This chapter describes the method used by the author to estimate the county of birth of English and Welsh emigrants in the late nineteenth century. It discusses the problems inherent in the data, the assumptions that were necessary and the degree of error in the final estimates. The chapter contains no sophisticated mathematics but it would be foolish to pretend that it makes easy (or even, interesting) reading. Any reader who is prepared to take the author's estimates largely on trust could read the next section which explains how they are derived in general terms and then proceed to chapters 5 to 10 which discuss how the new data affect our knowledge of English and Welsh migration history. It should be remembered, however, that almost all the conclusions of the book stand or fall by the accuracy of these estimates – although as will be explained they are sufficiently robust to stand a degree of error. In other words, the main conclusions of the book do not depend on the estimates being exceptionally accurate.

The method used to estimate the county of birth of the emigrants is, in theory, extremely simple. The first stage is to estimate the migration of natives out of a particular county, in a decade. That is, the movement of people who had been born in that county who left it for other counties or to go overseas. The second stage is to estimate the movement of these natives into all the other counties of England and Wales.