Sex in the Census: The UK is losing robust data on sex and the framing of the sex question in the 2021 census will make the situation worse.
The next census in England and Wales will take place on 21 March 2021. For the past few years, both expert data users and feminist groups have been taking a close interest in the sex question and how this is framed. While the question itself will simply ask ‘What is your sex?’ with male/female response options, the issue at stake here is how sex is defined.
With only a matter of weeks to go before Census day, the ONS confirmed on 12th February that the guidance accompanying the long-standing sex question will carry the following wording:
“If you are considering how to answer, use the sex recorded on one of your legal documents such as a birth certificate, Gender Recognition Certificate, or passport.”
While this is a tighter definition than the guidance used in 2011, the wording is ambiguous. The guidance provides three examples, only two of which are consistent with the EHRC definition of ‘legal sex’ (birth certificate or a GRC). In the case of passports, a person’s sex marker can be changed without acquisition of a GRC. The words ‘such as’ suggest that there are other ‘legal documents’ that could form the basis for answering the sex question, all of which can be updated by self-declaration.
The ONS formally abandoned the principle of gathering data on biological sex in 2011 and, against the strong advice of experts in quantitative research and analysis, have not returned to it for the 2021 census.
Following a public consultation on GRA reform the government decided legal sex should not be self-identified. Shockingly, the ONS has done the opposite and decided it can redefine sex to include self-declared gender identity. This is sex self-ID through the back door.
After obtaining expert legal advice we believe the ONS has acted unlawfully. The ONS has now been served with a pre-action letter inviting them to back down and withdraw the offending elements of the guidance accompanying the sex question, failing which we will be applying for a expedition order in the high court for an emergency judicial review. ONS ignored the experts and listened to Stonewall. But they won’t be able to ignore the Court.
News of our legal challenge was reported in the Times on Saturday 20th February here
Data on sex is important so it needs to be accurate
Sex is a fundamental demographic variable, essential for projections regarding fertility and life expectancy. We need accurate data, disaggregated by sex, in order to understand differences in the lives of women and men, and in order to tackle sexism. Sex matters from the start of life, as illustrated by international differences in the sex ratio at birth due to son preference. Sex is a significant factor in almost every dimension of social life: education, the labour market, political attitudes and behaviour, religion, crime, physical health, mental health, cultural tastes and consumption – the list goes on. It is difficult to think of an area of life where sex is not an important dimension for analysis. A lack of sex-disaggregated data often leads to the needs of women and girls being ignored.
ONS casually dismiss our concerns, arguing that the trans population is so small that the impact of allowing the sex question to be answered according to how someone self-identifies will be negligible. However, we currently have no robust data on the trans community as a whole or within sub-groups, and, crucially, it is impossible to predict how this may change over time. It is unlikely that the trans population will be evenly distributed, for example by age, sex and geography. This means small group analysis can and will be affected. Here are some examples of how small errors can have a big impact.
Crime
Male people are responsible for most crimes, particularly violent and sexual crimes. This means that even if a small number of males are misclassified as female it can have large distorting effects on female statistics. For example, HMPPS prison statistics show ~13,000 male sex offenders and 125 female sex offenders are currently in prison in England and Wales. 76 of the male sex offenders identify as female. While misclassification of 76 males would have a negligible impact on the male total it has a huge impact on the female figures by almost doubling them.
Young people
The Tavistock and Portman NHS Trust claims that between 1.2% and 2.7% of children and young people are ‘gender-diverse’. The trans population is growing rapidly, particularly among young females, and the reasons for this are not well understood, and require investigation. This means we need good data on which young people are identifying as transgender and why. A sex difference has been identified so we need birth sex data for these young people to monitor and analysis the female bias.
Employment
Some careers are predominantly male with very low female representation. We need to know when, where and why. A small number of males misclassified as female won’t impact the male totals but could make a big difference to the female numbers and obscure a problem with poor accessibility for women in some careers. This type of equality monitoring is used to identify when and how women are discriminated against. If we don’t identify the problem it means policy solutions won’t follow.
Same applies to gender pay gap data. Companies currently report pay gap data based on the self-declared sex of employees. Self-declared non-binary people are excluded from the data set. It is likely that the highest-paid transgender people will be late-transitioning males in established careers whereas the lowest paid transgender people will be young females who identify as male or non-binary. This differential means that in some careers misclassification of highly-paid males could distort and obscure the sex pay gap.
Some argue that this employment discrimination is based on perceived sex or gender identity rather than birth sex. However, without the data on all of these characteristics we can’t do the studies to determine what drives the discrimination.
