On 29/1/20 I sent an email to Professor Roger Halliday, Chief Statistician & Data Officer for the Scottish Government, for the attention of all the members of the Sex and Gender in data Working Group. Below is the full text:

Dear Roger Halliday,

I have recently become aware of your discussions during the Sex and Gender in data Working Group meeting on 28/11/19, the note of meeting states:

“Members discussed the principles that will be important in putting together guidance on the collection of data on sex and gender to give organisations and individuals’ confidence, and to support their decisions on delivering and improving public services. This includes the use of clear definitions that are inclusive, and ensuring that guidance is easy to interpret and relevant to real-life scenarios.”

I am writing to you as I am concerned about what will be the groups interpretation of “clear definitions that are inclusive”. Recently, the National Records of Scotland published their report about the Census Sex Question and its accompanying guidance and in it they have recommended that the Sex Question guidance directs respondents to answer based on their self-identified sex or gender identity. This was despite the strong recommendations from academics and experts in statistical analysis to not conflate sex with gender identity.

I believe that continuity and clarity is needed within Scotland as to the definitions of terms like Sex, Gender and Gender Identity, as it is clear from the wider public discourse there has been considerable criticism of the changing use of these terms in recent years. Having looked across European and Global organisations to identify whether there is a consensus on what definitions national organisations and Governments should use I have discovered the United Nations have produced such guidance. The UN Gender Statistics Manual details how to incorporate accurate and reliable statistics regarding gender, including information on how this should operate at organisational level, within Governments and national statistical offices, like those responsible for the Census.

Below are some quotes from the UN Gender Statistics Manual:

  • Confusion between “sex” and “gender” still persists among producers and users of statistics.
  • The word “sex” refers to biological differences between women and men. Biological differences are fixed and unchangeable and do not vary across cultures or over time.
  • “Gender”, meanwhile, refers to socially‐constructed differences in the attributes and opportunities associated with being female or male and to social interactions and relationships between women and men.
  • The term “gender” has often been wrongly used in association with data. “Gender disaggregation” or “data disaggregated by gender” are incorrect terms.
  • Gender statistics are disaggregated by sex, an individual‐level characteristic commonly recorded in censuses, surveys and administrative records, not by gender, a social concept relevant at the level of a population group.
  • When data on demographic, social or economic characteristics are collected in the field, it is the sex of a person that is recorded, as female (woman) or male (man), not the gender.
  • All national statistical systems produce statistics disaggregated by sex.

The UN Gender Statistics Manual very clearly highlights the differences between “sex” and “gender” and stresses the importance of why these should be used consistently by all nations. It is therefore concerning to discover that the National Records of Scotland have opted instead not to use the UN definition of “Sex” within the next Census. This will no doubt set a precedent for all other data collecting surveys in the country. It also raises the question of when Scotland’s statistics are thereafter forwarded to Eurostat how will they then be able to identify important information e.g. the number of same sex marriages, or civil partnerships. If Scotland’s national organisations no longer regard “Sex” as referring to the biological differences between men and women then this will have a major impact on ALL data that relies on this most important variable. The redefinition of a value like sex to something other than biological does not align with the global standards of mainstreaming data collection.

European Institute for Gender Equality (EIGE)

“The European Institute for Gender Equality (EIGE) was established to contribute to and strengthen the promotion of gender equality, including gender mainstreaming in all EU policies and the resulting national policies, and the fight against discrimination based on sex.’

The EIGE refers to the United Nations Gender Statistics Manual for its definitions and other information on Gender Statistics. See here.

  • EIGE definition of Sex: Biological and physiological characteristics that define humans as female or male.
  • EIGE operates within the framework of European Union policies and initiatives.
  • The European Parliament and the Council of the European Union defined the grounds for the Institute’s objectives and tasks in its Founding Regulation and assigned it the central role of addressing the challenges of and promoting equality between women and men across the European Union.

The EIGE Glossary uses the correct term sex-disaggregated data and does not use incorrect terms like “Gender disaggregation” or “data disaggregated by gender”.

The EIGE are very clear that gender statistics are disaggregated by sex. See here.

Equality Act 2010 Protected characteristic ‘sex’

The members of the Working Group might be interested to know that last year I carried out a review of the Equality Impact Assessments (EQIA) published on the Scottish Government’s website. The purpose of this was to identify how many had accurately described the Equality Act 2010 protected characteristic ‘sex’. Over a 12 month period there were forty-one EQIA’s listed, yet many were vague and did not reference any of the individual protected characteristics, therefore it was impossible to know if during the assessment process they had been accurately described. Further to those unknown, there were eleven that incorrectly referred to the protected characteristic as ‘gender’ where they presumably meant ‘sex’; or they had grouped the protected characteristics of ‘sex’ and ‘gender reassignment’ together. This inaccurate description and/or omission of the protected characteristic ‘sex’ didn’t just occur within the EQIA’s, as this was repeated across numerous other documents and publications on the Government website, for example, the Scottish Draft Budget 2017/18 Equality Statement omitted ‘sex’ entirely then incorrectly states that both Gender and Gender Identity are the protected characteristics. For further details of these findings see ‘What’s the priority, Gender or Sex?’.

The repeated conflation of sex and gender along with the apparent inconsistency in the understanding of the differences between these terms has a negative impact on any measures that are then introduced to end discrimination against women and girls. For example, the Government fully acknowledge that women and girls are discriminated against because of our sex as when they introduced to Parliament a Bill to target the under-representation of women on public boards, amongst the relevant reasons for doing so they made reference to both the International Human Rights and Treaty Obligations and also the Convention on the Elimination of All Forms of Discrimination Against Women. (For further details of this Bill see the Policy Memorandum.)

