To view this media, you will require Adobe Flash 9 or higher and must have Javascript enabled.

Duration 00:50:06

Webinar: Missing from the census?

Census records are invaluable sources of information for all family historians. But what if you can’t find your ancestor on the census? This webinar will look at useful search tips that will help you to locate hard-to-find people, as well as exploring possible reasons for their absence and providing pointers for where else you could look.


Hello. This webinar is all about missing from the census. How to find people who are missing from the census or that you think are missing from the census.

‘They’re not in the census!’ It’s a familiar cry from the genealogist. But are the people you’re looking for really missing or are they just hard to find? The information that is given in the census might not be completely accurate, and some people completely misunderstood the question and gave them very peculiar answers. So you might be able to find people that you think are missing because they’re not all but sometimes they might be genuinely missing, in which case I do have a few suggestions on how you might find the information that would be in the census if they were in there to be found.

The first thing to remember is less is more. When you’re searching online, which is how most of us look at the census these days, you’ll see very inviting looking boxes to fill in lots of detail. You should resist the temptation to fill in all these boxes or even half of them. You’re looking here at main search boxes on Find My Past and on Ancestry. Those are our two partner sites for the census records. You will find the census on other sites as well but the same kind of rules apply.

If you are speaking to a human and you’re asking them to help you find some information, the more you tell them, the more it helps; the more information they’ve got so they can work out what the answer might be. But when you’re looking at the database like this, it doesn’t work like that. If you’re giving a lot of information to a computer you will only confuse it if you give it too much. The computer is very fast and it’s very efficient but it’s very stupid. So it will take everything that you say very very literally. So if you put in a piece of information which is more or less right but isn’t exactly what is in the database then it will reject it. So even if you know everything about a person, you might have their birth certificate sitting on the table next to you, but if you put in all the perfectly accurate information from that, you might fail to find them in the census because maybe the information on the census is not quite the same as on the birth certificate.

So that’s the first thing to remember. I would generally start with no more than the person’s first name, no middle names, just their first name and their last name and maybe an approximate year of birth, and it should be approximate because a lot of ages in the census are out by a year or two. If you get a really large number of results then you might want to refine it a bit. But always start with the bare minimum and then try to narrow it down from there.

Now some of the information in the census you can find using some wildcards. That’s quite useful where you’ve got names that have got various spellings. The general convention is that a question mark will stand for a single character in a name – that’s useful if you’ve got a surname that maybe has a vowel in it but it could be pretty much any vowel, you can use just a question mark for that. But the asterisk, which will stand for any characters or no characters at all, that’s generally more useful. That’s what I use most of the time if I’m doing wildcard searches. That’s a general rule and it won’t always apply to every kind of search. It does depend on what the search engine is capable of. Sometimes you might not be able to use a wildcard at the very beginning of a word or one or two characters in. So you may need to experiment a bit. But those are the general rules.

If you are searching for information other than name and approximate age, you can use birthplace but it’s a very, very tricky thing to use because the birthplace on the census, remember this is from somebody writing down the information that somebody volunteered, they didn’t select it from a list, that might be perfectly accurate and your version of where they were born might also be perfectly accurate, but again, if they don’t match then they’re not going to come up in your results.

I’ve just got a few examples here. In the very first one, we’ve got some of these very typical helpful birthplace information which is like ‘London’. Well that’s really useful because London is an awfully big place and ideally you would want something that narrows it down a little bit.

The middle example is really going to the other extreme. An actual street address, which is unnecessarily detailed and helpful, but you’re not necessarily going to find that because you will probably wouldn’t think of putting down 92 Avendale Square, Old Kent Road in the birthplace field, because it’s not what you would expect to find.

The final one, that I particularly like, is someone who was born at sea; doesn’t just say ‘at sea’, you’ve got the precise compass bearing of where they were born. Again, useful information, when you find it but it’s not something that you can sensibly use to search by.

Now here’s another example of information where you’ve got some fairly dodgy birthplaces – a lot of ‘nk’s which is ‘not known’. But it’s worse than that because if you look at the names column, you have a lot of people who are just listed by an initial of a surname or a couple of initials. You will often find this in places like lodging houses – which is what this is, where the pride of the lodging house is putting down all of the information, and they can only put down the information that they’ve got from their lodgers, and that might not be particularly helpful or accurate. And if the place in question is an inn or a pub, people have been drinking there all night and then stayed for the night, their answers might not be the most accurate that you would hope for.

