Sex and gender - draft guidance: consultation analysis

Summary of Responses to Draft Guidance

Overall, the feedback portrayed a sense that most people thought it was important to collect data on sex, and a number of reasons were cited such as it being a protected characteristic, that it is already commonly used in data collection and an understood term, that data on sex is needed for understanding differences between men and women, where discrimination and inequality lies. There was less clear agreement on how this data should be collected, for example, some responses advocated that it should be biological sex, others legal sex, lived sex or sex registered at birth.

Historically the terms sex and gender have been used to mean the same thing, however, there was a clear view (and this is supported by draft guidance published by Office for Statistics Regulation) that the terms should no longer be used interchangeably. Linked to this was feedback in a number of responses that the definitions set out in the draft guidance were used inconsistently and not in keeping with their use in practice or law, or common understanding.

Linked to this distinction between sex and gender identity was a clear view that public bodies should be collecting data on both sex and on gender identity and/or trans status so that there is evidence on women and men and the trans population to use when designing services, tackling discrimination and improving better services for the public. There was opposition to the suggestion that public bodies should primarily collect data on gender identity rather than sex.

One of the key purposes of developing this guidance is to encourage public bodies to gather and use data to advance women’s equality and rights, and support them to decide how and why to collect and use this data. Some responses felt that the focus within the draft guidance was too much on how the questions should be asked, and not enough emphasis on this principal issue.

As well as feedback on the substance of the draft guidance, there were also a number of comments about how the guidance could be made more helpful for users, for example improved presentation, simplified language, more examples of when certain questions might be asked.

Rather than answering specific questions, people were invited to share feedback on an open basis. The analysis presented below summarises the most common and recurring points that respondents’ mentioned in their responses.

Definitions

• The difficulty of trying to strictly define the term sex, and concept of gender/gender identity when there is no single agreed definition in law and/or practice. Conflation of these terms in the draft guidance.

A number of responses commented on the definitions of sex, gender and gender identity that are set out in the draft guidance, highlighting that these were unclear, used inconsistently throughout the guidance and contrary to their definition in law and/or practice.

For example, some stakeholders and respondents to the consultation referred to gender as a social construct rather than a personal attribute (e.g. Engender) and felt this was conflated with gender identity. It was recognised that historically the terms sex and gender have been used interchangeably to refer to whether someone was a man or a woman, though more recently the meanings ascribed to the terms has begun to change.

In particular, responses highlighted that gender identity is not defined in law and has no legal standing, whereas sex is defined (e.g. in the 2010 Equality Act).

The conflation of the terms sex, gender and gender identity throughout the draft guidance and in the recommended questions was seen to undermine the messaging in the guidance, and the lack of consistency in their use within the guidance is generally unhelpful.

“Gender and gender identity are muddled in the draft guidance.”

• Some responses voiced concerned that the proposals ‘advocate collecting data relating to sex by asking for responses in relation to gender identity’, thereby conflating two protected characteristics (i.e. sex and gender reassignment) and impacting on the ability of bodies to meet obligations under the 2010 Equality Act.

Sex

• Failure to collect data on sex strongly highlighted as an issue. The importance of collecting data on both sex and gender identity/trans status so that the needs and experiences of different groups of people can be understood and measured.

There was a consensus that collecting data on sex was important (though there were differences of view on how data on sex should be collected).

A majority of the responses received were concerned that the draft guidance implied that in most situations where data on men and women is required, that a question on gender identity should be asked, with a sex question only asked in a small number of scenarios where it is directly relevant, for example, to a person’s medical treatment.

This view was shared across a range of respondents who were concerned that it would impact on the ability to carry out [gendered] analysis of sex-disaggregated data, that it would prevent public bodies from meeting their duties under the PSED by failing to collect data on sex (one example referred to it undermining the Scottish Government’s ambitions on gender mainstreaming and gender competence within policymaking) and that it would have negative impacts for data quality and reliability given that gender identity is seen by many as a subjective concept that could change over time.

