Gender recognition reform and data collection

At the next evidence session on the Gender Recognition Reform (Scotland) Bill (7 June) the Scottish Parliament’s Equalities, Human Rights and Civil Justice (EHRCJ) Committee will take evidence from witnesses on data and prisons. In relation to data, members will only hear from two witnesses, both of whom support replacing data on sex with data on self-declared gender identity, and neither of whom submitted written evidence.
This blog looks at current data collection practices in Scotland and argues that decision-making by Committee members is likely to influence the longer-term direction of travel: either bolstering the loss of robust data on sex or helping to reverse what is a damaging and ideologically driven trend.
The loss of sex in data collection
The replacement of sex with self-declared gender identity is already widely embedded in Scottish data collection practices, to the clear detriment of women and girls, policymaking, and research. It means that Scotland lacks the necessary information to monitor discrimination and disadvantage experienced by women on the basis of sex.
Scotland’s 2022 census proceeded with a binary “sex” question, but accompanied this with guidance which framed it as a question about self-declared gender identity, against expert academic advice. This means that the definition of this key demographic variable – the item of data most used from census outputs – will be unstable, conflating two different demographic concepts to an unknowable degree. These data will not be directly comparable to data on sex collected via the other UK censuses. Nor will the data be suitable for equality monitoring purposes, as it will not match ‘sex’ in line with its meaning under the Equality Act 2010. Remarkably, the Scottish Government went to court, twice, to defend its ability to do this. At the Appeal hearing of Fair Play For Women vs the Registrar General and the Scottish Ministers in February 2022, counsel for the Scottish Government made the extraordinary claim:
“It may once have been thought that sex at birth is immutable. It is no longer so”.
Douglas Ross QC, Scottish Government counsel, 23 February 2022
In September 2021 the Scottish Government published guidance for public bodies that advised against collecting data on biological sex except in very limited circumstances. The guidance recommends a sex question that defines sex in terms of biological, legal and self-identity ‘aspects’, conflating different concepts in a single variable. As we have previously documented, the definition was changed to include subjective self-identification at very a late stage, to align with the census, and counter to the Chief Statistician’s previously stated preference for an objective measure of sex.
The casual attitude of the Scottish Government to collecting accurate data on sex should be seen in the context of wider criticisms of its attitude towards data collection, in specific public services and most recently with the census more generally.
Losing count
The replacement of sex with self-declared gender identity in data collection means that many Scottish public bodies will be unable to meet the requirements of the Public Sector Equality Duty (PSED) under the Equality Act 2010.
The PSED relates to the nine protected characteristics. In February 2022, the judgment in For Women Scotland v the Lord Advocate and the Scottish Ministers ruled that, under the Equality Act 2010, the protected characteristic of ‘sex’ refers to biological males and females, and is separate to gender reassignment. This definition of sex was reaffirmed in the Fair Play For Women Appeal case, where the judges argued that a biological definition of sex may be necessary ‘in prescribed circumstances involving status, proof of identity or other important rights’.
Bodies subject to the more detailed Specific Duties (for example, the National Records of Scotland, NHS Boards, Police Scotland) are required to: report on mainstreaming the equality duty; publish equality outcomes and report progress; assess and review policies and practices; gather and use employee information; publish gender pay gap information; publish statements on equal pay; consider award criteria and conditions in relation to public procurement; and publish required information in an accessible manner.
It also means that it will not be possible to evaluate the impacts of gender recognition reform on women, which would require separate data on birth sex and gender identity. For example, any incidents of abuse will be masked by police recording practices that conflate sex and gender identity in a single category. Victimisation data on transgender identity is however collected for the purposes of hate crime legislation: we think MSPs should be worried about this imbalance.
Where does the Bill fit in?
The Gender Recognition Reform (Scotland) Bill as currently drafted is likely to cement the loss of data on biological sex in Scotland, in two ways. First, as currently drafted, it signals to public bodies that the Scottish Parliament endorses the idea that sex as a biological category is secondary to self-declared gender identity. The power of that message should not be under-estimated: it is one reason why advocates of self-declaration are so keen to see the law changed. This general message will then be reinforced by the specific legal privacy protections that Gender Recognition Certificate (GRC) holders have (under section 22 of the Gender Recognition Act 2004), which have an inhibiting effect on public bodies.
Sex has been overwritten with gender identity explicitly in anticipation of a legal change in multiple settings, including the Scottish Household Survey, the main source of population data after the census. Both NRS and Police Scotland state that this direction was taken as part of future-proofing for the current Gender Recognition Reform (Scotland) Bill:
It is important to recognise that the development of Scotland’s Census sits within a wider Scottish Government context. NRS has worked closely with colleagues in the Scottish Government undertaking the Gender Recognition Act consultation and consultation with colleagues in Equality workstreams to ensure a cohesive approach to the census questions.