So that’s just a couple of examples. This one, well you’re not going to find your ancestors obviously but you kind of hope you find your in laws’ ancestors. Prisons and the House of Detention here you don’t get any names at all, you just get the initials so you don’t even have a surname. Some nice detailed occupations and birthplaces, and the ages might be accurate, but you’ll be doing very well to find somebody just using their initials. So that’s another potential hazard and another reason why you might not find somebody even though you put in perfectly accurate information.

Then there’s the question of age. Every census year in supposedly humorous magazines like Punch you’ll get cartoons much like this. They’re usually having a pop at women lying about their ages, although I have seen ones about men who are vague about their ages. Sadly it is true and absolutely verifiable that women have lied consistently about their ages. If you look at the statistical returns, the number of women in their twenties in one census is not really matched by the number that are ten years older in their thirties in the next census. There’s always a little bit of a bulge that way. So I always allow a year at least either side of the year I think somebody was born and sometimes I allow even more than that. That’s where people are either deliberately lying about their age or they genuinely don’t know exactly how old they are so they give you answer which is close but is not quite right.

Then there’s a different kind of problem. When you’re searching the census and you’re either searching by age, or you’re looking at the results to see what sort of ages come up to see if any of them are about the right age of the person you want. This is where you’re at the mercy of whoever has been transcribing the census.

There are three lovely examples here. That top one there, the head of the household, is actually 66. Now this was transcribed as 106 and when you look at it you can make 106 out of it even though 66 is much the more probable answer. In the next one below that we’ve got a servant who is actually aged 27, which you can see if you look at some of the other 2s on that page; I’ll only show a snippet here – if you look at the whole page it’s very obvious. Looking at this page in isolation that 2 does look quite like a 9. But I think you’ll find that not many people would not employ a servant who is 97 years old. So I would go for the 27 here but again it’s transcribed as 97. Then the one at the bottom, the artist is 45 years old not 115. You have to look very closely and you can see that the bottom stroke of the 4 is exactly on the printed horizontal line. So it is 45, but it does look like 115. Again, that was how it was originally transcribed on one side. Though these are just three examples and it didn’t take me very long to find these three to show you. So this is the sort of thing you can look out for and that’s the reason you might want to search without an age at all, just in case you’ve got some odd transcription like that.

This is one of my favourites, this is one of Albert Square in the East End of London in 1871. The returns there are a little on the patchy side because it doesn’t say what the establishment was, but the head of the household was described as ‘brothel keeper’, so you can work it out from there. The details are about the various men, who are mainly sailors, and the other inhabitants, the female inhabitants, are described as ‘fallen women’, a nice Victorian term there. The enumerator helps. We’ve got a comment at the bottom of the page, which is quite hard to read, but it says ‘these fallen women very often don’t know either the ages’ or something else which I still can’t read of the men who spent the night with them therefore they’re only guessed at or words to that effect. That’s an extreme case but it’s a good reminder that the person filling in the details of somebody else in their household might not know the information. They might be making their best guess.

Here’s another kind of thing that you might find and this is nothing to do with somebody transcribing it badly or giving the wrong information. This is for a family which I know quite a lot about. You can see that they’re at 93 Lisson Grove and their surname is very, very clearly Hyatt and you get lots of shots of that because the enumerator has written the surname down on very line. That’s a bonus. Very often they would just ditto it so you only get one shot at the surname, and if that’s hard to read and somebody gets it wrong, then all of the rest of the family would be listed under the wrong name. So this is a really clear piece of writing. However, this is what the enumerator copied from the household schedule, the head of the household Mr John Odem Wyatt who filled in.

Here is the same family, still at 93 Lisson Grove, and you can see that their surname, it’s not particularly clear but the husband and wife have got their names written out in full and you can see that it is definitely ‘Wyatt’ not ‘Hyatt’. And I promise you that it is Wyatt because I’ve researched this family, I know an awful lot about them. So you can search until the cows came home looking for Mr and Mrs Wyatt in the other census year, but you wouldn’t find them because enumerator has copied their details down wrongly. We’re always very quick to blame the transcribers of the commercial firms. Sometimes it’s not their fault – that was transcribed perfectly. It was transcribed perfectly, wrong information.

