Sex harmonised standard

What we mean by harmonisation

Harmonisation is the process of making statistics and data more comparable, consistent and coherent by using consistent definitions, survey questions, data presentations and information for data users. The Government Statistical Service (GSS) Harmonisation Team works across the UK to promote and facilitate harmonisation. To improve harmonisation, analysts, researchers and data owners should aim to:

• use consistent definitions and survey questions in data collection
• improve understanding of how data and statistics can be combined
• present data and statistics in a comparable and coherent way
• provide guidance where consistency is not appropriate – for example, where there are specific user requirements

We provide guidance encouraging the use of standard data collection approaches across government departments and the wider public sector, where these are appropriate to meet user needs. Being harmonised is about aligning with others, not necessarily being identical to them. As such, some of our harmonised standards can be adapted for specific situations. It is not always possible to have completely harmonised data and statistics for all topics. This is because of differences in data collection, legislation across UK nations and user needs. We think of harmonisation as a spectrum, which can include actions like:

• communicating what sources can and cannot be compared
• indicating where our ability to successfully compare data and statistics is limited
• harmonising output categories in statistics

These are all types of harmonisation. Each of these methods help users get a better understanding of data and statistics. They also help users to make appropriate decisions about the data. This means statistics do not always have to be standardised to be harmonised.

This standard will continue to be reviewed and expanded where necessary in line with changing respondent and user needs, in line with Recommendation 7 of the National Statistician’s Inclusive Data Taskforce.

Who this guidance is for

This guidance is for producers of official statistics and GSS researchers who are currently using, or thinking about including, a sex question in their surveys or data collection. Organisations that produce official statistics are not required to use these standards, but doing so can improve the consistency, comparability and coherence of data and statistics.

What we mean by sex

Various ways of asking respondents about their sex have been used within official surveys and data collections over time. This has led to variation in the way that sex is defined, explicitly or implicitly, including:

• variation in terminology (for example, referring to gender instead of sex)
• variations in guidance (for example, for a trans or non-binary respondent on how they should interpret the sex question)

When we use the term ‘sex’ in this guidance, we are referring to a binary variable categorised as female or male. In the UK, an individual’s legal sex is recorded at birth based on their biological characteristics. The Gender Recognition Act 2004 enables people over 18 to apply for a Gender Recognition Certificate (GRC), which allows them to change their legally recognised sex for certain purposes, including on their birth certificate. Where an individual has a GRC, their legally recognised sex may differ from their sex recorded at birth.

In April 2025, the UK Supreme Court ruled that sex, for the purposes of the Equality Act 2010, refers to biological sex unmodified by a GRC. This means that references to the sex-based protections in the Equality Act are references to biological sex, that is, the sex of a person at birth.

For the purposes of this harmonised standard on sex, we have developed questions for two target concepts of sex. These target concepts are aligned with different legislative instruments which have a bearing on an individual’s official records and status:

1. Biological sex (sex at birth) aligned with the definition for the purposes of the Equality Act
   and
2. Certificated sex (sex at birth or as modified by a Gender Recognition Certificate) aligned with the legal recognition provided by the Gender Recognition Act

This reflects the urgent user need for statistical guidance shared with us by official statistics producers following the UK Supreme Court ruling. We recognise that user needs and the context for official data collection continue to evolve. Therefore, further testing may be needed to update the harmonised standard as circumstances change.

We use the phrase ‘sex at birth’ rather than ‘biological sex’ because of differences in the way that potential respondents interpret the meaning of biological sex. The UK Supreme Court defined its use of the term biological sex as describing “the sex of a person at birth”. Since this term was less likely to be subject to variations in respondent interpretation, we chose to directly refer to the Supreme Court’s definition as a target concept for data collection.

We use the phrase ‘certificated sex’ to mean a person’s sex at birth, or as modified by a GRC. An individual may change their sex on some official documents, including a driving licence, a passport, medical records, employment records, and a bank account. However, these changes do not require them to have changed their legally recognised sex with a GRC, and are not included within this definition.

