Survey of Contemporary Surnames

The Survey was commenced in 1980, by the leading lexicographer -Patrick Hanks - in order to establish a database of names for the Oxford University Press's Dictionary of Surnames. The work continued until about 1996, with the help of Kate Hardcastle and others, until the advent of on-line computer directories made manual extraction uneconomic.

The result is an alphabetic list of the 16,000 most frequent surnames of Britain and Ireland, with the areas in which they occur with greatest comparative frequency. Surnames with a frequency of less than four per hundred thousand in any one area were generally not collected,except in London, where the threshold was two per hundred thousand. Generally, the location of the highest absolute frequency has been left in, even when it is not statistically significant. As a result, London and -- to a lesser extent -- Birmingham, Glasgow, and other large cities are given rather more prominence than other places.

Method

Patrick describes his methodology :-

What we did was: noted the frequencies for each name in the areas of highest frequency. I subsequently normalized this to allow for differences in directory sizes: the database expresses frequencies both as absolute counts, and also (in brackets) as a frequency per 100,000. We deleted (or did not bother to enter) low-frequency occurrences of names for which we were able to find significantly higher frequencies elsewhere. Most areas of the British Isles are covered. (except for Staffordshire and Blackburn, if I remember rightly.)

A fuller description appears in his article in Nomina

Patrick Hanks 'The present-day distribution of surnames in the British Isles' Nomina 16 pp79-98

Interestingly, Patrick has gone on to compile a headlist of US family names for the Dictionary of American Family Names. The database being created from the US telephone directory on Cdrom

Access the Data

Areas surveyed (20.3.98)