Getting in a Flap or Turning off the Tap in Dunedin?

T-flapping data from the formal-register Dunedin Survey of NZE are combined with Holmes' analysis of this variable in interview and casual conversational style drawn from the Wellington Corpus of Spoken New Zealand English. My goal was to see how this variable was related to some 15 other previously analysed phonological variables, and to arrive at an overall stylistic portrait of this variable in NZE. The two datasets fit together well, even though Holmes used Varbrul in her analysis while I relied on factor analysis and MANOVA. A number of Holmes' specific hypotheses are supported by the more formal Dunedin data.