Studying the transgender population
The 2021 Census will capture information on the trans population but the usefulness of this ‘gender identity’ question is limited because it is VOLUNTARY. This means we can’t be sure the data will capture the whole trans population. In any case, the open text nature of the answer means we won’t necessarily get good data on the type of trans person (in particular, whether they male at birth or female at birth). It is also only open to over-16s so there will be no data captured at all on young people who identify as transgender. Researchers need other ways to identify these important groups in society. This can only be done by identifying a mismatch in records of birth sex and records of gender identity. This approach has been used before by Swedish academics using birth sex data from the Swedish Census.
It’s odd that groups who claim to advocate for the transgender community are arguing against the collection of high quality data. There is a paucity of evidence-based research on trans outcomes like surgery regret, suicide rates and health. A full and rich data set on trans people will help any inequalities to be identified and policy makers to make trans lives better. Instead Stonewall say it’s not enough that trans people have their own gender identity question. They think transgender people should be allowed to answer BOTH questions according to gender identity. They say it’s simply too offensive and hurtful to ask a trans person a factual question about the sex they were born.
Gender Identity lobbyists claim that gender identity trumps sex, and therefore we don’t need data on sex. We say, on the contrary, we need accurate data on both sex and gender identity in order to investigate the way both factors influence people’s lives. It’s possible that such analysis will reveal that sex is unimportant in certain contexts – only then should we stop collecting data on sex in those contexts
Studying the LGB community
The protected characteristic of sexual orientation is referenced in relation to sex. Sexual orientation is refined in law as meaning a person’s sexual orientation towards— persons of the same sex, persons of the opposite sex, or persons of either sex. For the first time in the census there will be a voluntary question asking about sexual orientation. The question lumps all homosexual people together in to one group of ‘gay or lesbian’. But we know forms of discrimination against lesbians will be different from forms of discrimination against gay men. So we need to be able to split the data by sex not gender identity. A significant proportion of transwomen identify as lesbians despite their sexual orientation technically being heterosexual males. So losing data on birth sex risks obscuring the data needed to identify discrimination and inequality suffered by lesbians, i.e. same-sex attracted females.
This is about so much more than the Census
ONS sets the gold standard for data collection. Its approach to the sex question will be followed by all other data collectors. Without a clear signal from ONS that it is both lawful and important to be able to collect birth sex data, others will inevitably take a cautious approach and allow self-declared gender even if they really want and need birth sex data.
This is madness while we are in the midst of the Covid pandemic. We know sex matters in all Covid-related research – not just both medical but social outcomes too. We already have evidence that men are at higher risk of death from Covid, while women are more likely to be impacted by the social and economic of lockdowns. Good data on sex matters now more than ever
Medical researchers tell us they want to collect data on biological sex data but they are concerned they can’t. The Telegraph covered this story here.
“I think there is a concern among scientists that there are acceptable ways you can ask these questions. The ONS will make it more difficult to ask about sex registered at birth.”
“There is a trepidation among researchers that we don’t want to upset people, don’t want to do something wrong. We’re not experts in law. We’ll tend to follow what’s done by the main government agencies.”
Even the Zoe Covid symptom tracker is not immune from ideological interference. When the Zoe App launched in March 2020 it asked “your sex at birth?” – Options Male or Female Sex was treated in the same way as all the other mandatory core variables such as age, height, weight, location. Within a month it had changed to “What sex were you assigned at birth” with additional options ‘intersex’ and ‘prefer not to say’. (Of course, intersex is a blanket term for the congenital differences known as VSDs, not a third sex.) The question “what sex were you born” is fast becoming taboo even in medical applications.
Enough is enough
We have to stop this erosion of sex data. If we don’t have good data on sex we can’t monitor inequalities due to sex. If we can’t prove inequality exists we can’t remedy it. Sex matters.
This is why we have launched our emergency legal challenge against the ONS to correct their definition of sex in the Census. ONS ignored the experts and listened to Stonewall. But they won’t be able to ignore the Court.
After obtaining expert legal advice we believe the ONS has acted unlawfully. The ONS will be invited to back down and withdraw the offending elements of the guidance accompanying the sex question, failing which we will be applying for a expedition order in the high court for an emergency judicial review.
Given the unusual speed at which this case has to take place we will need to prepare evidence on an at risk basis. We need to seek a trial before the Census on 21st March.
Preliminary legal opinion was paid for from our existing funds. We launched our crowdfunder on 17 February, and raised an additional £101,000 in 2 weeks to cover legal costs. That shows that people know this is important. We are grateful for the support, and working hard to make it count.
DONATE NOW
It cannot be wrong to ask people what sex they were born. We need to know it. We need be able to ask it. It really is as simple as that.
Our Chief Statistician should be prioritising facts not feelings and the public needs to trust that our national statistics are free from interference by ideological pressure groups.