As the registrar general, I am responsible for taking the census, so as there was some commentary on it, I thought it would be appropriate for me to note a few points.
The first point is about legislation. To be clear, in 2011, the whole of the UK used a self-identified sex question, and it was the intention of the Office for National Statistics to do the same as it went into its 2021 census. That was subject to a court challenge, and an interim decision in the High Court of England and Wales ruled against its approach, so the ONS took the decision not to contest that judgment and instead to amend its guidance.
As we are talking about data, it is important to note that that came into effect after ONS had received millions of returns. That change in the guidance happened after quite a large proportion of the population in England and Wales had already submitted their returns using the existing guidance.
In Scotland, the case was subject to both a judicial review and an appeal, and in both of those cases our guidance was found to be compatible with the legislation. The judges who were involved in those decisions noted that there was a wide variety of use and interpretation of sex and gender across a range of different pieces of legislation. To be clear, there is a clear legislative position in Scotland around the issue as it relates to the census, and we acted compatibly with the legislation. Obviously, I cannot comment on a decision made in a court in another part of the UK, based on the information that it had, but it was an interim proceeding, not a full judicial review outcome, in that case.
In terms of the comments that have been made about the data, I am not going to pretend that I am happy that I have not got a 90-plus per cent return rate in the census. However, there seems to be great weight of interpretation that because we have not achieved that the census is in some way a shambles—that is one word that I have heard among various others. That is not the case. Our organisation is the census-taking expert; we have been doing it for more than 200 years. We can produce a good-quality census outcome with the 87-plus per cent return rate that we have received now.
I appreciate that this has been a source of speculation, so I commissioned an international panel of experts chaired by Professor James Brown from Sydney, who is a professor in official statistics, and including the UK national statistician, Sir Ian Diamond. Last week, they issued a statement saying that the census had achieved a solid foundation and that it was appropriate to move on to the next stages.
Modern censuses are not just about the collection of data. That is how it used to be done a long time ago, but for the past 30 years we have been using census collection, something called a census coverage survey, which we are about to start running in Scotland, and a range of other statistical techniques to produce high-quality outcomes. We will use those, advised by the international panel that I have convened, to ensure that we provide high-quality census outputs.
I thought that it would be helpful to clarify those points, because they have come up in discussion.