‘Widespread loss’ of sex data in gender question confusion

Inconsistent approaches to how sex is recorded in publicly funded research has led to a “widespread loss” of data over the past decade, according to a government-commissioned study that has urged UK public bodies to collect information on both sex and gender identity.

The independent review, led by UCL sociology professor Alice Sullivan, recommends that the “default target of any sex question should be sex – in other words, biological sex, natal sex, sex at birth”, with researchers asked to list only “male” and “female” among the possible responses provided to respondents. The option of a non-response such as “prefer not to say” could be included, if appropriate.

Researchers should also collect data on gender identity but avoid combining questions on sex, adds the study, which recommends clarity in the phrasing of questions. “Do you consider yourself to be transgender?” is suggested as a potential format for such a question.

The phrase “sex assigned at birth” should not be used in data collection, the study continues, stating that this is “inaccurate and misleading”, implying that sex is “merely an assigned label rather than an inborn characteristic”. Instead, it suggests asking about “your sex at birth.”

The 232-page study, published by the Department for Science, Innovation and Technology on 19 March, was commissioned in February 2024 by then-science secretary Michelle Donelan, who had used a Conservative Party conference speech to criticise the rise of “woke ideology” in science four months earlier.

The Cabinet minister said it was important to “[safeguard] scientific research from the denial of biology and the steady creep of political correctness”, claiming that “scientists [have been] told by university bureaucrats that they cannot ask legitimate research questions about biological sex.”

The review by Sullivan, a prominent gender-critical academic, highlights numerous examples of where the “instability” over the meaning of sex, as collected by researchers, has led to a “widespread loss of data”.

That loss of data “poses a risk of individuals” in some cases, stating such risks are “especially high in the case of minors”.

More broadly, “how sex is defined is rarely made apparent in published outputs, including accredited statistics”, while some publications that present binary data on sex include additional responses from those using options such as “other”, explains the report.

The review explains how the word “gender” began to replace the word “sex” in data collection in the 1990s – albeit the two terms were broadly interchangeable – but, from about 2015, gender started to be understood in terms of gender identity and not binary sex.

That has meant robust and accurate data on biological sex was lost from some government and public body survey data, says the review, which found that about two-thirds of reports held by the UK Data Service, a national collection of social sciences studies, conducted since 2020, referred to gender or gender identity, compared with about 5 per cent of studies from between 2010-2014.

That practice did not meet the basic data collection standards given the need to distinguish between biological sex and gender identity in data collection, said Sullivan, who is head of research at UCL Social Research Institute.

“Rather than removing data on sex, government and other data owners should collect data on both sex and transgender and gender-diverse identities,” said Sullivan.

“This will help develop a better understanding of the influence of both factors and the intersection between them, and this is crucial for research and policymaking.”

The study also has wide-ranging recommendations for government and UK public bodies, research journals, research funders and polling firms, noting how inconsistent practices used by different bodies in relation to gender recording had the potential to undermine trust in research.

Among the recommendations is the need for “clear language in legislation, guidance and discourse”.

For government officials, for instance, given how gender, “previously a polite synonym for sex”, now has “multiple distinct meanings” that are “open to misinterpretation”, legislation should “refer clearly to sex and/or to gender reassignment as appropriate rather than using the term ‘gender’”, it says.

Guidance for employers, for example, on the gender pay gap should refer to the “sex pay gap”, it recommends.

Opinion polling should also consider whether the use of “language which is familiar to some groups of respondents may be unfamiliar or misunderstood by others”. Questions around whether transgender athletes should be allowed to compete in female sports categories should ask whether “males who identify as women be allowed to compete in female sports categories”, the study says.

Ministers should “consider the vulnerability of government and public bodies to internal activism that seeks to influence outward-facing policy, including through staff networks, and whether stronger safeguards are needed”, says the study on concerns that official statistics could be swayed by campaign groups.

jack.grove@timeshighereducation.com