Links in This Section
The links in this section include some traditional sources used for plotting surnames (censuses) as well as some early sources that will not cover an entire country but nevertheless illustrate some geographic occurrences before the time of censuses.
Why Study Distribution and Frequency?
There are two main reasons to collect information on the distribution and frequency (rate per area) of the surname.
1. For those who don't know where their family is from, noting areas where its distribution and frequency are heaviest provides reasonable locations to begin searching. If you know one or more other surnames that married into the family in the same country, you can look for areas where both surnames were prevalent.
2. To seek a best guess of the surname's geographic origins. “Best guess” is the operative phrase here for the Bowes surname and its variants for a couple reasons:
a. because we have no documentary proof of our geographic origins, and
b. because even when an area of heavy concentration can be located by mapping the surname data, the surname's founder could have lived somewhere where the name subsequently died out while it gained prominence in a new area(s) as the founder's descendants expanded. This is referred to as “genetic drift.”
Mapping Surname Origins–Frequency vs. Absolute Numbers
Typically, when plotting the distribution and approximating the origins of a surname by mapping data, its frequency per x number of people in the study area (county, parish or poor law union) is more helpful than simply the number of individuals in a location bearing the surname. The number of individuals method will bias more populated areas where you would naturally expect to find a higher number of people for each surname, while the information you're after is where the surname is strong based on geography itself, not population. Plotting by frequency is referred to as “correcting for population size.”
Once the rate of the surname per area has been determined, the Banwell index can be applied. A Banwell index of <1 means a lower prevalence of the surname in that area than in the country as a whole and an index of >1 means a higher prevalence. For example an area with a Banwell index of 3 has three times the national average prevalence of the surname.
That said, the Banwell index and reporting by frequency as opposed to absolute numbers is more useful with surnames of a higher frequency than Bowes and its variants.