That comes to the whole business of how do you spell your surname? Now these are some nice examples from one particular person, part of a family that I researched many years ago, and it did strike me that, as spelling variants go, this is an exceptionally good one. The person concerned, he’s just a child in the top example but the surname is next to his father’s name. In 1861, it’s Haveyard, h-a-v-e-y-a-r-d. Then in 1881, there he is with his own family and the surname is almost the same but the ‘h’ has gone, it’s ‘Aveyard’. Then in 1891, again it’s the same family because he has a very distinctive first name, which is ‘Demain’ and he very kindly married a woman named ‘Darinda’, so they’re fairly easy to spot. But it’s the same people and their surname this time, the ‘h’ is back but instead of ‘ave’, it’s ‘alf’ so it’s ‘Halfyard’.

I got so interested in this that I looked at some other sources and I found this man. There are various different ways which his surname is spelt. At his birth, his baptism, so two different spellings for almost the same event in the same year. Then when he married, it’s become ‘Hargard’. I have no idea how that came about except it’s the function of people copying things that are written down and sometimes it’s the function of people writing down what they’ve heard. This is in Yorkshire and I’m not going to attempt a Yorkshire accent but if you try and think of someone in the north, someone in Yorkshire pronouncing ‘Aveyard’ and ‘Halfyard’, they don’t really sound different in my southern accent but in a northern accent they’re very, very similar.

So try to think of the way something looks when it’s written down and how somebody might have copied it and I think that’s how ‘Hargard’ came about. I can’t make it sound like any of the others but all the other variations they probably all sound much the same. People didn’t necessarily have a particularly fixed idea about how their name was spelt; they may not have had to write it down very much and even if they did they might not have cared and they might not have been literate in the first place. So if you tell somebody what your name is and they say ‘How do you spell that?’, you might say ‘I have no idea. I can’t read and write’.

So names do change so it’s worth trying a few different approaches to see how something might be spelt and how it might appear when it’s written down. As you get to know a particular name, you will get to know the particular weird variations and the odd misspellings that other people make of it.

Now sometimes the information in the census itself, apart from spellings, ages and so on, sometimes the information there is just completely wrong. In this one case, I think it’s because the people filling in the household schedule didn’t quite understand the instructions and made a bit of a mess of it. It’s the family that’s right down the very bottom of this page and I’m showing you the detail here, just their names and their ages. Now it’s headed by Emma Weston, so far so good, she’s the head. ‘W’ for widow and she’s 36 years old. That’s fine. But then the next person is called Sarah Anne Parkenson. It is a misspelling – this woman’s name was Parkinson with an ‘I’. But she’s down as ‘mother’. The relationship is supposed to be the relationship to the head of the household. Well she could be Emma Weston’s mother and she’s also a widow, except that she’s 25 years old. So if she’s Emma Weston’s mother she can’t be 25, and if she’s 25 she can’t be Emma Weston’s mother. So that’s a puzzle. Just some extra details. The person at the bottom, which is spelt b-r-e-a-t-e-s, her name is actually Beatrice. Again, it’s a family I’ve done some research on so I know what it’s meant to say. And she is described as the granddaughter and again she’s fairly unlikely to be the granddaughter of the head of the household. What I think has happened here is that somebody has read some of the instructions and they got a bit confused, and they’ve tried to write something down that made sense and ended up with something that makes no sense. This is maybe worth pointing out, that this is in Maidstone, and at this point in Maidstone there was literally a pub on almost every corner, and I’m saying no more than that.

The real answer to this one is that Emma Weston and Sarah Anne Parkinson, they were sisters, and Sarah Anne Parkinson was the mother of Beatrice Parkinson not the grandmother. I tried to work out how they came up with those precise results and I still can’t make any sense of it. So you may find that somebody in your family has done something a little bit weird like that. That is certainly not the only example I’ve found – I’ve found an even better one in my own family. That one is quite enough to be getting on with.

Another thing you might want to do, and I strongly suggest you do, if you can’t find somebody and you’ve tried all the suggestions that I’ve already named, is try a different site. If you’re subscribed to one site, whether it’s findmypast or Ancestry or one of the other sites that have the census on, try using a different site. We can get very comfortable and get very used to the way a particular site works, and we may be fond of using that and be reluctant to use a different one. But it is often worth doing because whenever you have two different sets of people indexing the same material, you will get differences because where it’s really, really clear and easy to transcribe it will probably be identical. But where bits of the writing are a little bit fuzzy, a bit difficult to read, or the information is a bit ambiguous, you will get differences of opinion on how it should be transcribed. Sometimes the site that you’ve been using has got it wrong and the site that you don’t normally use has actually come up trumps. I would never say that one site is any better than another because I’ve used several different sites, and they all make blunders sometimes and they sometimes come up trumps and have the best transcriptions. So even if you think that somewhere is normally very good, maybe it is, but it might just have fallen down on the one that you really, really want.

