Workshopping: How to Model Variation
As a variationist sociolinguist, my research focuses on the way language varies and changes in communities of speakers and concentrates in particular on the interaction of social factors (such as a speaker’s gender, ethnicity, age, degree of integration into their community, etc) and linguistic structures (such as sounds, grammatical forms, intonation features, words, etc). As such, when trying to visually represent and statistically model the effects of various variables, many a variationist has simply defaulted to a statistician, throwing in the towel and hanging our head in shame.
Thankfully, I think my field has found a savior and is finally able to walk out of the dark ages. I attended a workshop at the latest conference I attended (NWAV 45) and learned how to model variation through the use of an interactive application built as a Shiny app with various statistical and graphical R packages (a programming language I am already quite familiar with). The Language Variation Suite “allows one to handle imbalanced data, measure individual and group variation and rank variables according to their significance”. The best thing about this suite is that it is easy to use and requiring minimal programming skills. Much of the interface requires only a drag/drop process.
I am in the process of writing my second qualifying paper for my degree and have low-key been procrastinating because I have had no idea how to deal with the interaction of all the variables I care about. This suite has come at the perfect time in my academic career and I can’t wait to try it out!