How to work out whether census records are missing

What would we do without census returns?

How could we even begin to research our 19th and early 20th-century English and Welsh and indeed Scottish ancestry without access to that remarkable resource?

Where else would we find records that list our forebears in neat, multi-generational family packages, along with their names, ages, relationships and birthplaces?

We sometimes need to remind ourselves that the censuses weren’t actually taken with us in mind; that they had a different purpose which was all about gathering statistical information that would help successive governments to understand the people they were governing and to plan for future services.

The census returns provide us with regular snapshots of our ancestors, allowing us to track their movements across the decades and giving us invaluable clues about family births, deaths and marriages that we can then follow up in other records. But we do need to remember that they are just snapshots; we see our families as they were once every ten years (or approximately 3,650 days). The nephew or niece who’s listed in the 1871 Census may have arrived with the family a few days before the census was taken and returned home early the following week; they may not have been a regular fixture in the household.

[This guide concentrates largely on the census records of England and Wales. For details of missing census records for England, Wales and Scotland, see the Findmypast guide. For information about Irish census records, see our guide to Irish research]

The returns from no fewer than nine decennial censuses are now available to us, presenting an unparalleled resource for family and local historians. And it would be nice to report that the surviving census returns for England and Wales taken between 1841 and 1921 provide a comprehensive record of those who were enumerated at the time. Unfortunately, however, there are some gaps in the records and it’s important that we’re aware of them when we’re carrying out our research.

Why are some English & Welsh census records missing?

I’m not talking about those urban myths concerning an ancestor who hid under the stairs when the census enumerator came calling. For a start, stories like that show a complete misunderstanding of the process behind the English and Welsh censuses: the enumerator’s role was to deliver and then collect the schedules, not to stand on the doorstep asking questions of the householders. Instead, the schedules were usually completed by a member of the household and were, in the normal course of events, waiting, ready for the enumerator’s knock on the door on the Monday morning.

The chances of a house or an entire road being missed are also very slim. Many of the same people were involved in carrying out successive censuses and the system put in place by the General Register Office was fine tuned and highly efficient. The use of local people to do the legwork was a great way of ensuring as close to 100% coverage as possible. That’s not to say that some people weren’t accidentally omitted but this was probably most likely to come about as a result of clerical errors made by the enumerators when copying the details from the householders’ schedules into their summary books. The schedules for the years 1841 to 1901 were later destroyed so in those cases we only have the summary books to go by. In 1911 and 1921 it is, of course, the householders’ schedules that we see.

So while there must always have been some degree of under-enumeration, what’s more of a concern to us is those sections of the census that, for one reason or another, have become lost over the years. Again, it’s not a huge problem but if your ancestors were listed in a part of the census that is now missing there’s not a lot of point in continuing to search for them. 
 

What’s the extent of the loss of the census records for England and Wales?

But how much of a problem is it and how concerned should we be? The good news is that, with the exception of 1841 (approximately 1.5% missing) and 1861 (approximately 2.5%), we’re talking about less than 1% of the returns each year. The 1861 Census is known to be the worst affected but it’s better than it might have been: the summary books for the whole of the 1861 Census were thought to have been lost before they were discovered in 1904 in an attic room in the Houses of Parliament! 

In the early census years there are some quite significant gaps. The 1841 Census returns for the Middlesex parish of Paddington have not survived and the 1851 returns for the whole of the Dunmow (Essex) registration district are missing, as are big chunks of Islington (Middlesex), Woolwich (Kent) and Ely (Cambridgeshire) in 1861. The 1861 returns for the sub-district of Belgravia (Westminster) are believed to have been stolen during the time that they were stored in the Houses 
of Parliament.

Wales is disproportionately affected across the 1841, 1851 and 1861 censuses but other than a small part of Fife in 1841, the Scottish census returns are virtually complete. From 1871 onwards, there are very few significant gaps in any of the censuses, the main exception being the curious loss of the 1901 Census for the town of Deal in Kent.
These large gaps are most likely to be the result of damage – storage conditions in the 19th century left a lot to be desired and a large section of the 1851 Census for Manchester and the surrounding area was badly damaged by water. Thanks to the efforts of the Manchester and Lancashire Family History Society these returns have now been digitally recovered and transcribed.

