News & How-To
Formerly branded as GlobalGazette.ca
Articles, press releases,and how-to information for everyone interested in genealogy and history
Subscribe to our free newsletter
[an error occurred while processing this directive]
ENGLISH & WELSH ROOTS
Article posted: August 12, 1999
By: Fawne Stratford-Devai Biography & Archived Articles
When I first started the English and Welsh Roots Column in the Gazette I promised guest authors who were experts in a variety of English and Welsh family history topics. I am very pleased to welcome the first guest author Barney Tyrwhitt-Drake. Well known on many British genealogy email discussion lists for his patience and helpful resource information, Barney is also a regular contributor to Family Tree Magazine and a variety of genealogical publications. More importantly, Barney is known and respected for his computer expertise in both the use and creation of genealogical software. Barney's article this issue articulates both the problems associated with the LDS 1881 English Census CDs and timely solutions for making the most effective use of this invaluable resource. I look forward to many more such articles from Barney in future issues of English and Welsh Roots.
- Fawne Stratford-DevaiThe 1881 British Census on CD , Problems and Solutions
By Barney Tyrwhitt Drake
The release of the index to the 1881 British census on CD-ROM by the LDS Church is welcomed by family historians around the world. At last we have an index to locate all of our ancestors and their relatives living in Britain at that date. A recent article in the UK publication 'Family Tree Magazine' by the respected genealogist and author Susan Lumas, suggested that the error rate in the 1881 index was well below 1%. The basis of this estimate was the number of errors reported by users to the FFHS (Federation of Family History Societies) for the microfiche version. I believe this is an underestimate and that the chance of finding an inaccurate individual entry in the CD-ROM index is an order of magnitude or more higher. This article describes some of the common problems you may come across using the 1881 British census on CD-ROM, and how you can overcome them.
The installation instructions that come with the CD-ROM set are a model of brevity, but unfortunately not of clarity. Many users have discovered that they install the National Index, find a person of interest, double click on that person's entry, and up comes an error message. What the installation instructions didn't make crystal clear is that the install routine is a 3 step process.
1. Install the LDS Resource File viewer program.
2. Install the 1881 British census National index.
3. Install the 1881 British census Regional index.
If you also have the British Isles Vital Records Index, you may want to install that as a fourth step also. The way to install in steps 2. and 3. is to ignore the statement <Click here to add a Resource File to the list>. When you click nothing happens. What it should say is <Double click here to add a Resource File to the list>. Do this for both Disk 1 of the National index to complete step 2. and then repeat it with any of the Disk 1s from the Regional indexes to complete step 3. The East Anglia CD is the obvious one to use. When you've finished you should have entries in the Resource File Viewer for both '1881 British Census' and '1881 British Census - National index'. Now when you try to move between a National index entry and their full record in one of the Regional indexes, you will be successful.
Other users have reported that even when they do this, the program will not run. The problem here is often a clash between the install routine and your anti-virus software. The best way is to run the uninstall program from the Windows Start | Programs | Family History | Resource File Viewer - Uninstall menu, deactivate (or even remove) your anti- virus software, and begin the Census install process all over again. Some anti-virus software is more problematic than others. There seems to be a particular problem with Norton anti-virus but almost none with McAfee anti-virus.
FIRST LETTER PROBLEMS
These problems surface when you try to find someone and they simply are not there! A friend of mine was searching for the family of Samuel and Harriet RYCROFT who were most probably in either Yorkshire or Lancashire in 1881. Samuel worked for a railway company making rolling stock and coaches. Neither on the microfiche nor on the CD-ROM could she find them. One of the most frequent problems with transcribing any of the 19th century CEBs (Census Enumerators' Books) is that they are the original manuscript of the enumerator. The style of writing then was such that initial capital letters to words could easily be confused. A 'R' could be mistaken for a 'P' or a 'B' or even an 'H'. The rapidly written vertical strokes meant that the letters 'M' and 'W' were often confused. The single vertical stroke and a lower loop in both 'L' and 'S' means that they can be taken for each other. My ancestor was a lawyer you discover! In fact he was cutting up wood and was a sawyer.
There are several ways you could solve this kind of problem. You could try entering the surname BYCROFT instead of RYCROFT and see what comes up. Probably quicker and certainly more thorough is to use wild cards. A wild card is a symbol that replaces one or more letters in a word you are searching for. The ? symbol means any single letter of the alphabet, while the * symbol means a string of as many letters as you like from zero upwards. Instead of entering a name to search on like RYCROFT you could enter ?YCROFT and this would find RYCROFT, BYCROFT and PYCROFT, or any other first letter variant. This policy paid off in the search for Samuel RYCROFT. His census entry had been mistranscribed as PYCROFT, and he was found under this name living in Lancashire.