For some individuals, their personal sense of their own gender may differ from their sex recorded at birth, their legally recognised sex, or both. We use the term ‘gender identity’ to refer to this individual experience of gender, and we use the term ‘trans’ or ‘transgender’ to describe people whose gender identity differs from their sex recorded at birth. Gender identity is not constrained to binary male/man and female/woman categories and includes others such as non-binary and gender fluid. Some, but not all, trans people want to undergo a process to assign their sex, which may involve changing physiological or other attributes of sex. The Equality Act states that an individual has the protected characteristic of gender reassignment if they are proposing to undergo, are undergoing, or have undergone such a process. The GSS Harmonisation team have separate guidance for collecting gender identity, and are working to produce a new standard which takes into account the needs of data producers, data analysts, and respondents, on these topics.

Considerations when choosing whether and how to collect data on sex

Many statistical collections will routinely gather and publish information on sociodemographic characteristics, such as sex. Nevertheless, producers of official statistics and researchers must have a strong understanding of their own data collections and how these meet the needs of their users. This is necessary so that you can appropriately analyse, identify and minimise the data protection risks associated with processing personal data, for example as part of a Data Protection Impact Assessment (DPIA).

The Office for Statistics Regulation (OSR) has provided guidance to support official statistics producers in the applying the principles of the Code of Practice for Statistics when making decisions about the collection and reporting of statistics and data about sex and gender identity. This includes encouraging data collectors to ask themselves the following questions when planning to report on sex:

• Do you have a clear reason to collect information on sex?
• Are you clear on how you are defining sex?
• Does your definition of sex meet the needs of the different users of your statistics?
• Do you understand what data you can, or should, collect about an individual’s sex to comply with relevant legislation?

This harmonised standard includes questions aligned with two target concepts of sex. Organisations should determine which question is most appropriate based on their assessment of user needs and the intended purpose of the statistics being produced. In some circumstances, it may be appropriate to collect biological sex (sex at birth), for example where statistics are produced to support monitoring under the Equality Act, such as gender pay gap reporting, or where this definition of biological sex is relevant to the aims of medical or scientific research. Where statistics could be adequately provided through the collection of less sensitive data, it may be more appropriate to collect certificated sex. You can read more about this in the next section of this standard, titled “General Data Protection Regulation (GDPR) and the data minimisation principle”. The choice of question should always be informed by the concept the statistics are intending to measure and the needs of data users.

You should be clear about what information you are collecting and why, and explain this both to respondents and data users using plain language. It is the responsibility of producers of official statistics and researchers to understand and be able to explain to users the strengths and limitations of their data collections, data, and statistics. They must also be able to explain any effect these strengths and limitations may have had when making decisions on collecting data on the topic of sex. This is in line with the Code of Practice for Statistics principle of Trustworthiness, specifically that data collectors will collect, use and share data in line with legal obligations and ethical principles and be open about choices made.

General Data Protection Regulation (GDPR) and the data minimisation principle

When collecting or processing data about individuals, statistical producers should refer to the General Data Protection Regulation (UK GDPR) and the Data Protection Act 2018 (DPA 2018). The Information Commissioner’s Office has produced guidance discussing the research provisions in the UK GDPR and DPA 2018 in detail.

You are responsible for determining whether there is a legitimate reason to collect data about sex, and how you will do so including the target definition for respondents to report on. To make this decision, you should carefully consider:

• your user needs
• relevant official guidance and legislation
• what this information will be used for
• what other data or statistics, if any, you plan to compare or combine your data with

Data minimisation is a cornerstone principle of data protection. Researchers and statistical producers should seek to minimise the volume and sensitivity of data collected in accordance with this principle. This will include considering not only what concept of sex you intend to collect, but what other data are collected at the same time and may already be held. Data on a person’s sex is not by itself considered special category data, meriting specific protection under GDPR. However, it can involve special category data where it reveals specific details about sensitive personal data alongside other known information. An example would be information about an individual’s health and medical care.

