Table 7. Multivariate and univariate linear models, investigating whether ~4.35 Hz PSD power at 11 months predicted either infant-led or parent-estimated language outcomes.
The table details the multivariate linear models followed by the univariate linear models describing whether ~4.35 Hz PSD power at 11 months predicted a) infant-led language measures or b) the parent-estimated language measures. Bonferroni correction for multiple comparisons in the univariate models led to modified significant alpha levels of, p = <0.01 for infant-led and p = <0.025 for parent-estimated (denoted by bold italic text). Modified alpha level for trends are p = <0.02 for infant-led and p = <0.05 for parent-estimated (denoted by bold text).
| ~4.35 Hz PSD power at 11 months a, Infant-led language measures |
||||
|---|---|---|---|---|
| Multivariate linear model | Df num | Df denom | F | p |
| Infant-led language measures | 5 | 65 | 1.76 | 0.13 |
| Univariate linear models | Df | β estimate | SE | p |
| NWR Consonants, 24 months | 1,79 | -4.93x10-4 | 3.69x10-4 | 0.19 |
| NWR Syllables, 24 months | 1,79 | -8.36x10-4 | 3.65x10-4 | 0.025 |
| NWR Stress, 24 months | 1,79 | -5.31x10-4 | 3.40x10-4 | 0.12 |
| CCT, 18 months | 1,79 | -1.62x10-4 | 2.76x10-4 | 0.56 |
| Pointing, 12 months | 1,90 | -2.95x10-4 | -5.98x10-4 | 0.62 |
|
b, Parent-estimated language measures |
||||
| Multivariate linear model | Df num | Df denom | F | p |
| Parent-estimated language measures | 2 | 75 | 3.18 | 0.047 |
| Univariate linear models | Df | β estimate | SE | p |
| CDI comprehension, 24 months | 1,76 | -0.086 | 0.209 | 0.68 |
| CDI production, 24 months | 1,76 | -0.346 | 0.215 | 0.11 |