We have loaded a fresh version of the searchable data onto the website, with a number of enhancements:

• Transcription errors reported up to February have been checked and corrected where necessary
• Data standardisation has been applied to first names to correct common mistakes such as Geroge for George
• Data standardisation has been applied to ages to make them standard: for example the various ways that householders may have written “months” has been standardised

Address searching often requires a degree of lateral thinking to get the best results. Here’s a few extra tips and also some new features on the horizon which aim to make your searching easier. The post below is based in replying to questions from a customer searching in Dorking, Surrey but the points apply equally to addresses across the country.

The source of address details on the census is that taken from the original form filled in by the householder (this contrasts with previous censuses, where the forms were compiled by the enumerator, thus introducing some level of standardisation in recording). Unfortunately, several factors conspire to make the historical document problematic for finding addresses using 1911 census returns. 

The first is that in 1911, the concept of a full postal address with a number and street was less evolved than it is today. Many houses simply carried names and householders would then place the town afterwards. To take an example, looking at modern-day Pixham Lane in Dorking, Surrey, the majority of the houses carried names but most householders simply included their postal address as “name of house, Dorking” and this is the information that we transcribe. Unfortunately this was compounded by the small space on the original form left for the address, meaning the householder would often abbreviate the address to make it fit. Have a look at an example of an Original Page to see how small the space was for your ancestors to enter their address. 

The second is that many householders used abbreviations for words (as we do today), such as “Rd” for “Road”. Again using an example of Lincon Road in Dorking (around the corner from Pixham Lane) if you search for “Lincoln” on its own in Dorking, Surrey all 44 properties are returned sequentially, some listed as “Lincoln Road” others as “Lincoln Rd”. Try searching for just the first part of the address and leaving off lanes, Roads, Crescents etc, but narrow the search area by county and district first.

We will be applying many data enhancements and standardisation processes over the coming months to compensate for these common inconsistencies in the originals and to make the data more easily searchable. However, the transcriptions are in this case accurate based on the original documents. To get the best out of any historical document, a degree of lateral thinking often has to be applied. 

Thirdly, place names and spellings change: in the case above, Pixham had an alternative spelling of “Pixholme” and 35 properties are found in Dorking under this listing. If you can find contemporary maps of the area you are searching, either online or in local libraries and archives, these can prove useful as the name today may be utterly different.

 Finally, with 8 million different sets of handwriting, deciphering becomes extremely difficult and what may appear to be transcription errors (and in some cases are) occur. Thus we found one property transcribed as “Pischolme”. However, when examining the householder’s writing, the awful way he had formed the X would lead any person to transcribe it this way.

 We are working on a number of ways to make searching by address simpler in face of the difficulties posed by the original records, but the unique nature of the 1911 census means these methods have had to be worked out afresh for this census, and the census is very much work in progress, although to date hundreds of thousands of researchers have successfully used the service to identify the records they want to view.

 As well as applying many enhancements to the data to attempt to smooth over the inconsistencies of our ancestors, we will also release the RG78 Enumerators Summary Books soon (current estimate is April), which list the households and heads in each area: this information is invaluable for identifying neighbouring houses when the address information left by our ancestors makes this hard to recover. If you have already paid to view a household image, you will be able to view the linked Enumerators images for free, by returning to your saved records. You will not be required to make further payment to view these.

We will also be adding a wildcard search to the street field to allow you to search laterally and many more data standardisations will be applied over the coming months.

We have released a number of new search features this morning - we hope you find them useful.

1. “Last Name” is no longer a compulsory field - you can search using “First Name” or “Census reference” instead
2. Up to 5,000 search results can now be viewed
3. Searches with up to 1,000 results are now displayed automatically, without the “gateway” screen being shown first
4. You can choose to enter a Census reference number instead of an address in Place Search (see the help file next to this field for further details)

We have turned on some more search features for you. Firstly, you can now use a ‘leading wildcard’ (i.e include a wildcard as the first letter of the name). This is helpful as the initial character of names is more liable to be mistranscribed.

Secondly, we have increased the range of dates that you can search: you can now search +/- 5 years on Date of Marriage and +/- 10 years on Date of Birth.

More goodies coming next week - keep your eyes peeled.

We are happy to tell you that we have now applied changes to the data on 1911 Census to allow you to search for variants for both first names and last names.

The variants search is essentially a large thesaurus that identifies common variants, mis-spellings and alternative spellings for many of the names within the census. This should be of great help in tracking down elusive ancestors that you have so far not been able to identify. 

Click on the “Show advanced fields” button to make use of this search feature.

Coming next: allow a wildcard as the first character, increase range of years available (currently set at a maximum of +/- 2 years).

Enjoy!

We’re very excited to tell you that we have started unlocking the search features.

You can now search using wildcards, or using the “name starting with” feature. To try these out, go to “Search” and click on the “Show advanced fields” button at the bottom.