Information about a person’s sex as recorded on their birth certificate before an application for a GRC was made, or information about their application for a GRC, is protected under Section 22 of the Gender Recognition Act, which provides specific information on how disclosure of such information should be treated.

Questions and response options (inputs)

Question testing evidence relating to the development of the following questions for data collection on sex can be found in the research report accompanying this publication. These questions were developed to support respondents to more easily understand the target concept being asked within a question on sex, in order to improve data quality and manage respondent burden. Respondent burden refers to the degree to which a survey respondent perceives participation in a survey as difficult, time consuming, or emotionally stressful.

Question for biological sex (sex at birth) aligned with the definition for the purposes of the Equality Act

If the purpose of your data collection includes the collection of biological sex to produce statistics to support monitoring under the Equality Act, or where this definition of biological sex is relevant to the aims of medical or scientific research, the recommended sex question is as follows:

This question is recommended for online self-completion mode. Further testing for paper self-completion mode, telephone and face to face interviewing will follow.

Question for certificated sex (sex at birth or as modified by a Gender Recognition Certificate)

If the purpose of your data collection includes the collection of certificated sex or could be adequately provided through the collection of less sensitive data (as discussed in the previous section of this standard, titled “General Data Protection Regulation (GDPR) and the data minimisation principle”), the recommended sex question is as follows:

This question is recommended for online self-completion mode. Further testing for paper self-completion mode, telephone and face to face interviewing will follow.

Respondent guidance

In general, we recommend that questionnaires should avoid relying on respondents reading additional guidance to help them provide a response. However, it is not always possible to remove all guidance from a questionnaire. In the sex harmonised standard additional respondent guidance (“by sex we mean…”) is presented alongside the question text in both the question for biological sex (aligned with the definition for the purposes of the Equality Act) and the question for certificated sex. This approach was carefully tested with participants and proved to be more effective at guiding responses and improving the acceptability of questions than questions without any such guidance.

The harmonised standard sets out that this additional guidance should be presented in the same place as the question text, and not on a separate linked guidance page. This is to ensure that the definition for sex which aligns to the user need for data collection is visible without expecting the respondent to take further actions or to navigate away from the main survey page, which would add burden to the questionnaire. In most cases, respondents will not need additional information to help them answer a question about their sex. For those that do, research has shown that respondents might access question guidance held on a separate page sporadically or not at all, meaning that there would be inconsistency in the information respondents use when they provide an answer.

Information relating to the purpose for collecting sex data may optionally be presented on a separate linked page. You can find out more about this in the “Demonstrating trustworthiness” section of this harmonised standard.

Response option ordering

Response options were updated from “1. Male, 2. Female” to “1. Female, 2. Male” in:

• Census 2021 in England and Wales
• Scotland’s Census 2022
• Census 2021 in Northern Ireland

Census development reports for England and Wales describe that participants in respondent testing noted that the display of the response options, with “Male” first, did not conform to rules applied to other questions. Response options are typically displayed alphabetically, or with the largest group first, unless there is a reason to order response options in a different way. The harmonised standard follows this approach to alphabetical ordering, as no evidence was found in respondent testing to indicate this led to any loss of data quality or additional respondent burden.

Using this standard

Demonstrating trustworthiness

Across all our testing for the sex harmonised standard we heard feedback from respondents that indicated they would want to know the justification for collecting:

• data on sex
• data on sex rather than data on gender
• data on sex based on specific definitions

Questions on the purpose for data collection were frequently raised by transgender respondents, but also by some non-transgender respondents who wanted to be sure there was a legitimate reason for being asked this question. Unless it was clear to respondents that the definition of sex being collected was relevant to the purpose for collecting the data, they may be less likely to continue engaging. Respondents described that their expectations about the trustworthiness of sex data collection would vary by the identity of the data collector – for example, having greater trust in a large statistical survey than in data collection for commercial purposes.