The other reason is even if the information is identical to the transcription, different sites may allow different kinds of searches and they might be more helpful for the particular thing you want to do. It’s always worth trying at intervals to do this because sites do from time to time change the way their search engine works, sometimes improve it.

One of the things that one site has done is radically improve the way you can use wildcards. It used to be very difficult to use wildcards on Ancestry at one point, because you couldn’t use it right at the beginning of a word. But now you can – you can in the census and that’s made a tremendous difference. And there’s all sorts of other things that a site might do: it might just make its search forms a bit more user friendly or it might add a field that you can search on. So it’s always worth going back and having another look just in case the functionality of the site has improved.

Another thing you might spot is that the image quality can vary. Sometimes, particularly with the early census years, you will find that the quality of the images, which have been scanned from the microfilm, which in turn was done often several decades ago. The thing that you’re seeing on the screen might be quite hard to read but if you try another site, you will sometimes find that they have a better quality image. Sometimes the site may have rather good image editing software that can actually bring out things which are very, very hard to read, even on the master copy of the microfilm. They have very clever machines which can see things that even I can’t. They can somehow bring out details that weren’t previously visible to the naked eye. So that can help. That can also help make a better transcription, if their transcribers can read the image better, then they have a better chance of transcribing something.

There’s also another approach that you can take. Although it’s always very, very important to look at the image because a transcription will only be what somebody else thinks it says, and they might be very good, but you really want to make your own judgment so you always look at the image. But sometimes using indexes which are not linked to images can be very helpful because they might be very good indeed.

The site FreeCEN is quite a good place to go and do this. Once you’ve got what you hope is the answer, then don’t rely on the transcription even the transcription of the household with lots of detail in it. Do go and look at the original or at least the original image just to satisfy yourself that it really does say what the transcribers think it said. That’s what the FreeCEN search screen looks like. It’s a sister site to FreeBMD which is much, much better known. But it’s very well worth having a look at FreeCEN to try and find something that you haven’t been able to find in a site that’s completely images. To translate something from a site such as FreeCEN or possibly even from a printed index, a lot of family history societies have been producing census indexes for many, many decades and some of those indexes are very, very good indeed. So if you can track one of those down, or sometimes family history societies may still have them for sale in booklet form or CDs, or you might find them in a genealogical library. It’s often worth looking at those. Lots of snazzy new tools doesn’t mean discard all the old ones.

If you’re using something like a printed index or an index without images, the way to find the image will be to use the census records. That’s usually how a good transcription will direct you to it. I’m just going to show you here about what the census records is. It used to be many years ago that to find anything in the census, you had to know how census records would work, otherwise you couldn’t find anything at all. Now you’ve got wonderful tools that will search by name and that’s pretty much what we expect most of the time. That will parachute you straight to the right page, but then you might have a problem if you want to try and find it again because you might not be able to remember exactly what search you did to get that particular result.

It’s always worth learning how a census reference works and how to use it because that will never change. Websites may come and go. But as soon as the census enumerator books were acquired by what was then the Public Record Office as original paper books, these references were created and they are still the references. Now this one is from the 1891 census. A reference starts with a series number. You can see that down there – the series is RG12. And the piece, this is a cataloguing term we use in The National Archives. We’ve got lots of other series in which RG12 is just one of them. And a piece number within a series is really what we could describe as an ‘orderable unit’. And that might be a book, folder, a single sheet of paper. In the case of the census, it’s usually a number of enumeration books. That will direct you to part of the census.

It varies a bit from census year to census year. In 1851 the pieces are extremely large – in later census years they were a bit smaller. But once you’ve got a piece, and you’ll see this tag when you’re looking at the census on microfilm, that tag with the series and piece number on without all my red markings all over it, you’ll see that on every frame of the census. It might be at the top or the bottom or either side but it will always be there on the census page.

