The subjectivity and non-probability based nature of unit selection (., selecting people, cases/organisations, etc.) in purposive sampling means that it can be difficult to defend the representativeness of the sample. In other words, it can be difficult to convince the reader that the judgement you used to select units to study was appropriate. For this reason, it can also be difficult to convince the reader that research using purposive sampling achieved theoretical/analytic/logical generalisation . After all, if different units had been selected, would the results and any generalisations have been the same?
Like all writers, scholars depend upon words used as precisely as possible. In contemporary academic English, “thesis” and “dissertation” are almost interchangeable, and in this book I’ll use them that way merely to provide some variety. A thesis can, of course, be a master’s thesis or an undergraduate thesis, but a dissertation is always written for a doctoral degree. The dictionary’s succinct definition of a dissertation omits any mention of a proposition to be defended, and length seems to be the dissertation’s principal characteristic. A thesis might be very brief indeed. Martin Luther came up with ninety-five of them, and crammed them all onto a document correctly sized for a church door. For modern-day academics, a dissertation is expected to contain a thesis, that is, this lengthy exposition of evidence and analysis is supposed to contain a core argument. It might be said that the thesis inhabits and animates the dissertation. Unfortunately, it sometimes seems, at least to publishers, that the thesis—the heart of the dissertation—has stopped ticking. Argument gone, all that is left is length.