Operationalizing Style: Quantifying the Use of Style Shift in the Speech of African American Adolescents

The vast majority of research to date on African American Vernacular English style shift has taken the form of qualitative analyses of individual case studies; however, despite its great success, in focusing on individual rather than group style and style shifting, such work by itself is unable to answer key questions about style and style shift at the level of social groups, communities of practice, and broader based communities. Recent quantitative analyses, such as Craig and Washington's (2006) Dialect Density Measure (DDM), have sought to capture stylistic variation at the group level by analyzing dozens of linguistic features meant to represent a dialect, but use of such large numbers of features severely restricts the types of statistical analyses that can be applied to a given data set and therefore limits the utility of the technique. To test whether a smaller subset of features can be used to quantify stylistic variation, we analyzed a sample of 108 sixth-grade students observed in two conditions that differed in formality. Three measures were used to track changes in style, two large-scale DDMs constructed from a set of more than 40 variables and a subset measure that used only 6 variables. Analyses indicate that the larger DDMs were highly correlated with the subset measure, thus indicating that a small number of features can be used to reliably reflect shifting styles.