The debate over the definition of ‘Sex’ within the Census continues or more accurately, the debate over the accompanying guidance. This guidance explains that the person can select “whichever answer best describes your sex” ONS or “the answer you give can be different from what is on your birth certificate” NRS , thereby transforming the question to one not of sex but to one of Gender Identity, or as the NRS describes it Self Identified Sex.

Thinking about how, and why, a vitally important characteristic ‘sex’ has been able to be redefined I was reminded that this process was initially driven by the need to collect statistics on the people with the protected characteristic Gender Reassignment as defined with the Equality Act 2010., for equality monitoring purposes.  It’s also worth remembering that the Equality Act didn’t come into effect until October 2010, only 5 months before the 2011 Census.  With the Equality Act the Public Sector Equality Duty was also introduced which meant for the first time public bodies were required to have due regard to the need to eliminate unlawful discrimination, advance equality of opportunity, and foster good relations across all aspects of equality protected by the Act.  Thus, the importance of collecting and monitoring equality data on characteristics not previously included i.e. Sexual Orientation and Gender Reassignment, was then established.  The background of the legal requirements introduced after the 2011 Census diminishes the current position of the ONS and NRS, that is that an online FAQ response for transsexuals on how to answer the ‘sex’ question had established a precedent for every Census thereafter.  It didn’t and it wasn’t until about 6 years later that serious consideration was given to including such a question within the 2021 Census.

Yet, discussions on how to capture data of the trans population in surveys had been going on for a lot longer, even before the anticipated duties of the Equality Act before had been finalised. Around the same time, trans groups were campaigning for the definition of Gender Reassignment within the Equality Bill to be a much broader umbrella term, ranging from cross-dressers to transsexuals, and other gender variant people.

The ONS Trans Data Position Paper 2009 highlighted that the UK Government were very clear that the Equality Act should not include the wider group of people who describe themselves as trans, and that it was to be strictly restricted to transsexual people. The Government’s response to the consultation stated:

“We are clear that transsexual people must be protected against discrimination arising from their gender reassignment, whether or not they undergo medical supervision. However, our intention is not to protect a wider group such as transvestite people or others who have no intention or commitment to live life permanently in the sex opposite to their birth sex. Extending protection to perceived gender reassignment would encompass that wider group.”

Adding further to the difficulties of collecting data on this demographic was the realisation that no matter the manner in which questions were asked, the majority of trans people would never disclose they had the protected characteristic of Gender Reassignment.

“in a recent informal poll undertaken by Press for Change among transsexual people, findings showed that 55 per cent of respondents would refuse to answer a question that may lead to their gender history being disclosed under any circumstances”

ONS therefore concluded that:

“data collection via household surveys is not the most appropriate method of meeting these requirements.”

Research continued into how to best capture those with the protected characteristic of gender reassignment as defined in the Equality Act, in particular those who had obtained a Gender Recognition Certificate, but may choose not to disclose their trans history; and also those who identified as trans within the wider definition, but without the protected characteristic of gender reassignment.

This EHRC research on question testing confirmed earlier findings that regardless of the wording there will always be trans people who choose not to answer questions on their trans history.  However, a suite of four questions were recommended as the most effective method of capturing data, and these were able to be further split depending on the scope of the data required, namely:

  1. questions to capture data on transgender history and gender identity
  2. questions that captured data on the legal definition of gender reassignment

Suitable questions had now been developed that could then be recommended for use in data collection surveys; and the ONS had already ruled out the Census as being an appropriate method.

But then that all changed.

In January 2016 the Women and Equalities Committee: Transgender Equality Inquiry  recommended:

”the use of the terms “gender reassignment” and “transsexual” in the Act is outdated and misleading; and may not cover wider members of the trans community. The protected characteristic in respect of trans people under the Equality Act should be amended to that of “gender identity”. This would improve the law by bringing the language in the Act up-to-date, making it compliant with Council of Europe Resolution 2048; and make it significantly clearer that protection is afforded to anyone who might experience discrimination because of their gender identity.”

As a result, further exploration of the inclusion of a Census question on Gender Identity had to once again be considered, taking into account the recommendation of the WESC Committee. A consultation was then held to assess the user requirements for asking a question on Gender Identity – notably not a question on Gender Reassignment.

As could be expected respondents to the consultation referred to gender identity as including everyone who identified as trans irrespective of whether they had the legal definition of gender reassignment.

EHRC’s suite of four questions was considered to be a high burden for the respondents, and in order to lessen this burden the EHRC recommended that “the first would replace the standard mandatory sex question”

It was now, as a result of the WESC recommendation, being seriously proposed that the mandatory sex question could be amended to include responses relating to Gender Identity i.e. Male / Female / Other. 