For more on problems associated with the census Enumerators Books refer to the earlier two part discussion on census records in English and Welsh Roots:
Census Records Part 1
Census Records Part 2.
OTHER LETTER PROBLEMS
Although first letters are the ones that are mistaken most often, there are still problems with other letters, particularly with vowels. The letters 'i' and 'e' can look similar since enumerators often forgot to dot the former. Similarly the letters 'l' and 't' can look the same as enumerators forgot to cross the latter. Once again, judicious searching using the ? wildcard can overcome these mistakes. The LDS search engine is not totally flexible when it comes to wild cards. It will not let you use the * wild card at the beginning of a word, nor does it allow you to combine wild cards one after the other.
A number of census entries have either an alias as well as a surname, or a name such as 'JONES or SMITH'. These people appear twice in the indexes, firstly under JONES and secondly under SMITH. This is fine as far as being able to find them under either name is concerned, but you need to be careful if you are using the census to find populations in 1881 for specific villages or parishes. If you search with the name blank and just the place name for the village in the census place, you will double count anyone whose surname is of this doubled variety.
The original Census Enumberators Books (CEBs) had two columns for ages for males and ages for females. The 1881 census index has changed this to a single age column and a sex column. The original CEBs are not as pristine as they were when they left the enumerator. Census office clerks were responsible for summing various categories in order to prepare the central statistics that were the whole purpose of the census in the first place. Frequently they counted by scoring a thick pencil stroke through the number in one of the age columns - an action that makes some of the figures hard to read. So, if you see a widow aged 11 with a son aged 44, the probability is that she was really 71. Children under the age of 1 had their ages recorded in terms of months, weeks and days. By far the most common are ages such as '5m' meaning 5 months or '3w' meaning 3 weeks. Occasionally you'll find someone over the age of 1 recorded as '18m'. If you print out a census entry in the RTF (Rich Text Format) the age column sometimes runs together with the gender one giving what looks like '5mF'. You can normally interpret these correctly from their context.
Another feature of the 1881 census index is to show a possible birth year for everyone that is calculated by simply subtracting their age from 1881. Thus if someone is recorded with an age of 2, their birth year will be shown as 1879. However the 1881 census was taken on the evening of Sunday 3rd April. People born between 1 January 1879 and 3 April 1879 will be aged 2 and will have their birth year shown correctly. People born between 4 April 1878 and 31 December 1878 will also be aged 2 but will have their birth year shown incorrectly by one year. And of course there are more 2 year olds in the latter category than the former. You need to be very careful when searching on a year of birth and setting the variance allowed to plus or minus 0 years.
PLACE NAME PROBLEMS
You can search the Census index by both the place where the census took place and the place where someone said they were born. All the search results, both on the National index and Regional indexes show a county or country in their abbreviated forms in the Index pane. Unfortunately the LDS have chosen a completely non-standard set of county and country codes instead of the well-known Chapman County Codes and the British Standard County Codes. You should be able to interpret most of these without too much problem, but please don't quote them in genealogical correspondence or we'll suffer from a surfeit of 'standards'.
In several cases there have been some indexing disasters in the LDS computing centres. Some civil parishes have been placed in totally the wrong county, and probably the worst of all is what has been done to the birth places in the Scottish Highland county of Sutherland. Run a search with the Scottish Highlands CD in the drive and no name for the person, but the county of Sutherland in the census place. You'll find 22,402 matches showing the complete population for that county. Now erase that search and keeping the name blank, enter Durham, England as the Birthplace county. This shows 18,827 people born in that county living in Sutherland, and indeed 42,149 from Durham living in the Highlands altogether. At a time when there was a massive migration of people into Durham as a centre for coal mining and shipbuilding, this exodus seems rather strange! It turns out that someone has confused the city of Sunderland in County Durham with the county of Sutherland in Scotland and decided they are synonyms, when of course they are not. There was no large scale migration of this type.
All of which points to another weakness of the indexes that you need to be aware of and able to work round. The LDS sense of British geography is, how can I put this politely, eccentric to say the least. Despite extensive feedback from beta testers in the British Isles, they decided they knew better... One of the most confusing things are county names. A county like Warwickshire is everywhere referred to as Warwick. The confusion arises because the city of Warwick is also in that county. When someone records their place of birth as Warwick we have no idea which one they mean. To be fair, this was also true in the original CEB, however the convention there was to place the county name first, so if the enumerator used Warwick he probably meant that the person was born in Warwickshire rather than in the city of Warwick. In today's maps and speech we always say Warwickshire to identify the county and Warwick to identify the city. In 1881 things were not so clear cut. So, if you find places such as Warwick, Buckingham, York, Derby and Nottingham, don't automatically assume they are towns or cities, they could be counties or vice versa.
