As genealogists we know that the historic records we rely upon were not created with us in mind. Always, there was a significant contemporary reason behind the collection of data. The list of heads of household may have been compiled to keep track of who had paid taxes, or rentals. The list of adult males may have been required for selection of local militia. Parish registers, while being a record of religious rites, were also a means of keeping track of the population, and so on.

When the first censuses were carried out in 1801-1831 the purpose was to collect basic information about numbers of inhabited/ uninhabited houses, numbers of families occupying them, numbers of people (male/female) with a very basic breakdown of their occupations, and information about numbers of baptisms, marriages and burials. It was not until 1841 that names, ages and occupations of individual household members were collected, along with a rough indication of birthplace. The real need at that time was to get a handle on the size of the population (for military purposes and food requirements) and to understand migration within the country. Since then, with every new decade, changing societal conditions and increased understanding of links between, for example, living conditions and health, led to additional questions. In more recent decades, there has been a need to understand religious diversity, language needs as a result of immigration, gender identity, and so on. This then feeds into planning of services and facilities.

However, there has been another huge change in the past thirty years: information technology. Such is the ‘ready’ availability of timely data that the need for a decennial snapshot of the nation is in question. In June of this year the Office for National Statistics (ONS) produced a consultation paper on its proposals for The future of population and migration statistics in England and Wales.

In it, they explain their proposals to replace reliance on a decennial census, creating information that is essentially out of date as soon as it is published by instead using administrative data collected by a range of state organisations as a by-product of their work. This, they say, would enable the ONS to provide completely accurate, up-to-date information, when needed, and at a fraction of the cost. It is estimated that the cost of the 2021 Census was around £1 billion.

In response to these proposals a group of leading academics whose work draws heavily on data provided by the censuses have published an Open Letter in which they express their concerns.

Whereas we, as genealogists, are interested in personal information – and we’re not permitted to access that personal information until one hundred years have lapsed – impersonal statistical information from each census is released very soon after the information is collected. Already, a good deal of data is available following the 2021 census, and undoubtedly this is the data essential to the work of the sociologists, medical sociologists, criminologists, epidemiologists, social policy, population and health experts who signed the Open Letter.

Reading their letter, it is interesting to note that, despite the differences in our use of the census data, they broadly use the census in a very similar way to us: as a decennial benchmark against which the representativeness of other data may be assessed. For them, that would be statistics and other studies; for us it is Civil BMDs or parish register entries. Administrative data produced as a by-product of a service, they say, cannot fulfil this function. Although the census completion is not actually 100 percent, it is a legal requirement and completion is in fact very high. By contrast there is no mechanism for administrative data owners to be held accountable for the quality of their data. There are also opt-outs, not to mention the fact that some people are not even registered with the services in the first place and would therefore slip through the net. Their suggestion is that if the alternative ‘patchwork of administrative data’ is seriously to be considered as a replacement for the census, then at least in 2031 the two systems should be run in tandem, to ensure complete coverage and a smooth handover. They also assess that the cost of collection of data from these other sources and bringing them together as a seamless whole would be about the same cost as the decennial census.

Essentially, of course, our interest is in the genealogists of 2131 and beyond. It is with them in mind, as well as my own privacy concerns, that I am worried about the future of family history. The contents of our medical records and other information collected or created about us in the operation of public services should remain private – but a census is different. Whilst fully mindful of the real, administrative benefits to the nation of collecting all that information, on a personal level I enjoy sharing my information (my choice what to share) on the census. I think of it as a message to my descendants. In Ireland, last year, the 2022 Census even had space for a Time Capsule Message to be available to descendants in 2122. How wonderful would it be to open that!

Anyway… what do you think? Have you been involved in any of the consultation surrounding this? Do you have strong views either way? Do leave a comment.