However, in its paper “Assessment of initial user requirements on content for England and Wales May 2016” the ONS assessed the impact on data quality of including such a Gender Identity question as “High”, highlighting that:

“Sex, as biologically determined, is one of the most frequently used and important characteristics the census collects as it is used in most multivariate analysis of data and feeds into the UK population projections.*

It is critical that the collection of information on gender identity for a small population (estimated to be less than 1%) does not jeopardise the quality of data collected on sex for the population who don’t have trans identities or the protected characteristics of gender reassignment.

Any question on gender identity would need to be understood by the whole population.

Following testing, the EHRC concluded that some of the responses which appear to come from trans people show that this may not be the case; suggesting potential respondent error. For example none of those that stated a birth sex of female and a gender identity of male gave an answer indicating that they had the protected characteristic of gender reassignment.

Therefore, as the population of interest is very small the quality implications may create errors bigger than the population we are trying to estimate.

Inclusion of a question on this topic may discourage some usual residents from participating in the 2021 Census. This is a key concern for ONS. For example, trans people obtaining a Gender Recognition Certificate cannot be required to reveal their birth sex or gender history. The ONS ‘Trans data position paper (May 2009)’ reported that 55 per cent of trans respondents would refuse under any circumstances to answer a question that may lead to their gender history being disclosed”

The ONS emphasised the importance of the sex variable as reflected in the Government Statistical Service policy, that is, when collecting data on the variable ‘sex’ the UK has to match the agreed definitions from the United Nations and the European Statistical System. See here and here for more details.

The conclusion of the ONS 2016 report therefore remained unchanged from their earlier 2009 paper, repeating that:

“data collection via household surveys is not the most appropriate method of meeting these requirements.”

They further stated that:

“the consultation responses were not clear on the exact output categories required. Hence, the exact concepts to be measured need to be clarified Additionally, other considerations regarding data quality, public acceptability and respondent burden are ‘high’….

… there are specific legislative issues linked with collecting data on gender reassignment, and it is possible that the inclusion of a question on gender reassignment in the census would require an amendment to primary legislation.”

The Report explained the reasons why the ONS held this view:

“The Equality Act 2010 includes gender reassignment as a protected characteristic but does not include those with other trans identities such as those with a non-binary gender identity. As such, any development of questions related to gender identity, intended to collect data on those with the protected characteristic of gender reassignment should ensure that this sub-group can be identified separately from the wider trans community.

In addition, the collection of data on this protected characteristic should not have a detrimental impact on the collection of other protected characteristics, such as sex.”

The publication of this report by the ONS was also timed with the publication of a plan on how they would take this work forward, Gender Identity Research and testing plan May 2016, It included engaging with Stakeholders and referring to “changes in UK legislation, such as the Equality Act 2010, that relate to gender identity.”.  Reading other, more easily found, reports on this topic have all significantly downplayed, or just outright omitted to mention these ONS findings. It is clear that this was progressed against the advice of the very authority whose expertise should have been heeded.

We can not underestimate the extent and reach of the recommendations from the WESC Transgender Inquiry.  Most bizarrely, it didn’t seem to matter to anyone that the Equality Act 2010 had remained unchanged from its inception only 5 years earlier, yet this new concept of Gender Identity would continue on and take over almost every sector across the UK, trashing the definitions of both protected groups ‘Sex’ and ‘Gender Reassignment’ along the way.

In normal times I don’t think anyone would have found the conclusions by the ONS in their 2016 report to be at all controversial.  They made it clear what their legal requirements were in respect of collecting data, the difficulties in the various interpretations of ‘trans’ and the impact such a question would have on the protected characteristic ‘sex’. 

Countless women have since went on to make those very same points, and as a result we have suffered abuse, harassment and other forms of discrimination, with no support or even condemnation of that abuse from any organisation or public body.

Yet, here we are.

I would like to know how a tiny demographic have managed to break through all these sectors and make changes on such a grand scale that no other protected group could ever have imagined possible, and in rapid time too. Who would have the power to make that happen?

Remembering that this had started way back in 2009 when Press for Change said “that 55 per cent of respondents would refuse to answer a question that may lead to their gender history being disclosed under any circumstances”.

There has never been a demand for a question to record a persons transgender history.

The recent demand was to redefine sex to include Gender Identity, and this is what has been happening everywhere else for years.

We now have a sex question in the Census that no longer asks about the characteristic ‘sex’, having now been replaced with gender identity.  

WE know it’s a gender identity question and THEY know it too. 

After more than 10 years of question testing, consultations and workshops, are the ONS and NRS confident that the proposed Census question will capture those with the protected characteristic Gender Reassignment?

I doubt they are. I also wonder if those responsible will feel that all this effort, and the wider impacts, will have been worth it.