NRS, 2019: para. 3.29
In 2019, to prepare for the Gender Recognition Act reform which was taking place at the time, Safer Communities E&D team considered a draft Police Scotland policy on Gender Identification and assessed that a Police Scotland position statement was required. A position statement (below) was developed by Police Scotland. This was approved by the Senior Leadership Board in November 2019 for use in response to enquiries while Police Scotland awaited direction and guidance from Scottish Government on the identification and recording of sex and/or gender, which would emerge from a review by the Scottish Government’s Chief Statistician.
Gender self-declaration position statement:
“The sex/gender identification of individuals who come into contact with the police will be based on how they present or how they self-declare, which is consistent with the values of the organisation.
Police Scotland requires no evidence or certification as proof of biological sex or gender identity other than a person’s self-declaration, unless it is pertinent to any investigation with which they are linked as a victim, witness or accused and it is evidentially critical that we legally require this proof, or there is reason for further enquiry based on risk. We would look for the most sensitive way to acquire this information.”
Police Scotland, 28 January 2022
Whose voices count?
This process of loss of data on sex has happened because the Scottish Government has repeatedly ignored expert opinion on data collection.
In December 2019, 80 experts in social statistics and users of population level data, including some of the UK’s leading social scientists and ten Fellows of the British Academy, wrote to the Scottish Parliament’s Culture, Tourism, Europe and External Affairs (CTEEA) Committee, setting out their concerns:
As experts in social statistics and users of population level data, we call on the UK’s census authorities to retain the integrity of the category of sex, and not to conflate this with gender identity.
18 December 2019
In February 2021, 91 quantitative social scientists wrote to Scotland’s Chief Statistician to express concern about the proposed loss of data on sex:
We find the conclusion that sex-based data should rarely be collected astonishing. 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 (Chao et al. 2019). Sex is a powerful predictor of almost every dimension of social life: education (Stoet et al. 2016), the labour market (Joshi et al. 2020, Bryson et al. 2020), political attitudes and behaviour (Green and Prosser 2018), religion (Voas et al. 2013, Voas 2015), crime (Ministry of Justice 2017), physical health (Koblinsky, Campbell and Harlow 2018), mental health (Ploubidis et al. 2017), cultural tastes and consumption (Sullivan and Brown 2015)–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 (Perez 2019).”
Sullivan et al. 12 February 2021
Instead, the Scottish Government has given precedence to assertions put forward by activists who are not close to having an equivalent level of expertise in dealing with population data:
Reversing the damage
The loss of robust data on sex in Scotland over a relatively short time can be attributed to lobbying to prioritise self-declared gender identity over sex, coupled with a failure of government and other public bodies to balance different sets of interests, and listen to the concerns of academic data experts.
This has led to a position where it is no longer possible to say what most of the data collected by Scottish public authorities marked as ‘sex’ is actually recording.
The Scottish Government has failed to demonstrate why it rejects the position held overwhelmingly by senior quantitative social scientists and instead believes sex at birth is no longer a valuable explanatory variable across a range of public policy outcomes and other social phenomena. The only way to test robustly how sex and gender identity each relate to various life experiences, and to each other, is to collect data on both: the Scottish Government however appears resistant to this approach.
In behaving as it is, the Scottish Government is damaging our ability to understand how sex operates in Scotland as a determinant of social, economic, and physical outcomes, and to redress discrimination on the basis of sex. It is also inhibiting a better understanding of how self-declared gender identity compares to sex as a predictor of experiences. Future Scottish governments now stand to inherit this damaged data legacy.
At present, the EHRCJ Committee appears to be planning to hear only from witnesses who support replacing sex with gender identity in public data collection. For more detail on the approach to witnesses to date, see here:
Committee members could however use the Bill process to reverse rather than cement in the damage already done. As Professor Alice Sullivan states in her written evidence to the Committee:
The bill should be amended to provide explicit protection for sex-based data collection. Without sex-disaggregated data, it will be impossible to assess the effects of the proposed reform. Serious policy evaluation will be rendered unfeasible.
Gender Recognition Reform (Scotland) Bill (Detailed): Professor Alice Sullivan
Putting data collection in Scotland more generally back onto a solid basis is a much larger task, but decisions made in the Scottish Parliament will be influential. Adopting Professor Sullivan’s approach would make clear that, in modernising the process for obtaining a GRC, the Scottish Parliament did not mean to suggest that biological sex does not still matter, or that privacy protections are intended as a general impediment to collecting data on sex, where needed.