We therefore recommend that researchers ensure information is available to respondents setting out the purpose of data collection for sex within their study. Providing information for the respondent on the purpose for the data collection, and the reason for defining sex in a specific way, increases the acceptability of the sex question as it allows respondents to understand why these data are needed and how they will be used. Where the purpose of a research question and the associated data collection are viewed more positively by respondents, risks to data quality and research drop-out can be reduced.

Proxy response, or completing the question on behalf of someone else

When collecting data on personal characteristics, including sex, you should aim to get an answer directly from the respondent themselves. This is part of good data collection practice. We advise putting in place appropriate ways to maintain privacy and protect respondent confidentiality.

It is important that you consider balancing data needs with respect for individual privacy, and whether it is appropriate to enable a proxy response.

Asking related questions

Gender identity

The GSS Harmonisation team is continuing to follow its workplan on the topic of gender identity. This work may also affect our future guidance on sex. We will update this section of the standard if additional advice or alternative options are needed for statistical producers intending to present both sex and gender identity questions in the same data collection instrument.

Presenting and reporting the data

The presentation of data on biological sex (sex at birth) aligned with the definition for the purposes of the Equality Act should follow the order of responses in the following box, and include a clear table header referencing the target concept collected, as shown:

Where appropriate, accompanying metadata should state the full definition of the target concept being presented as: “Biological sex (sex at birth) aligned with the definition for the purposes of the Equality Act”.

The presentation of data on certificated sex (sex at birth, or as modified by a GRC) should follow the order of responses in the following box, and include a clear table header referencing the target concept collected, as shown:

Where appropriate, accompanying metadata should state the full definition of the target concept being presented as: “Certificated sex (sex at birth, or as modified by a Gender Recognition Certificate) aligned with legal recognition provided by the Gender Recognition Act”.

If you are changing systems from older data structures where the response ordering was reversed, you should include clear information alongside your tables indicating the response codes which apply to each option to prevent data processing errors.

Comparability with other sex data collection instruments

Sex data are collected across many UK and international surveys. Sex has been a core UK census topic since 1801, with the question wording “What is your sex?” remaining largely unchanged over recent decades. Most UK surveys provide binary response categories to their sex data collection questions, with “Female” and “Male” being the typical response options.

Data collected using a guided target concept of sex (biological or certificated) and data collected using an unguided question on sex (that is, where respondents were left to select the response option which best matched their subjective definition of sex) will be differentiated by the proportion of respondents in your sample who hold subjective definitions which are broader than these biological and legal components.

Our research indicates that many transgender people differentiate between sex and gender identity when answering survey questions. Consequently, we cannot estimate the true size of the population who would answer an unguided sex question differently from a question guided for responses aligned to these definitions of biological or certificated sex. When comparing data collected using the harmonised standard for sex with data collected using an unguided target concept of sex, you should consider both the likely proportion of transgender individuals in the sample for each data collection and the impact of this for standard error rate and confidence intervals. In doing so, you should also consider the impact of smaller sample sizes resulting from intersectional analysis.

We are aware that many UK researchers and statistical producers are currently using a question consistent with Census 2021 in England and Wales. Respondent guidance presented online for the Census 2021 sex question in England and Wales reflected the definition of ‘sex’ for the purposes of the Census Act 1920 (as it applies in England and Wales), as agreed in a consent order made by the High Court in 2021. The guidance stated: “If you are considering how to answer, use the sex recorded on your birth certificate or Gender Recognition Certificate”. This question guidance directed respondents towards providing certificated sex (sex at birth or as modified by a GRC) as previously defined.

Data collected using the target concept of biological sex aligned with the definition for the purposes of the Equality Act, and data collected using the target concept of certificated sex will be differentiated by the proportion of respondents in your sample who hold GRCs. This will vary depending on the characteristics of your sample. You can access Tribunals Statistics Quarterly for more information on the number of GRCs that have been issued in the UK. In the period from April 2005 to March 2026, 11,223 full GRCs and 289 interim GRCs have been granted. It should be noted that this is the total number of GRCs granted during this period and does not take account of people who have since left or arrived into the UK and those who have died. Research which primarily samples respondents from LGBTQ+ populations would reasonably be expected to include a greater number of individuals with GRCs than population level data collection where data collected using either biological sex or certificated sex would be highly comparable. When comparing data for smaller populations, you should take into account the impact of standard error rate and confidence intervals.