“it is clear to us that a failure to gather, analyse, and use gender-sensitive data has been a serious impediment to gender mainstreaming within policy and programme development. (Engender)”

A couple of responses referred to the draft guidance being at odds with the UN Gender Statistics Manual^[2] and the European Institute of Gender Equality Gender Statistics Database^[3] which together gives a list of the features of competent gender-sensitive sex-disaggregated data. These are:

a) Data are collected and presented by sex as a primary and overall classification;

b) Data reflect gender issues, such as childcare spending and provision;

c) Data are based on concepts and definitions that adequately reflect the diversity of women and men and capture all aspects of their lives; and

d) Data collection methods take into account stereotypes and social and cultural factors that may induce gender bias in the data.

A number of responses suggested that public bodies should collect data on both sex and gender identity/trans status, arguing that this will allow them to better understand the needs of distinct groups of people.

“The UK LGBT survey demonstrates how much more can be done with a data set which collects both sets of data [sex and gender identity]. (MBM policy analysis collective)”

“accurate data on sex, in combination with data on gender identity, has the potential to improve our understanding of the discrimination faced by trans people of either sex.”

“concern that the current [proposals] concerning the collection of sex-disaggregated data in Scotland appear to suggest that this would rarely be required and only collected in exceptional circumstances…[sex is a] key demographic variable [this would make it] impossible for quantitative data analysts and social scientists to accurately describe trends”

In terms of the gender identity question, some feedback suggested that this should have additional response categories ‘non-binary’ for people who identify as neither female or male, and ‘none’ for people who reject the concept of a gender identity.

• How data on sex is collected

There was very little support for a question on legal sex. A number of reasons why were cited, including because it is out of step with how sex is defined in law (i.e. male or female); very few (if any) other data collections use this question to collect data on sex; public bodies may consider there to be privacy concerns with collecting and holding data on a person’s legal sex and therefore be put off collecting the data at all; for trans people, it was highlighted that a number do not obtain gender recognition certificates and may see being asked to disclose their sex at birth intrusive.

“It is not clear from the guidance why and when public bodies should be asking questions about ‘legal sex’…[it] does not present any use cases for this information specifically”

“Any definition in the guidance of “sex” as meaning current birth certificate sex would therefore remain anomalous and inconsistent with other practice.”

A number of responses suggested that sex should be collected on biological sex, whereas other responses suggested that a lived sex question would be most appropriate. Others simply stated that data on ‘sex’ is important.

“In terms of monitoring change over time and across space, it is vital that we retain biological sex as a variable”

“Lived sex is the most useful and relevant thing for public bodies to know about a trans person. This is because trans people, regardless of whether or not they have a GRC, and regardless of their sex registered at birth, are lawfully able to, and do, interact with public bodies in line with how they are living.”

It was highlighted that for trans people, asking one question cannot give you all of this information. That is true whether you ask a question on their sex registered at birth, physical and biological sex characteristics, the sex on their current birth certificate, or how a person lives – these will not align in the “typical” way that they do for the vast majority.

“Whilst we agree that this is the correct approach to gathering data on sex/gender, we are not convinced that this question needs to be labelled “gender identity” – we would be comfortable with it being labelled “sex”, “gender” or “gender identity” (or in fact to have no label at all).”

A small number of responses suggested that the inclusion of intersex on a sex question was important.

Some responses highlighted instances when it might be necessary to collect data on a person’s biological or legal sex, including for healthcare purposes, and during interaction with the criminal justice system. The guidance needs to be clearer on examples of when this may be necessary and proportionate (in line with the Equality Act 2010).

Quality

• Potential impacts of the proposals in the draft guidance on data quality, reliability, consistency and harmonisation.

Concerns were voiced that the proposal ‘in most situations data should be collected on the basis of gender identity’, would impact on data reliability because gender identity is seen as a subjective concept, compared to for example, a person’s sex. It was suggested that this could impact on data consistency over time.

Relevant to this is a recent Scottish Government commissioned report on understanding equality data collection in the Scottish Public Sector which looked at how data on equality characteristics, including sex and gender is collected by a sample of public sector bodies. That report found that there was a lack of standardised approaches to collecting this data in terms of definitions and terminology used, as well as response category options. In terms of data collection methods, organisations that collected equality data directly from their customers through an online system generally reported higher quality, more complete data. The advantages of online data collection included: (i) the facility for individuals to update their own information, (ii) the (in general) higher quality of self-reported data (compared with data collected through an intermediary or through observation), (iii) greater efficiency and cost-effectiveness, and (iv) the greater ability to build quality control procedures into data collection processes. Therefore, it is important to bear this in mind when considering current levels of data reliability and quality.

