302 Thomas C. Purnell
Table 13.3 Results of Stepwise Discriminant Analysis Using a Forward Method for
Four Test Groups of WI English Speakers and One Control Group
Group, Variable Partial r^2 F p > F
Ave r age
squared
canonical
correlation
1: 1866–1892 (df=4, 41; N voiced=30, voiceless=16)
% glottal pulsing 0.446 35.4 0.0001 0.446*
vowel duration 0.070 3.2 0.0787 0.485*
change in F1 0.070 3.3 0.0756 0.523*
change in F0 0.057 2.5 0.1231 0.550*
2: 1899–1918 (df=4, 58; N voiced=38; voiceless=25)
% glottal pulsing 0.460 51.9 0.0001 0.460*
pulsing duration 0.030 1.8 0.1830 0.476*
C gap duration 0.005 0.3 0.5945 0.478*
V:C ratio 0.002 0.1 0.7658 0.479*
3: 1920–1939 (df=4, 147; N voiced=79, voiceless=73)
% glottal pulsing 0.347 79.6 0.0001 0.347*
vowel duration 0.248 49.1 0.0001 0.509*
change in F0 0.013 1.9 0.1684 0.515*
change in F1 0.006 0.8 0.3640 0.518*
4: Perception (df=4, 55; N voiced=42, voiceless=18)
vowel duration 0.209 15.3 0.0002 0.209*
% glottal pulsing 0.152 10.2 0.0023 0.329*
V:C ratio 0.138 8.9 0.0042 0.421*
pulsing duration 0.063 3.7 0.0600 0.457*
5: 1966–1986 (df=4, 61; N voiced=42; voiceless=24)
% glottal pulsing 0.283 25.3 0.0001 0.283*
vowel duration 0.068 4.6 0.0366 0.331*
change in F0 0.016 1.0 0.3196 0.342*
C gap duration 0.014 0.9 0.3505 0.351*
(continued)