ABSTRACT
The neglect of sociolinguistic patterns of language variation in U. S. Spanish is due at least in part to the abiding preoccupation with contact-induced change and a methodological predilection for acceptability judgments, experimental tasks, or cherry-picked examples. This chapter explores the scant number of reports on social factors in U. S. Spanish, first for stable linguistic variables—those with distribution patterns that persist across time and communities—and then for possible changes in progress—where age distributions are gradient. It describes the statistical procedures, such as principal component analysis and regression analysis, to a community-based corpus of New Mexican Spanish to infer and test social factors relevant to conditioning language variation. A contributing factor to the paucity of studies has been the familiar problem that social characteristics of speakers are generally less well defined than linguistic categories, particularly in minority language situations, and that social factors are often highly interdependent.
