Radical is a civil-rights campaign for truth and freedom on matters of sex and ‘gender’, committed to free expression and equal respect, founded by Rebecca Lowe and Victoria Hewson. This Radical piece is written by Rebecca, the former director of FREER.

Have you been following the census ‘sex question’ fiasco? If not, then you’ve been missing out on one of the most important political stories of recent years — and my use of the term ‘political’ signifies a justification for this grandiose claim.

The census, after all, should be an apolitical matter, you might think. Surely, you’d say, it’s a matter reserved for data geeks analysing information collected directly from the person in the street (or wherever the person happens to be on Census Day). But, aside from the fact that, within a political society, no matter of public business is fully apolitical, this year’s census has become very much politicised. And this is something that should concern us all.

At Radical, we first wrote at length on the topic of the ‘sex question’ back in October, when we issued the following warning:

“Gender ideology has penetrated our institutions so deeply that, even without self-identification becoming a matter of law, the insidious idea that one’s sex is solely a matter of personal demand is seeping into policy and practice, almost unnoticed. Yet the damaging effects of this will be far-reaching, and one of the most worrying examples regards the case of the upcoming census.”

Back then, we reported that social scientists and statisticians were concerned about the guidance that was being produced to accompany the 2021 census. In a Sunday Times letter, almost 100 academics noted that this guidance would “effectively transform the sex question into one about gender identity”; they highlighted their fear that this would “undermine data reliability on a key demographic variable”.

As one social scientist, whom I spoke with this week, emphasises:

“Political changes to the census do not provide evidence for the claims activists wish to make, but they do impact on the quality of the research tools depended on by genuine social scientists. Changing census questions reduces its comparability over time, for instance. It should also be pointed out that advocacy groups use statistics, often misuse them, in order to justify their own incomes.”

Some important progress has been made on the sex question, however. Following a recent successful legal challenge by the campaign organisation Fair Play for Women (FPW), the Office for National Statistics, which is in charge of the census in England and Wales, has, as reported by FPW, now ‘conceded that the proper meaning of Sex in the Census means sex as recognised by law’.

An urgent change, to this effect, has been made to the guidance that the ONS had already issued. However, many questions remain. Here are three, to begin with:

1) How did the ONS settle on its initial 2021 guidance?

This is a question not only about the particular wording of the guidance, but also about how that wording was determined. As we discussed in October, questions remain about the formation of the ONS’s 2011 guidance on the sex question.

The approach that the ONS took appears not to have been subject to consultation; according to the ONS, it was done “at the request of the LGBT community”. Of course, input to the deliberative process of public business should always be welcome, but consultation should be wide-ranging, and the opportunity to respond should be open. Contentious questions of sex and gender are not the preserve of groups whose work is particularly focused on these matters — they are of relevance and importance to everyone.

The policy analysis organisation MBM has raised similar questions about the approach of the National Records of Scotland to the upcoming Scottish census (which will take place in 2022), reporting that

[d]uring the question development phase for the sex question in the [2022] census, NRS met only with LGBT advocacy bodies. There is no evidence of consultation with independent statisticians or census data users in this period (see FOI correspondence)”. 

Conflicting statements and rapid shifts of approach from the ONS on their 2021 sex question guidance haven’t helped, either. Having confirmed in January that the question would relate ‘very simply to your legal sex’, it became clear the following month that one’s passport would be included in the exemplar documents, listed in the guidance, which respondents would be recommended to use to inform their answer.

And, as FPW emphasised, passports do not record legal sex — in that you can change the sex recorded on your passport, as you see fit, without having obtained a Gender Recognition Certificate (GRC). Now, why did the ONS change its approach? And what is being done to address serious concerns about institutional capture?

2) What is the purpose of the sex question?

While it is good news that the ONS has corrected its guidance, it should be noted that one’s legal sex and one’s biological sex are not identical.

The 2004 Gender Recognition Act (GRA) allows people, on having met certain conditions, to obtain a GRC, which entitles them to be treated as belonging to the opposite sex set for (almost all) legal purposes. This does not mean, however, that it is possible to change one’s biological sex. It is unfashionable (to put it mildly) to state this so baldly, but it is the truth, nonetheless.

And if it is the case that reliable census data is required regarding the membership of the two biological sex sets, not least for the purposes of UK public policy — including the allocation of vital medical resources — then being asked to record one’s legal sex, rather than one’s biological sex, is, at least superficially, a peculiar choice.

Now, this is not to suggest that legal sex, or other descriptors, should not also be recorded. Indeed, this year, the ONS has also included an optional census question pertaining to gender identity, and there are good reasons for including such questions. But conflating these matters is the kind of bad practice that leads to unfortunate unintended consequences — in this case, not least regarding public respect for biological truth. And, again, the process for determining what questions are asked in the census, and how these questions are worded, should be transparent and open.

3) What about other forms of data collection?

As we’ve discussed previously, census data is extremely important. But the census is neither the only means of collecting public data, nor arguably the most important for some of the kinds of public-policy-making typically referred to in discussion of these matters.

Rather, that accolade could be afforded to the Labour Force Survey (LFS), which is also run by the ONS. This is because the census, whilst crucial for geographical and long-term historical purposes, goes out of date extremely quickly, and isn’t as helpful for analysing short-term trends over time, or addressing pressing matters such as evidencing discrimination patterns. This is not to undermine the census’s importance.

But the LFS reports quarterly, and, as the ONS explains, ‘provides data at a level of precision not matched by any other regular household survey’. Now, it’s unsurprising that there are many differences between the ways in which the LFS and the census are designed and carried out. But when the ONS makes significant changes to census categories, it’s naive to think this won’t have a significant impact on its other forms of data collection. So, what might this year’s census debacle mean for the LFS, and for vital work depending on reliable data sets, more generally?