Then to get your precise page in the census, that’s a little bit trickier. Though every page has got a page number, the page number on its own isn’t enough to find it because a piece, although it varies in size, will usually consist of a number of different books. So if it’s got five books in it, every book will have a page 1, a page 2, a page 3. So if you just give the page number, that’s not narrowing it down to the right book.

Once the paper books came into the Public Records Office they were given folio numbers. The folio number is stamped on the top right-hand corner on every right-hand page on a double-page spread. So it’s not on every image, it will be on alternate images. In this case I’ve shown you a page that actually has got a folio stamp on it. In this case, it’s folio 156 and the page number is 5. Now the next page on, page 6, that won’t have a folio number stamped on it. But its reference will be folio 156, page 6. So the folio number is the really important one. Page number is a bonus – it will narrow you down to a single page. But if you’ve got a folio number, that will take you to a pair of pages and that’s pretty good. So the series, the piece and the folio number are really important ones and you will also be able to find some in the census, and the worst thing you will have to do is look at two pages for your information but that’s not too difficult.

So that’s how a census reference works. It’s a good idea to get in the habit of always noting the census reference or reference of anything you might want to find again, or give the information to somebody else so that they can find it. You can certainly locate the page of the census using these references on both Ancestry and findmypast, and obviously if you happen to be using the microfilm which some people may still have in the library or record office, you will still be able to find it that way.

There is another way you can find a street in the census. This is what we used to have to do in the olden days when there weren’t name indexes. Sometimes it’s still a good idea because you might literally want to be finding who was living at a particular address. Or you might think, I’ve tried all these clever things, I can’t find somebody via the name search engine no matter what tricks I use, so I’m going to do it the old-fashioned way and look at their address and see if there’s somebody there that looks like the person I’m looking for, maybe their names didn’t match because of the transcriber.

That was actually how I found the Wyatt family, masquerading as the Hyatt family, because I looked at the address I had for them which helped when it was the same as in another census year. Now in The National Archives we have typed script street indexes which again use the census records that I’ve just explained, so that’s another reason for learning how to use them. A number of years ago we had a number of these indexes, not every single one but quite a lot of them. We had them typed up into a database and put online. Now they’re part of the government web archive and we do have a link to where you can find these street indexes in our own online census guide. Because trying to give you a precise url to get to this page will be horribly complicated, it’s very, very long. But if you follow the link in our census guide it will take you there and then you can search where there is a street index. There isn’t one for every town for every year – it’s mainly for larger towns and villages. But you can search by census reference and by address.

Now, it is possible to search by address for some census years. You can certainly do this for every census year on findmypast and you can do it in some census years on Ancestry. But Ancestry have also got quite a useful field in their census search which is called ‘keyword’. So it may be worth putting a street name into the keyword search. This isn’t always going to work but it can be a good way of getting you to a street address if you’re using Ancestry and not findmypast. Again they may have transcribed something accurately and findmypast may have got it wrong or the other way round. So that’s a number of different ways that you can locate a street.

Another way you can find a street, whether through a street index or perhaps the keyword hasn’t worked, is you can browse by place for places that are too small to have had a street index and that is places with roughly a population of 40,000 or more. What you can do is browse through the title pages of all the enumeration districts and hopefully in most cases it will have a list of the streets included in that district.

So rather than scrawling through or clicking through every single page of seven or eight enumeration districts, you can look at the first page of each one to see whether the street you want is included there, and that’s much easier than trawling through the whole thing. Again, it won’t always work because sometimes, particularly in the early census years, you won’t get a very detailed description, but in some cases you might do and if you already know the geography of the place that you’re looking at, that makes it that much easier because you will recognise nearby streets and that will speed up the process a bit more. That’s quite a few hints to searching by place which might help you find the person or the family that you want, where all the tips that searching by person or name haven’t worked.

This is something that won’t help you find somebody in the census but it might give you at least reassurance that it’s not there to be found. Over the years we’ve been able to accumulate information about bits of the census that we know are missing. Every census year have got some bits missing, sometimes it’s a whole district or very often it’s just odd pages. For a variety of reasons the 1861 census has suffered worse than any other census year. And although we can’t replace the pages that are missing forever, by at least identifying where they would be we can save people doing a lot of fruitless searches, or sometimes explain why you’ve tried everything you can and you still can’t find the person you want. And if it turns out that they’re in a district where half of it is missing, you might just have to come to the conclusion, oh well, they’re probably in that missing bit, at least I know it’s missing. You might go and try and do something more rewarding than carry on looking for something you’re probably not going to find.

