outcomes—sometimes with life-or-death consequences. Gender can still be useful as a separate category for understanding social dynamics, but it is a less reliable metric because it is often conflated with sex and is a continuous, self-reported variable. A more precise approach would be to have respondents select their biological sex (male or female) while allowing them to write in their gender identity separately. The real problem arises when only one of these variables is collected, as it creates ambiguity in the data—are respondents indicating their biological reality or their social identity? Without clear distinctions, the data become unreliable and lose their scientific value.
If biological sex is as important as you say, which it is, then why do you suggest that respondents SELECT their biological sex? You don’t choose it, you get it at conception.
You’re absolutely right in pointing out that biological sex is determined at conception and is an immutable characteristic. The reason for asking respondents to select their biological sex in certain contexts is rooted in the scientific and medical need for accurate data to understand and address specific health conditions and outcomes. Biological sex—male or female—plays a critical role in areas such as disease prevalence, response to treatments, and overall health outcomes. These differences are not just theoretical; they can have real, life-or-death consequences when it comes to diagnosis, treatment plans, and medical research.
While gender identity is an important social and psychological factor, it is a separate and more fluid variable that doesn’t always align with biological sex. Gender, being a self-reported and subjective metric, can vary across cultures, communities, and individuals, and is less consistent in its scientific utility. That’s why it’s important to distinguish between the two variables when collecting data. If respondents were only asked to select one variable—either sex or gender—it could lead to ambiguity. For example, is a respondent reporting their biological sex or their gender identity? This lack of clarity in the data could compromise its usefulness for research, public health analysis, and medical decision-making.
A more precise approach is to have respondents identify their biological sex (male or female) and then provide an opportunity for them to report their gender identity separately. This would allow for both biological and social factors to be accounted for, while preserving the scientific integrity of the data. The real issue arises when only one of these factors is collected, leading to confusion and potentially unreliable conclusions. Clear distinctions between biological sex and gender are essential for ensuring that the data collected serves its intended purpose and provides valuable insights in medical, scientific, and social contexts.