ABSTRACT
In the broadest terms, there are two possible types of explanation for the influence of education in the data analysed here. In the first, if the relationship between education and the responses of the participants in the test is direct and straightforward, then the results can be explained by hypothesising that in school, speakers are taught that only singular agreement is acceptable. In the second, if the relationship between education and the cloze test results is indirect, a qualitative analysis may be more revealing. This chapter reports on analyses which explore these two possibilities. It presents a survey of school grammars, which aimed to establish what the norms taught in school about this kind of agreement are, and whether these have changed over time. The chapter discusses a qualitative analysis of some of the sociolinguistic study data, which looks at some individual speakers of interest. The chapter presents a qualitative follow-up study, in which a subset of participants are re-interviewed.