Discovering that a part of the census on which your ancestors would have appeared is missing is, naturally, very frustrating. A missing census doubles the gap between an ancestors’ appearance in the returns from ten to twenty years (or from around 3,650 days to approximately 7,300 days) potentially leaving a huge hole in the story of their life. But it’s not always instantly obvious that the reason that you’re unable to find someone in the census is because the relevant returns are missing.
 

How to identify the gaps in the census records

There’s a very useful (although certainly not comprehensive) list of missing sections of the census covering England, Wales and Scotland on Findmypast.

You can also search the National Archives catalogue using the search term ‘missing’, combined with the reference for a particular census year, to find out which sections are known to be missing (1841 and 1851, HO 107; 1861, RG 9; 1871, RG 10; 1881, RG 11; 1891, RG 12; 1901, RG 13; 1911, RG 14; & 1921, RG 15).

But, with the exception of the 1861 Census, on which extensive work has been carried out over the years (particularly with the returns for London), finding information about the odd missing page or two is not easy. This is where a bit of local knowledge comes in handy, along with a whole lot of evidence-based research, using a variety of sources such as trade directories, electoral registers, rate books, birth, marriage and death certificates, parish registers, wills and newspapers (to name but a few).

Before the days of blanket online coverage of fully indexed census returns, the only way to find an ancestor in the records was to carry out a physical search. And in order to find them, you really had to have a good idea of where they were living in the first place. Yes, there were some name indexes, produced by family and local history societies but in most cases you had to use paper lists to identify the microfilm which covered the part of the country that they were likely to be living in. There were street indexes for the larger towns and cities but if your ancestors were living in a small town or a more rural area you might have a long search on your hands.

You might have an address from a birth or death certificate close to the date of one of the decennial censuses which told you where the family were living at the time but if not, you might have to trawl through the returns for a whole parish or a small town, hoping that your ancestors would jump out at you. 

It wasn’t a quick process but like so many things that were done ‘by hand’ in the past, it taught you a lot about the records that you were looking at. And it’s these skills that we need to employ, and these sources that we need to consult, if we want to attempt to explain why we’re unable to find our missing ancestors in the census returns in today’s digital world.

The old paper-based street indexes are still available if you’re able to get to the National Archives in Kew but digitised versions can be accessed online (for 1841, 1851, 1861, 1871 and 1891) on the archived ‘Historical Streets Project’.

Findmypast offer an address search  that can help to pinpoint the returns for a particular address, although the house numbers haven’t always been captured so you may still have a bit of a search to do.

Case study: How to proceed when census records are missing

Let’s have a look at how it works in practice.

A few months ago, I was looking for a family in 1891 who I was sure would have been living in the Edmonton (Middlesex) area but my searches on Ancestry, Findmypast and TheGenealogist had all failed to reveal the family’s whereabouts. I had found the Bowes family living at 68 Brettenham Avenue, Edmonton in 1901 and I suspected that they wouldn’t have been too far away in 1891. I noticed that a child had been born around 1891 and her birth certificate confirmed my suspicions: she was born on 5 December 1890 at 68 Brettenham Avenue. So, why wasn’t I finding the family in the 1891 Census?

The online street indexes and the Findmypast address search both led me to the section of the 1891 Census that included the returns for Brettenham Avenue, and after a little bit of browsing I soon found the problem. The last couple of pages (which I could work out were numbered 51 and 52) were badly damaged. It’s often the first or the last pages in the books that are missing or damaged, reflecting the wear and tear inflicted on them over the years.
The addresses on page 51 were missing but I could see that the returns for 46, 48 and 50 Brettenham Avenue were included on page 50. Scrolling back to the start of the census enumeration book I was able to confirm from the ‘Description of Enumeration District’ page that the whole of Brettenham Avenue was included in this district and I also noted that it was the last road named in the ‘Contents’ section. Turning to the fifth of the introductory pages (headed ‘Abstract of the Totals on the Following Pages’) I could quickly see that this enumeration district had originally comprised 55 pages and that four pages (i.e. two sheets of paper) were therefore missing. The returns for 68 Brettenham Avenue and the Bowes family were undoubtedly on one of these pages. I hadn’t found the family but at least I knew that I could stop looking for them.
The figures on the abstract page tell us that 98 names were recorded on the four missing pages, a small number in the greater scheme of things but if your ancestor was amongst those missing you’d clearly want to know.