Statistics meets Linguistics

Statistics is the basis for any kind of data analysis. On the one hand, it is a means of describing data samples (descriptive statistics). On the other hand, statistical approaches allow for generalizations from a data sample to the general population (inferential statistics). Such generalizations are needed to determine whether the results we draw from our data sample have general value, e.g. for subjects not part of the data sample but part of the population we investigate (e.g. all speakers of a particular variety of English). For modern quantitative linguistics (as well as for many daily situations and tasks), statistical approaches have become an indispensable element of analysis. In this class, we introduce students to both descriptive and inferential statistics. We combine the two approaches to understand and interpret statistical distributions, relationships between variables (such as the realization of linguistic characteristics and age, gender, etc.), and the prediction of such variables. Starting from linguistic research questions, we present different statistical methods to investigate and answer such questions. The class will be co-taught by a statistician and a linguist in order to demonstrate how statistics meets linguistics.