“In addition to the moral and economic imperative for increasing women‘s representation on Scotland‘s public boards, taking action is also crucial to upholding women‘s human rights. The Scottish Government is committed to protecting, respecting and realising human rights and takes its international human rights and treaty obligations seriously. Taking action to ensure that women are properly represented on public boards is imperative to meeting these obligations.”

“Article 7 of CEDAW requires that State Parties ―shall take all appropriate measures to eliminate discrimination against women in the political and public life of the country and, in particular, shall ensure to women, on equal terms with men, the right:

(a) To vote in all elections and public referenda and to be eligible for election to all publicly elected bodies;

(b) To participate in the formulation of government policy and the implementation thereof and to hold public office and perform all public functions at all levels of government;

(c) To participate in non-governmental organisations and associations concerned with the public and political life of the country”

However, as this Bill progressed through parliament it was amended to such an extent that a definition of ‘woman’ had to be included. This resulted with the Gender Representation on Public Boards (Scotland) Act 2018 defining ‘woman’ as including people with the protected characteristic of Gender Reassignment. Crucially, this included trans people who self identify as women, as there is no requirement to have obtained legal gender recognition via the process of the Gender Recognition Act 2004. What began as an equality measure to tackle women’s sex based discrimination as per Article 7 of CEDAW has resulted with new legislation and a definition of a ‘woman’ that includes trans people who are both biologically and legally male. (For further details on this see this post.)

The conflation of the protected characteristic ‘sex’ with gender, and its resultant diluted effect, is not restricted to parliament and is occurring in public bodies, organisations and charities across Scotland. For example, the Equal Representation Project, a coalition of Scotland equalities organisations, describes the commonly understood term ‘sexism’ as ‘prejudice, stereotyping, or discrimination on the basis of a person’s gender’ and not their sex. It is known that some Local Authorities incorrectly believe that UK Law uses the terms ‘sex’ and ‘gender’ interchangeably and have therefore replicated that within their own publications. Also, along with COSLA, they have stated that they prefer to “use the term “gender” in equalities monitoring and other documents because we believe it leads to more accurate data being collected which helps us to better protect people who are likely to experience inequalities. We recognise that the Equality Act 2010 refers to ‘sex’ but this is the legislative minimum standard.” (For further details on this see this post.)

It is a requirement of the Public Sector Equality Duty to eliminate discrimination and to advance equality of opportunity, but when public bodies fail to understand the differences between sex and gender, they are also undermining the discrimination, harassment and victimisation sex based protections afforded to women and girls.

The widespread misunderstanding and conflation of the terms sex and gender is troubling enough, however with this comes the realisation that Scotland’s organisations and public bodies have somehow lost the crucial point as to why there is a need to collect and analyse separate data on women and men. The diminishing importance of the protected characteristic ‘sex’, along with the increasing popularity of terms like ‘self identified sex’ or ‘lived sex’ (terms that do not exist in any legal framework or policies within the UK) within our organisations is a direct conflict of the rights of women and girls in Scotland and is a regressive step in the Government’s aim to eliminate all forms of discrimination.

It is my hope that the working group recommend that clear definitions, as set by the EIGE and the United Nations, are respected, understood and implemented across all sectors in Scotland.

Yours sincerely

Susan Sinclair

On 7/2/20 I received an emailed reply from Professor Roger Halliday.

He agreed that the definitions are important, and “also agree(d) that the starting point for any data collection needs to be a consideration of why we’re collecting data in the first place, and this will very much be the basis of the guidance I’m putting together.”

He also stated that:

Underlying this work is a set of statistical principles, which were agreed at the first meeting of the Working Group, and these are taken from the UK Statistics Authority’s Code of Practice for Statistics, the European Statistics Code of Practice and the UN Principles for Official Statistics. I will use these principles, and potentially a wider set of other principles, to help us consider what the guidance should be.

See full email response here.

Terms of Reference and Membership of the Group

Update: on 24/5/21 Roger Halliday, Chief Statistician wrote to Ed Humpherson, Director General for Regulation OSR

Dear Ed

Gender identity and Scottish Household Survey

There has been significant interest over the last couple of years in how we collect, analyse and use data around sex and gender. This led to the review of these issues that I’m leading in Scotland. There has been particular interest in the use of a gender identity question in the Scottish Household Survey, which was introduced in 2018.

While I’m expecting my review on sex and gender data to report over the summer, I think it would be very helpful for me and colleagues in Government for the OSR to do a compliance check on the arrangements we have in place for collecting and reporting gender identity on this survey. Ideally this would cover the work done around the introduction of this question. This would help provide assurance to me, users of this data and survey participants that this has been well handled.

Can you confirm whether this is possible, and if so, what the arrangements would be for taking this forwards.

Yours sincerely

Roger Halliday
Chief Statistician

On 28/5/21 to Ed Humpherson wrote back:

Dear Roger,

Gender identity and Scottish Household Survey

Thank you for your letter of 24 May requesting a compliance check of the Scottish Household Survey focused on the collecting and reporting of data about gender identity.

As you say, there has been significant interest in the collection of data about sex and gender. I’m pleased to receive your request for this review of the Scottish Household Survey against the standards of the Code of Practice for Statistics with regard this matter, and I hope any findings or outcomes can feed into your wider review on sex and gender data in Scotland.

My team will take forward this work in the coming weeks and will be in touch with statisticians involved in the Scottish Household Survey shortly to agree arrangements.

Yours sincerely,

Ed Humpherson
Director General for Regulation

Update: The new Data Collection and publication Guidance was published on 22/9/21.

Now it turns out the Household Survey questions are being finalised for 2022. This was confirmed in October, after the publication date of the new Data collection and publication guidance, so i can only assume the questions will be in line with those recommendations.