What we have done is created a very big database for this, and a few years ago we were able to upload all of this information into Discovery, our catalogue. To find what’s missing from any particular census year you can do an advanced search and if you put in the series code for the census you’re interested in, in this case it’s the 1861 census which is RG9. So if you search within RG9, ‘search within’ is one of the fields in the advanced search, and then in the description you could ask it to search for the words ‘missing’ or ‘wanting’. There is a technical difference between ‘missing’ and ‘wanting’ and they won’t always be used perfectly accurately. So if you put in both, that will find not all of the bits that are missing but all of the bits we know for certain are missing.

This is still a work in progress – we still get people looking at a bit of census and saying ‘I think there are some pages missing, can you check this out’. And sometimes we check it out and we say ‘sorry, those pages are genuinely missing’, but if we didn’t know that little bit was missing before we can add it into the catalogue and that’s an ongoing process.

Sometimes it does happen; it happened a few months ago, somebody reported that a page was missing from a district that they were interested in and when we looked at it, it turned out it wasn’t missing. It had been invisible to the public gaze for several decades because when that bit of census was filmed, the camera operator had accidentally turned over two pages at once. And we were able to go back and get the original book out and get the missing page images scanned and then uploaded onto the commercial sites that have got the census on. So all because of one very sharp-eyed reader who was looking very carefully at a district and thought that the page was actually genuinely missing and alerted us and we were able to do that.

That doesn’t happen very often but I’m sure it will happen again and the general public searching are our eyes and ears in this regard. With all the millions of pages for the censuses, we can’t look at every single one in detail and check for missing pages, but all the family historians out there can do it for us. So if you do see there’s a page you think is genuinely missing, or you at least want to query it, you can send us a request which will end up on my desk and that’s fine, it keeps me in a job. If you just contact us through our contact us page as a research enquiry and just put the details on and it will be looked at. You will probably get the disappointing answer that, ‘sorry, it’s genuinely missing’, but you might not.

This is something else that can happen apart from pages being missing. Again this is 1861, it’s not just the census that has the most bits missing, it is the one that has the most undamaged bits. Here’s a nice example of a page where you’ve got the names and the addresses, quite a lot of information, but the right-hand bit is missing. So you’re not really going to get anybody’s birthplace and you’re losing a chunk of occupations. Well that’s a nuisance but at least you’ve got something. Of course for the reverse of that page you’ve got occupations and birthplaces, no problem, but for the people on the bottom half of the page you have no idea what their names are.

So that’s something that you will find. Fortunately not too often over the census as a whole but there are certainly pages where there are literally only half of it left. As damaged pages go that’s actually one of the slightly better ones – I have seen far worse than that. So again that is sometimes a reason why you might not find somebody. They’re there, but their name isn’t so you can never be sure it’s them.

And another thing that you really should do is search in every census year. When we do family history we are going backwards and you find a family, you find your ancestor, maybe as an elderly person with children and then you work back, and you find them as a child with their own parents, and then you find their parents as children with their own parents. But you should always come forward as well because even if you think you know quite a lot about a person, there might be some odd little detail that’s in a later census – perhaps they’re an elderly widow parent with one of their children that you hadn’t picked up in another one. And in an extended family there will be, or even just a very large family, there will be different family members present in different years as children are born, as they grow up and leave home, some of the poor little things die young.

You get a much better picture of the family if you try and find as many members of the family in as many censuses as you can. I came across one example where I had a married woman, and she was in the census in the town of Wye, that’s w-y-e, in Kent. And I have a look at three census years and I had great difficulty trying to find her birth, because in every census year it’s down that she was born in Wye. It was only when I found her as a much older woman in a later census, she wasn’t in Wye at the time, she had moved away from there to another part of Kent. Then in her birthplace column, instead of saying Wye, Kent, it said Rye, Sussex. I don’t know if somebody had a speech impediment or was a bit deaf, but if somebody’s in a particular town and they come from somewhere that sounds very like it, then the person filling in the details on the form, even if it’s a family member, they will tend to put down something that they are familiar with.

