The Frequency and Degree of “intrusive /r/” in New Zealand English 55That is—is social class playing a signi¿ cantly separate role for each, or is
there a more direct relationship between frequency and degree? We re¿ t the
model predicting the F3 of the intrusive /r/, this time including the rate of /r/
insertion for that speaker as a predictor. The effects of the “real” /r/ F3 and
the base word remained signi¿ cant. Social class was no longer signi¿ cant;
instead the rate of /r/ insertion proved a signi¿ cant factor (p<.01). This effect
is shown in Figure 2.9.
The more frequently an individual produced an intrusive /r/, the lower
their F3 was when they did so. Both high rates of intrusive /r/ and low F3s are
associated with speakers from lower classes. This may be because social class
independently affects both of these factors. Or it may be because there is some
more direct link between them (e.g., high rates of intrusive /r/ somehow more
directly lead to lower F3s). That the rate of /r/-insertion is a better predictor of
Figure 2.9 Model predictions for the F3 of /r/, on the basis of the percent /r/ produced
by that speaker. Dashed lines represent 95% con¿ dence intervals.