This geographical eccentricity goes one step further in the division of counties into the 13 regions on the LDS CDs. Some counties just don't turn up where you think they are going to be. Buckinghamshire (called Buckingham of course...) is the to NW of London, yet strangely it appears on the Greater London East CD. Northamptonshire, which is well to the West of East Anglia appears on that CD. The best advice here is to read the Regions list in the front of the CD package. In fact it is a good idea to make a photocopy of it and amend the county names to show their more normal forms.
Mobile parishes are the next problem. A number of civil parishes in Britain have always straddled county borders, but some of them have been placed in clearly the wrong counties in the 1881 census index. For example, the civil parish of Ickford in Buckinghamshire only ever had an odd house or two that strayed across the border into Oxfordshire. Yet the whole of the civil parish of Ickford has addresses in Ickford, Oxford, England. Strangely the people censused in Ickford and also born there always show their place of birth as Ickford, Buckingham, England. If you are looking for people you know may be in a border parish, it's always worth getting out your atlas and checking the county next door, which by a corollary of Murphy's Law will always be on another CD-ROM!
Looking for people by searching on their place of birth is fraught with problems, which are mostly due to the very variable way in which the enumerators spelled birthplace names. The instructions to the transcribers for the 1881 census index were to reproduce exactly what the enumerator wrote, so things that are plainly wrong will stay that way. You need to be able to work round these by knowing some of the common variations. Some are downright impossible. One enumerator in Buckinghamshire recorded a birthplace for the hamlet of Hyde Heath as Ideith (say it, and you'll see), probably because the person being enumerated was illiterate and just pronounced it that way. The enumerator didn't know the place so just took a guess. You'll find a lot like this. Beccles in Suffolk can be spelled as Beckles, Beckels or Beccels. Use of wild card searches for places such as Bec*l*s is a good strategy for eliminating these variants.
Many enumerators used abbreviations. The village of Little Missenden often appears as Lt Missenden. A wild card search for L*t* Missend?n should find most of them. The other common error is simple transposition of letters in a place name. Just occasionally you come across an enumerator with a sense of humour. There's one I've found in Buckinghamshire who recorded someone's birthplace as Aaaylesbury. Clearly the person had a strong accent and the enumerator was in a mood for fun.
Although you cannot search for people based on their sex, occasionally you will come across some surprising census entries, such as the one for 24 year old Elizabeth HAWKINS whose occupation was recorded as Farmer's daughter. A certain female, yet somehow her sex has been recorded as Male in the LDS database! She/he is not alone. In a small county like Buckinghamshire with just over 180,000 population, 26 Eliza/Elizabeths are recorded as male, and there is a fair sprinkling of female Richards as well. This won't affect your ability to find people in the index, but it serves as an extra reminder of the necessity of going back and looking at the original census for anything interesting you find.
SAVING FILES AND PRINTING
There are limits of 100 records at a time when it comes to saving or printing search results. In my experience this limit for 100 is an overestimate. Quite commonly you think you are saving 100 records just because you have selected 100, when in fact you will only save around 70. This is a bug in the LDS Software, but it's best to work round it by only trying to save or print a maximum of 50 records at a time.
The Help file, which has a lot of useful information in it, doesn't however make it clear that when you select say 50 personal records to save to file, you are in fact saving the 50 households that those 50 people are in. If you have more than one person in a household in your list of people to save, you will save that household twice or more in the file you create. This isn't really a problem, except that if you are going to try and put this information into one of your databases, you will need to remove the duplicate entries.
The only format that the LDS software saves complete records in is as an RTF file. This is fine if you want to print it from a word processor, but even here there can be problems. The RTF file that is produced has some of the personal details of people coded as hidden text. You will need to turn on the hidden text viewing and printing functions in your word processor to see the text correctly laid out. In Microsoft Word go to the Tools menu, select Option, and on the View and Print tabs of the dialog box that opens, make sure the 'Hidden text' checkboxes are checked.
