Census (Amendment) (Scotland) Bill: data user views

Background
The Census (Amendment) (Scotland) Bill has been introduced to change the law so that the 2021 census in Scotland can include new, voluntary questions to gather information about sexual orientation and, separately, about the transgender population.

The Policy Memorandum to the Bill also stated that Scottish Government planned to introduce a third option in the separate question for “sex”, to allow those identifying as neither male nor female to provide an alternative answer.  The government has since further stated that any non M/F answers will be randomly assigned to one of those two  categories before data are made available for use, although there is the possibility of a national summary of the total number answering in that way. The government has also stated its intention that, more generally, the M/F categories should be used to return self-identified rather than legal sex, where these conflict.  This means that Scottish Ministers wish to use a different interpretation of sex than is in current law (mainly the Equality Act 2010).

This is controversial because it would mean that while the 2021 census would have a question termed ‘sex’, it would not produce data on sex on the usual legal or scientific definition for the whole population. Instead, for an unknown number of people, it would gather a different class of information (gender identity) which may differ from their sex as defined in law. This would also set a precedent for the interpretation of sex in primary legislation which conflicts with that used in the 2010 Act and elsewhere. Advice was issued for the 2011 census in Scotland that individuals could answer M or F on the basis of self-identification but this was not on the face of the form itself and it is not clear how widely it was seen or taken up (further information on that is included here.)

The Stage One report on the Bill argued against the conflation of gender identity and sex, and recommended that the sex question remain binary and that data on gender identity should be gathered, but separately. The Scottish Government has since stated it does not intend to conflate sex and gender identity but has not ruled out a third option under the question on sex, and an explicit shift to asking for sex more generally based on self-identity. The final format will be decided over the next 12 months or so. At the Stage One debate on the Bill, Culture Secretary Fiona Hyslop stated that “the specific census questions are a work in progress”.

Evidence submitted to the Committee to date can be accessed here, and the Official Report from the Stage One debate is here.

Purpose of the census and data quality 
The purpose of the census is to provide robust, detailed evidence to develop policy, plan and run public services, and allocate money to public authorities. Census data are also widely used by academics, businesses, voluntary organisations and the public.

There has been disagreement as to whether changing from a binary sex question to one including gender identity would affect the quality of data collected for analytical purposes. Some evidence submitted to the Committee suggested it would make no difference or even improve data quality, while others argued that it would affect data reliability, including data continuity, and that the effects being unpredictable, particularly at sub-population level,  was itself a problem: see various contributions here.  (We have argued for the second of these positions.)

The UK government currently estimates that the total number of people in the UK with a transgender identity of some type is between 200,000 and 500,000, but is not confident about the accuracy of this estimate, and does not appear able to break it down further by any other characteristic. 

For the small population who already have a gender recognition certificate (GRC), of whom there are just under 5,000 across the UK, there may be specific legal issues of confidentiality, and for the purpose of this discussion it is assumed that, whatever was done,  they would be able to respond with whichever sex their GRC confers in law.

Assessing impact on data users  
To better understand the implications of changing the sex question, we would like to hear from service planners, researchers and analysts who use the census sex variable as part of their work. This includes analysis that uses sex as a single factor, or multivariate analysis, which combine sex with other variables such age, health, ethnicity, education level, employment status and so on.

We would like to know whether data users think that changing the question format to a combined sex/gender identity response would impact on their analysis, and if so how. We would also like to collect examples of research or analysis (published or ongoing projects) that uses the sex variable.

This evidence will be used to inform our ongoing work on the Bill (for example, briefing papers and reports), and a research paper that is currently in progress.

We can be contacted using the form below. We will only reference submissions where the ‘permission to publish’ box is checked. If you are happy for us to cite your evidence but wish to remain anonymous, please check the ‘anonymous’ box.

Many thanks in advance.

Kath, Lucy and Lisa