Our presentation will report on the recent activities of the Spanish Linguistic Variation in Digital Spaces (SLViDS) research group, which involves faculty and students (in varying combinations) from UB, la Universidad Isabel I in Burgos, Spain, and North Carolina State U. We compiled two corpora of digital Spanish text from (global) Twitter and el Corpus del Español del Siglo XXI (CORPES) from queries that targeted conjugations of certain classes of Spanish verbs selected on the basis of research questions related to Spanish morphology, syntax, semantics, and discourse. SLViDS is comprised of two teams: the programing/coding solutions team and the token-for-token qualitative analysis team. The former was responsible for extracting the text with Python, using spaCy to do part-of-speech and syntactic dependency tagging, and offering solutions to filter out tweets that did not contain an instance of the relevant linguistic variable. The latter team looked at each tweet and analyzed the morphosyntactic structure of the clauses it contains to determine if it constitutes an instance of the relevant linguistic variable and to categorize and tag its outcome; they also communicated to the programming/coding solutions team how this qualitative analysis procedure might be made more efficient by making it easier to spot the relevant linguistic variable in certain parts of the clause. Eventually, we will develop statistical models and data visualizations that will help us to understand dialectal patterns of variation & change in Spanish morphosyntax, semantics, and discourse, and to reflect on their cognitive and sociolinguistic underpinnings. Our presentation will introduce the linguistic variable of interest and walk you through our research process, highlighting our methodological challenges.

See all DSSN Events