Suppose you are collecting all instances of a particular surname from the 1881 census and you want to put these into either a spreadsheet, database, or your family history program. None of these will accept input from an RTF file. As I struggled with this in testing, I decided to write a Wizard program that allows you to convert an RTF file from the LDS 1881 British census into any of GEDCOM, CSV (delimited .TXT file), .DBF or .DB formats. One of these will be ideal for direct import of the census data into just about any spreadsheet, database or family history program. The RTF Wizard program allows you to do these conversions, and to append the results of processing multiple small RTF files into larger .TXT, .DBF or .DB files. If you are interested in extracting sets of data from the census for your research, there are details of RTF Wizard on my website at www.tdrake.demon.co.uk/rtfwiz.htm. RTF Wizard can be purchased directly from the Global Genealogy website at: RTF Wizard.
I seem to have dwelled upon the many problems in the 1881 British census CD-ROMs, but would like to finish by saying that this is a marvellous index. Provided you understand the problems, many of which go back to the original enumerator, you can find a way of working round them. Given the choice of searching the 1881 census on microfiche or on CD-ROM, the CD-ROM wins hands down due to its flexibility, speed and low cost. For every problem I hope I've suggested ways in which you can work round them to find the people you are looking for. All in all the LDS Church, the Federation of Family History Societies and the PRO (Public Record Office) are to be congratulated for creating an index for all of us.
Given that there are so many errors and variations in the 1881 census index, we should never forget that it is just an index. If you find anyone of potential interest, write down either the FHL Film number (if you are going to use a Mormon FHC) or the PRO Reference number (if you are going to use the Family Records Centre or a library that has the facsimiles of the CEBs). In either case you should view and make a photocopy of the facsimile of the CEB page for that ancestor or relative. This is the primary source you should be quoting, and not the CD-ROM index itself.
One frustration more advanced users of the 1881 British census may have felt is that it is not possible to search for people by either address or occupation. I hope to look at some methods and software in future articles that will find ways to overcome this limitation also.
CD 'S BY DRAKE SOFTWARE
RTF Wizard for the 1881 British Census on CD-ROM:
The LDS Church's release of the 1881 British census on a 24 CD-ROM set is a major event in genealogical computing. Unfortunately the only format in which events can be fully exported is as a Rich Text Format (.RTF) file for use in word processors. RTF Wizard converts these RTF files into GEDCOM, .DBF, .DB and .TXT (Comma separated variable) files that can be used in any spreadsheet, database or family history program.
DATA from Drake Software:
Pigot's 1834 Directory for County Durham and Northumberland- V 1.0
Pigot's 1830 Directory for Berkshire, Buckinghamshire and Oxfordshire- V1.0
New at the website of the Archives of the State of New South Wales Australia. The website of the Archives of the State of New South Wales now features 2 important online indexes. Even if you don't think you have a connection with someone sent to Austraila, you never can tell if a distant great-great-great-...uncle or other family member was sent to Austraila in the 1800s. In my own family, a brother of my direct ancestor received assisted passage to Austraila. Today there are at least 4 distant relations actively working on the genealogy of our shared Stratford family and others who are also interested. The work of all these Stratford decendents, on all continents, has been of benefit to everyone.
Index to Certificates of Freedom, 1823-69 now online. This is the first of a range of indexes relating to convicts which will be progressively added to the website of the Archives of the State of New South Wales. Visit the website of the Archives of the State of New South Wales Austraila at: www.records.nsw.gov.au/ Once at the main page, click on the "What's New" bar and scroll down until you see the link to the index. Don't forget to read the explanation of the convict system, what exactly a certificate of freedom was and how the index was compiled.
Index to Assisted Immigrants arriving in Sydney & Newcastle, NSW, Australia 1844-1859 now online. Visit the website of the Archives of the State of New South Wales Austraila at: www.records.nsw.gov.au/ Once at the main page, click on the "What's New" bar and scroll down until you see the link to the index. The indexes to assisted imigrants may be searched by the name of the immigrant or the ship of arrival. Indeed, once you have found the ship of arrival of an ancestor by conducting a surname search, you would do well to conduct a search by ship to see who else traveled to New South Wales with the family.
What is an assisted immigrant? The term 'assisted immigrant' refers to those people whose passage was subsidised or paid for through one of the several assisted immigration schemes which operated to Austraila, Canada and the U.S.A. from the United Kingdom and other countries. The importance of assisted immigration schemes and examples of some records available to researchers will be discussed more completely in future issues of English and Welsh Roots.
About Fawne Stratford-Devai
Fawne Stratford-Devai's work on Land Records and early Ontario records is well known in the genealogy community. A published author of several Canadian and UK research books, she has also contributed articles to the Ontario Genealogical Society's newsletter "Families" as well as writing for the online family history newsletter the "Global Gazette". Biography