Table 2.
MANOVA results.
| Sorted Dimension | P | % Variance | Function Loadings |
|---|---|---|---|
| 1 | < 10-3 | 42.1 | 0.860 Mean S + 0.676 Mean F + 0.363 Min F |
| 2 | < 10-3 | 22.2 (64.3) | 1.011 SD S - 0.787 Dur - 0.493 Q3 |
| 3 | < 10-3 | 11.6 (75.9) | 1.449 SD S - 1.200 Ent - 1.187 Mean F |
| 4 | < 10-3 | 8.2 (84.1) | 1.393 Skew + 0.870 Ent - 0.831 Kurt |
| 5 | < 10-3 | 6.5 (90.6) | 1.142 Skew + 1.001 CV F + 0.733 Q2 |
| 6 | < 10-3 | 3.2 (93.8) | 1.539 Kurt + -1.537 Max F - 1.520 Skew |
| 7 | < 10-3 | 2.6 (96.4) | -1.304 Mean F + 1.036 Q3 + 0.937 Q1 |
| 8 | 0.006 | 1.3 (97.7) | -7.824 Mean S + 3.597 Q3 + 2.053 Q2 |
The table shows the loading factors for the 8 significant linear discriminant functions that can be used to classify giggles notes by individual. Since the discriminant analysis is performed on z-scores and the functions are normalized so that the within-individual variance is 1, the coefficient of the loadings can be used to measure the effect of each acoustical parameter. The % variance is the percent of the between-individual variance of means that is explained. The number in parentheses is the cumulative variance explained.