Today I gave a talk that Betsy Sneller and I have been working on called “How Sample Size Impacts Pillai Scores – and What Sociophoneticians Should Do About It” at the 50th New Ways of Analyzing Variation conference in San Jose! This is an update to what we presented at ASA2021.

We have three things you can download.

  1. First is the actual powerpoint file. In the notes of each slide though you can see the actual script I read, so you can read every word that was said during the talk.
  1. Next, if that’s too much for you, you can download just a PDF of the talk. In case you want this ligherweight version of the slides.
  1. Finally, here is the current manuscript that is under review with the Journal of the Acoustical Society of America. The final product will likely change somewhat, but most of the information is there. [Edit (December 14, 2022): Here is the accepted veresion.]

If you need to calculate Pillai scores in R, I’ve got a two-part tutorial for you (here and here). I also did a blog post (here) about how Pillai scores don’t seem to change after normalization.