Both these features will give you a lot more flexibility. For example, I just searched for all footballers in England starting with S.

Wildcards are very useful if you are unsure of a spelling - you can substitute any or all of the parts of a name with a “*”. Searching for “R*B*T*S*” would give you a huge number of variants including Roberts,  Robertson, Rabbits, Reblatts and more. “AN*T*” will find Anthony, Annette, Antony, Anita and many more variations.

N.B You cannot search using a wildcard as the first character yet, but we will switch this on soon. Keep an eye on this blog for updates.

Coming soon - a variants search: this finds variants of first names and surnames. A variant search for Elizabeth would find you Liz, Eliza, Liza, Betty, Betsy, Beth and more

For more on search features, read this earlier post: http://blog.1911census.co.uk/2009/01/extra-search-features-coming-after-launch/

A few people have been having difficulty finding ancestors using the address search. Here are some tips to improve your chances of success!

The address search uses data from the RG14 (Household) schedules rather than the RG78 (Enumerators Summary Books). This means that the data being searched is what your ancestors would have written on the form rather than what the enumerator would have written. Because an enumerator would have been more likely to give a single, standard name to a particular street, there will be more variations in what the householders in a road have put on their form.

For example, the road near where I used to live is called “Clapham Manor Street”. Householders in this street have used various permutations such as Manor St, Manor Street, Clapham Manor, Clapham Manor St, Clapham Manor Street and more (including misspellings on the original household page itself).

We are using intelligent filtering to clean up the more obvious variations such as “Rd” for Road and “St” for Street but obviously, some lateral thinking may be required with these records. Here’s a few tips.

1. The wildcard search will be available soon (next few weeks). This will allow you to search for variants within a district (in my example, perhaps “*manor*” within the registration district to find all roads with ‘manor’ in the name in that area.

2. The RG78 Enumarator Summary Book images will also be available soon. These will let you find a family nearby, then look at the street summary to identify their neighbours (hopefully including your elusive ancestors). They will also give you fascinating details about the area your ancestors lived in.

Hope that helps!

We expect the site to be very busy at launch, and so we have taken a number of measures to make sure that as many people as possible can get their searches completed successfully.

The first is that we have deliberately restricted some of the search functions on the site, meaning that the search is not as flexible as it will be later on - please see our separate post on the extra search functionality that will be ‘unlocked’ after launch.

Secondly, we are only allowing users to download original pages, rather than view them directly on the website - an in-page viewer will be switched on shortly after launch.

Thirdly, only one image will be available at launch: the principal side of the original household (or institution) page. Most searches have up to seven associated images, which will be made available to you shortly after launch. These extra images are included in the cost (30 credits) of the original page so you will not need to pay extra to view them once available. The images include the rear of the household form and images from the Enumerators’ books (akin to images from earlier censuses but with more detail and in full colour). Please see our separate post about the extra images - they are really rather exciting and unique to the 1911 Census!

During launch, if the site becomes exceptionally busy, we will also restrict the number of new people entering the site. We want to allow the people who are already using the site to complete their searches rather than make the site impossibly slow for everybody. We have built the site to withstand a very large (but not infinite) number of visitors, so if you find that you are not being allowed in, please bear with us.

We call this ‘velvet roping’ - much as you would restrict access to a building to large numbers of visitors to ensure the comfort of those inside, we will try to make sure that the experience of those searching is not overwhelmed by new visitors if the site becomes very busy.  Real-time service updates can be found on this blog, which we advise you to check frequently for the latest news.

Obviously, we will ‘unlock’ the powerful search features, switch on image viewers and add the extra images as soon as possible after launch and let you know by email and via this blog when we have done this. We cannot commit to specific dates for this yet, so please check here for the latest news or sign up for our newsletter (in the “My Account” area on the website).

We have temporarily restricted the search facilities available at launch to allow a greater number of searches to take place at the same time. As soon as we can, we will ‘unlock’ the search feature to allow you to search much more flexibly and find those elusive ancestors. Here are the main features to be unlocked.

1. You will be able to perform a person search without the need for a surname, allowing you to find ancestors where you are unsure of the surname or to perform broader searches (finding all tram drivers within Manchester for example)
2. You will be able to perform ‘wildcard’ searches. This is where you can search using only part of a word if you are unsure of a spelling for examaple “an*t*” will find Anthony, Annette, Antony, Anita and many more variations
3. You will be able to perform ‘variant’ searches - this finds variants of first names and surnames. A variant search for Elizabeth would find you Liz, Eliza, Liza, Betty, Betsy, Beth and more
4. You will be able to perform an address search without the need for a street name allowing much more flexibility
5. You will be able to search date fields with a wider variation in date - for example birth year will allow to to search 5 years in either direction instead of the current 2 years

Many people have reported that the clear search button seems to be a bit erratic. It is working, but not perfectly. Make sure you click on the text in the middle of the button - clicking on the edges doesn’t seem to do anything at the moment. We will try to fix it soon!