More generally, a number of responses raised concerns that there had been a lack of engagement with academics and researchers who work with large population data sets, and therefore that the proposals in the draft guidance do not adequately consider possible impacts on data reliability and quality.

Related to this theme were concerns that the impact on sub-group analysis was not mentioned in the draft guidance. Specifically, comments highlighted that the size of the trans population is unknown and the draft guidance cites evidence from 2011 to support a statement that the trans population is small and therefore unlikely to have an impact on the quality or reliability of data.

“However the EHRC report is based on non-representative trial data (using panel members who had signed up to answer online surveys) and states that the data ‘cannot be used to estimate the percentage of the population in particular trans groups’ (2012: 4).[1] It is a concern that this clear caveat is ignored.”

“persistent international evidence of sharp growth in the numbers identifying as transgender, particularly among younger age groups. This is not acknowledged anywhere in the draft guidance… Without a reliable estimate as to the size and demographic distribution of Scotland’s trans population, the proposals in the draft guidance represent a clear risk in terms of data reliability.”

Some responses noted challenges to harmonisation resulting from a change to how public bodies are instructed to collect data on sex and/or gender.

“Trans people are already recorded on a range of administrative systems with their lived sex rather than legal sex. This will therefore generate further problems with comparing data across instruments and data sources.”

“the Working Group’s recommendation for a sex question that is externally-identified (i.e. linked to another identity document, a birth certificate) is a departure from the existing precedent in data collection activities undertaken by public bodies in Scotland”

Linked to this, the Census was mentioned in a number of responses. A fairly large number shared direct feedback on the sex question in the Census, and their views that this should be a biological sex question. The draft guidance is in no way related to the Census question development work.

Other feedback expressed concern that the draft guidance advocates a different approach to collecting data on sex and the trans population compared to Scotland’s Census 2022, and the impacts on comparability this could have.

“While it is understandable that proposed ‘sex’ question differs from the approach taken by the Census 2022… there is a sense of disappointment that direct comparisons with locally and national data may be made more difficult in future years because of this.”

“We also wish to express concern about the Working Group’s proposal to depart from the format of the sex question used in Scotland’s forthcoming census”

Evidence

• The need for an evidence driven approach – to set out clearly the evidence on which the proposals in the guidance are based.

Some feedback commented on the lack of evidence cited to support the proposals in the draft guidance. For example, in relation to the suggestion that a question on gender identity should replace a question on sex, the point was made that some people actively reject having a gender identity, and the evidence for there being a concept of gender identity questioned.

“The claim that gender identity should be collected in preference to sex appears to rely on an implicit assumption that gender identity is more important in determining (all) relevant outcomes than sex. But no evidence is provided to support such a proposition.”

“the guidance was a little light in justification to persuade public bodies to reject the “extensive testing and feedback” that led to the decision NRS made and instead adopt the question recommended in the guidance.”

“offers no theoretical argument as to why gender identity would better explain differences in the experiences of women and men, or empirical evidence that it does”

Some responses highlighted the conflict between conclusions made in the draft guidance and the statistical principles that underpin the work.

“That the case for collecting data on sex has been made by senior academics who routinely work with large population data sets, including statisticians, as noted earlier, is not acknowledged in the draft guidance.”

“As a wholly subjective category, grounded in a person’s feelings, compared to sex it has substantially more limitations and is less easily reconciled with the various statistical principles that are purported to underpin the proposals.”

Others drew attention to various statements made by Scottish Ministers, and the direct conflict between these and the proposals in the draft guidance.

“The draft guidance thus precludes data collection that ‘takes account of the differences—including biological and physical differences—between men and women, and their impact’, as noted by the Minister.”

“Cabinet Secretary for Health on 3rd March 2020, made a commitment to take account of the importance of data collection disaggregated by sex.”

Some responses suggested that providing a justification in the guidance for asking the trans question of those age 16 or over would be helpful, and conversely why that justification doesn’t apply to questions on legal sex and gender identity. This would assist public bodies to explain why they are/are not collecting this information from those under 16.

