- Subject index
The chapters in this innovative book introduce the quantitative analysis of linguistic survey data with specific reference to the Linguistic Atlas of the Middle and South Atlantic States (LAMSAS).
Topics covered include: the relation of sociolinguistics to the original conception of LAMSAS; the adaptation of LAMSAS to the needs of computerization and the research methods envisioned; the mechanics involved in computerizing LAMSAS; how to handle and analyze the data in the database management system; the creation of categories for analysis; and the logic of statistical testing.
Chapter 3: From Atlas to Database Structure
From Atlas to Database Structure
A questionnaire-based project like a linguistic atlas appears to lend itself fairly easily to computerization and automatic analysis because its data structure resembles the one that computers and database management systems prefer or require: a table. In fact, the first fascicles of LAMSAS (McDavid and O'Cain 1980) were published in tabular form, and Hans Kurath had been making and referring to handwritten list manuscripts, edited listings of responses to a question ordered by informant, since the 1930s (see Figure 3.1). Theoretically, the data set for any questionnaire study could be arranged quite easily in a simple matrix structure; for the atlas, the informants (or, more generally, localities) would constitute the rows, and the linguistic items elicited ...