So I have started with Marriages.
If I filter my name index I find that there are 5,307 marriage index records in my database. But this doesn't mean there are 5,307 marriages as each name referenced counts as a separate record. So my next filter is by surname to include those with surnames starting with TILL. This reduces the number to 3,160 names. I then removed any references where the person with the surname starting with TILL was the spouse and not the primary Tillin reference. Finally, I got to 3,061 records from the indexes between 1837 and 2005 (according to Ancestry.co.uk) in England and Wales.
The split by surname was as follows
TILLING - 2679
TILLIN - 238
TILLEN - 136
TILLON - 4
TILLYN - 2
TILLAN - 2
which is what I would have expected as TILLING is a much more common surname.
To further analyse the data I exported it to an Excel spreadsheet and used a pivot table for easy analysis.
By decade, it looks like the most popular decade for marriages was the 1960s.
If I look at the TILLIN name itself the top 5 registration districts are Camberwell, Reading, Windsor, Kettering and Bradfield. This is sort of what I would expect but the fact that Kettering has made the top 5 places was a bit of a surprise although all of these records were from 1936 onwards.
If I look at TILLING the top 5 registration districts are Bristol, Cheltenham, Swindon, Stroud and Warrington. This time Warrington is the surprise as I haven't had very many records from the North of England so that is a new area to look at.
The top 5 TILLEN registration districts were Newbury, Reading, Bradfield, Wokingham and Brentfield.
At the moment I'm not sure what this tells me but I find the geographical spread of the surnames quite interesting and maybe I need to look at some sort of mapping analysis.