The draft guidance offers no theoretical argument as to why gender identity would better explain [than sex] differences in the experiences of women and men, or empirical evidence that it does. (MBM Policy Analysis collective).

What does the law say?

• Legal framework and considerations linked to the collection of data on sex, in particular, where discussed in the draft guidance are unclear and in some parts potentially misleading.

A number of responses commented on the proposals in the draft guidance and their perceived mis-alignment with the legal frameworks around sex, equality Act. PSED and GDPR.

“That the guidance be clarified to support systematic decision-making about how and when to gather and use particular data, including appropriately contextualised information about regulation such as GDPR.”

“it is our understanding that legally, there is no difference in Scots, UK, or relevant EU or international law between sex and gender. Instead, the concepts of sex and gender in the law are used interchangeably, and are interrelated.”

“it is necessary to collect data on gender identity, but this needs to be carefully considered in terms of the possible impact on women's protected rights.”

“Public bodies have a legal duty to demonstrate what they are doing to comply with their public body duties under the equality act 2010 in regard to sex.”

“Sex is clearly defined in law and women's rights in law are tied in law to sex.”

A few responses pointed to concerns around some of the advice provided by the Equality and Human Rights Commission in Scotland to the Working Group on whether a public body should collect data on biological sex, citing alternative legal view that privacy rights are not absolute and collecting data on biological sex is permitted for a legitimate and proportionate aim.

“The draft ignores the fact that organisations sometimes need to know staff or users’ sex to operate sex-specific services, such as recently agreed for forensic examinations for rape victims.”

“It would be inappropriate for the [draft guidance] to recommend that single-sex services should be asking a question about current birth certificate sex in order to make a decision of “whether to offer a single-sex service”. The sole legal test in the Equality Act is whether, by treating the trans person less favourably or excluding them, the service is applying “a proportionate means of achieving a legitimate aim” [Equality Act 2010, schedule 3, paragraph 28].”

High level aim(s) of the guidance

• Agreement on the importance of collecting, analysing and using this data

There was widespread agreement that it was important to collect data on women, men, and the trans population.

Many reasons were cited why it was important, including:

• to support parliamentary scrutiny and the legislative process;
• to understand women’s experiences;
• to support gender mainstreaming and policy/programme development;
• to reflect the diversity of women and men and capture all aspects of their lives;
• to meet obligations under the UN Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW);
• to advance women’s equality and rights;
• to comply with the 2010 Equality Act in order to avoid discrimination, both direct and indirect, on the basis of sex;
• to tackle indirect discrimination against females/gendered inequalities;
• to build an evidence base about trans people’s experiences;
• to allow meaningful analysis into specific needs of the distinct groups.

Presentation

• Drafting of the guidance

Many responses shared feedback on the drafting of the guidance, in particular areas where this could be improved. Generally the drafting was seen as convoluted and overly complex, with lack of clarity on key areas such as the definitions, examples of when bodies may need to collect certain data, and appropriately contextualised legal and other statutory information.

This included omissions from the work – for example, the lack of Equality and Human Rights Impact Assessment to accompany the draft, lack of information on things public bodies should consider around good data handling from a GDPR point of view.

“The section on intersectionality needs considerable strengthening and expansion”

“the guidance risks losing sight of the principal issue: how and why public bodies should be gathering and using data to advance women’s equality and rights…[instead it] focuses disproportionately on the narrow questions around how to ask individuals about their sex and gender identity”

“it is vague and potentially misleading on some other important issues and generally unsystematic”

“That the guidance be clarified to support systematic decision-making about how and when to gather and use particular data, including appropriately contextualised information about regulation such as GDPR.”

“we note that the majority of the draft guidance is not actually focused on how to gather data that provides an evidence base for tackling gendered inequalities, but instead is focused on how trans people fit in the collection of sex/gender data”

“The guidance should advise public bodies to take the same approach to reporting on data about non-binary people as it does data on other groups that make up a small number of the general population”

“Without clarity, this description of definitions is open to interpretation, risks confusing public bodies, and is likely to result in divergent approaches to defining sex and gender”

“guidance could go further in setting out a core data collection plan which public bodies would be expected to adhere to unless they can justify deviating from it. A useful way of summarising the guidance might be to have a table setting out common variations of the questions and the pros and cons of each.”