When publishing outputs by sex you should be clear about what information you are collecting and why, and explain this using plain language. This helps ensure users understand the implications for onward data use, combination, and comparison.

Development of this standard

This standard was developed as part of a larger project to develop standards for both sex and gender identity. The methodology for the work we undertake to develop finalised harmonised standards uses Agile Development and Respondent Centred Design principles.

The Respondent Centred Design (RCD) Framework is a best practice approach that encourages the use of evidence gathered from research to design and develop data collection questions. The first stage of the RCD Framework is to establish the data user need, to understand how data users and analysts intend to use the data. Once user needs are established, RCD uses multiple methods and iterations of qualitative research with respondents to be sure that the wording and flow of the questionnaire meets the needs of respondents who will be answering the data collection question. For example, what do respondents need to:

• be able to understand a question
• answer quickly and easily
• provide accurate information for the data user

Testing for the sex standard has so far focused on development for an online self-completion mode, applicable for survey or administrative data collection in English. This is in line with the RCD framework which recommends taking an ‘optimode’ approach to design, where respondent materials and questionnaires are tailored for each mode and explored through combined cognitive and usability testing. When designing for multiple modes, we always design the online mode first and then move on to develop other modes. We do this because if a design can be easily used and understood in a self-complete mode, then it stands a high chance of being easily used and understood in an interviewer-led mode. We will continue our testing to explore how the sex standard may be tailored for other modes of data collection (for example, paper, telephone, or face-to-face), and to data collection in Welsh. This may result in the inclusion of alternate variants of the sex question specific to the requirements of different modes or languages. This is an approach we have used previously in other harmonised standards, for example presenting alternate variants of questions which use show-cards for face-to-face interviewing.

Our respondent testing for questions on sex was accelerated to meet priority user needs and was conducted over two iterations, as set out in our supporting research publication. As a result of this accelerated timetable for respondent testing, there are smaller population groups within the UK which we were unable to bring into our research sample. We have identified the need for additional follow-on research among populations who may require variation in questions to support their understanding of the question target, such as individuals:

• with a low English language proficiency
• with learning difficulties
• with variations in sex characteristics (VSC, sometimes referred to as intersex, or differences of sex development), who may have corrected their birth certificate or be considering doing so
• who have migrated to the UK from countries without comparable birth registration and certification processes
• who are proxy respondents providing answers on behalf of someone else

Data collected between 2014 and 2023 indicates that 77% of children worldwide have their births registered, but that coverage of birth registration varies by region. This is important to the wording of the harmonised standard on sex, since both questions within this standard refer to the sex that was, or is, recorded on the respondent’s birth certificate. We will therefore continue to conduct further research with particular focus on populations falling into these groups.

We will also continue to research user needs for data and questions on other topics related to sex, such as variations in sex characteristics (VSC, sometimes referred to as intersex, or differences of sex development). There is currently no consensus on where or how to collect data on populations with VSC, and so the GSS Harmonisation team expect to carry out further in-depth qualitative research in this area, exploring terminology, data risks and user needs.

While our research supports the use of the developed questions by GSS data collectors, we recognise that user needs and the context for official data collection continues to evolve. Therefore, our guidance may need to change further as circumstances change and to reflect further research with groups that we were unable to bring into our research sample. Our work to develop a harmonised standard for gender identity may also affect our future guidance on sex. For example, this guidance may be affected if additional advice is needed for statistical producers intending to present both sex and gender identity questions in the same data collection instrument. Our workplan sets out how and when we will work towards respondent testing for gender identity questions, including the implications for collecting both sex and gender identity together.

Contact us

If you would like more information about the sex harmonised standard, you can contact the GSS Harmonisation Team by emailing Harmonisation@statistics.gov.uk.

Find out more about harmonisation.