If somebody, for example, has a place name that is familiar for another part of the country – Kent is full of examples like this and there is a very small part of Chatham in Kent which is called Luton. Now people who don’t come from Chatham probably haven’t heard of Luton; if you say ‘Luton’ you think of Luton, Bedfordshire. So it would be very easy for somebody from the part of Chatham, when asked their birthplace, to say ‘Luton’ and the person listening to assume they meant Luton, Bedfordshire. There is also a Leeds in Kent and a much more famous one in Yorkshire. There is a Rainham in Kent and a much more famous one in Essex. I know this because I used to live in Kent so I’m fairly familiar with Kent place names. But you can probably do much the same for other places that you’re quite familiar with. There’s a lot of misunderstanding. It’s not like somebody was giving information the enumerator misunderstood. It could be that several years before they mentioned to someone in their family that they came from Luton, or Rye, or Leeds, or Rainham, and that person misunderstood and immediately misguided them to the wrong part of the country altogether.

One of my great grandmothers came from a place called Boness, which is part of Inverness, and in the census she is clearly down as Boness in Linlithgow, which is several counties away in the wrong place altogether. So if you have a look in your own family and look at the places they came from, and the places that you know, and you might well find similar misunderstandings like that.

This is an example: this is in many ways my very favourite census – the 1911 census – and I’ve put it here not just because it’s a very nice census that gives lots of information and where you see the household schedules, but there was one unique thing about the 1911 census, and it has a feature that I have seen in no other census, and that is it is only one that gives, or supposedly gives, exact birthplaces in Ireland. The Irish 1901 and 1911 census gives just the county. To my great annoyance the Scottish 1911 census doesn’t give exact birthplaces in Ireland, just the fact of Ireland. You might get some more detail but all that was required was just ‘Ireland’. But the English 1911 census did ask for the exact parish and county in Ireland. So if you have an Irish ancestor and they came from Ireland to England or Wales, and you haven’t been able to figure out which part of Ireland they came from, it’s worth coming forward and hoping they managed to live long enough to be in 1911 or failing that, if they have a brother or sister who lived long enough to be in 1911. So 1911 could be the clue – your one breakthrough for finding an Irish birthplace, which is one of the hardest things to do. So that’s one of the many reasons I like the 1911 census so much.

So that’s some of the ways you might be able to find somebody in the census when maybe you thought you couldn’t. But sometimes, for various reasons I’ve already given you, they might not be there to be found or that census return might have been destroyed never to be seen again. But you always should remember, when you’re doing family history, that you’re not looking for a particular document. What you want is the information that’s on it. If you can’t get the document you want, whether it’s a birth certificate or a census return, try to think of other ways where you might find the information that you would hope to find in this non-existent record.

If you’re looking at the census, one of the things you want to find is how people are related to each other, you want to reconstruct a family. It’s very nice if you find a family with two parents, a whole bunch of children, maybe some in-laws thrown in for good measure. You can virtually do a family tree from that. But there are other ways you can do that and wills are one of the best. This particular one is very, very long and it’s a man who had several children and he very decently noted them down in quite a lot of detail and described exactly how they were related to him. He mentioned grandchildren as being the sons or daughters of my son and so and so. This is quite important in wills because if a man has three or four sons and each one of them has a son named after him, he’s got to be very specific that his grandson Thomas Gash is the son of his son Richard, or his son William, or his son Thomas, to distinguish him from other grandsons of the same name. So wills can be tremendously useful for this. In this particular case, as well as this very informative will, the death duty register entry goes with this one is also very useful because it’s got dates of death for his unmarried daughters who died many years after he did, but that’s a subject for a completely different talk altogether.

So don’t get too focused on being determined to find that document because there might be some alternatives, and I’ve certainly in the past spent ages looking for a census return before there were name indexes, and try as I might I couldn’t find it because these wretched people just kept moving in a very built up area. But almost by accident I found an application for poor relief, which gave all the detail I would have found in the census and then some. So don’t give up if the elusive census return that you’re pinning your hopes on doesn’t exist, because you might be able to find that information somewhere else.

There is a lot more information about the census in our online guide on census records, which has just recently been updated, so there’s even more information in it than there was before. That has links to how you find the street indexes that I mentioned on the government web archive, and also we have a link to the blog post that I did about the subject of missing from the census, which repeats some of what I’ve said here but you may get some other hints from there.

So never give up – there might be an alternative, and do have a look at the census guide because there might be little hints in there that you haven’t previously thought of. So if at least a few people come away from this and try something they maybe hadn’t tried before I’d say my work here is done.

Thank you very much for listening.