Opening Statement, Professor Alice Sullivan

Equalities, Human Rights and Civil Justice Committee, Gender Recognition Reform (Scotland) Bill

Date: Tuesday, June 21, 2022, Session: 6

Written evidence submission.

Sex is a fundamental demographic variable and a powerful predictor of almost every dimension of social life.

Sex is a protected characteristic under Equality Act 2010, and is essential for equalities monitoring. EA2010 includes a public sector equality duty to monitor and publish data on the protected characteristic of sex.

We need consistent and accurate data on sex, to make comparisons over time and between countries, and to evaluate the effect of policy interventions.

Sex is not the same thing as gender identity. We need data on both of these variables.

I am a quantitative social scientist, with over twenty years of experience in the collection and analysis of population data. Between 2010 and 2020, I was Director of the 1970 British Cohort Study, often described as one of the jewels in the crown of British social science.

My views on the importance of sex as a basic variable which needs to be collected in both administrative and survey data are not controversial among quantitative social scientists.

In 2019, 80 quantitative social scientists signed a letter I co-ordinated  to the UK Census Authorities expressing concern regarding their plans to advise respondents that they may respond to the sex question in the census in terms of their self-identified gender identity. The signatories included demographers, epidemiologists, statisticians, sociologists and economists. They included ten Fellows of the British Academy and several past and present leaders of major datasets.

In 2021, Scotland’s Chief Statistician released draft guidance stating that that public bodies should not routinely collect data on sex.  91 eminent quantitative social scientists signed a letter I co-ordinated objecting to this guidance. These objections were ignored. The final guidance states that data on biological sex should only be recorded only in “a small number of instances”.

In contrast, the UK Statistics Authority guidance  recommends that “sex, age and ethnic group should be routinely collected and reported in all administrative data and in-service process data, and should clearly distinguish between the concepts of sex, gender and gender identity.

One might assume that there is no reason why the legislation we are considering would affect data collection. This would be an understandable assumption. But it would be a mistake.

When the 2004 Gender Recognition Act was introduced, it was designed to cater to a tiny number of transexual people suffering severe psychological distress. The act was not intended to raise barriers to the collection of data on sex. And yet this has been one of the unintended consequences of the legislation. Public bodies have taken the message from the wording of section 9 that people can switch legal sex over their lifetime not just for certain legal purposes, but completely. Section 22 of the Act has contributed to over-caution about collecting data on sex.

Legislators must give serious consideration to the likelihood that introducing gender self-declaration in law will reinforce an existing reluctance to ask about sex.

Even small numbers of cases where sex is misclassified can introduce substantial error in data analysis, particularly in cases where sex differentials are large, such as crime statistics.

And the number of trans identified people has risen rapidly, particularly among youth, especially girls. According to a large US study, one in 2,000 female undergraduates identified as transgender in 2008. This rose to five percent in 2021. The experiences of these young people are so important. We need accurate data on sex as well as gender identity to understand their outcomes, both for those who medically transition and the majority who don’t. Yet even within the health system, this data is being erased.  For example, the wellbeing of children transitioning at the GIDS service has not been followed up over time because their NHS number changes post-transition.

Where organisations adopt legal sex as a “safer” category to ask for than sex itself, as happened in the census in England and Wales, larger numbers of people with GRCs under self-declaration will imply a larger impact on data collection and analysis.

Finally, without data on sex as well as gender identity we will not be able to evaluate the impact of the proposed reform. Serious policy evaluation will be rendered unfeasible.

Leave a comment