ABSTRACT
Written surveys, or questionnaires, have long been an important means of gathering data on linguistic variation. The idea is simple enough: if you want to find out which words people use, or how they pronounce those words, or whether they find certain sentences grammatical, write up a list of the questions you want answers to, distribute it to members of the population whose speech you want to find out about, and ask them to fill it out and return it to you. This approach has been more commonly used in dialectology than in sociolinguistics, which reflects a belief that regional differences are less socially sensitive than social differences. Many respondents enjoy reflecting on how their speech differs from that of other regions but become self-conscious when asked about differences tied to levels of education or occupation, making responses to direct questioning more reliable and valuable in the former case than in the latter. While this belief is largely correct, I will argue that written surveys, which have advantages that complement their disadvantages, retain a useful role even in sociolinguistic methodology when they are deployed appropriately for certain purposes. The following discussion examines the advantages and disadvantages of surveys and the history of their use, with a special focus on the study of Canadian English, in which written questionnaires have played a major role. The chapter concludes with some methodological considerations that sociolinguists should bear in mind if they wish to maximize the benefits and minimize the drawbacks of surveys.
