Table 4. Results from logistic regression analyses predicting placement in the low group on five language and literacy outcomes measures for children in the recovered group (separate univariate analyses).
| Outcome Measure (n pairs) | Sex |
SES |
Voc-2 |
Voc-4 |
Gram-4 |
Abstr-4 |
||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B (SE) | OR | B (SE) | OR | B (SE) | OR | B (SE) | OR | B (SE) | OR | B (SE) | OR | |
| Oral language at age 7a (151) | 1.04* (0.51) | 2.8 | −0.67* (0.28) | 0.51 | −0.09 (0.07) | 0.91 | −0.02 (0.03) | 0.98 | 0.54 (1.07) | 1.71 | 0.06 (0.13) | 1.06 |
| Word decoding at age 7b (151) | −0.62 (0.77) | 0.54 | −0.82* (0.38) | 0.44 | −0.07 (0.09) | 0.93 | 0.08 (0.05) | 1.09 | — c | 0.00 (0.18) | 1.00 | |
| Receptive language at age 12d (104) | 1.00 (0.75) | 2.71 | −1.76* (.62) | 0.17 | −0.02 (0.10) | 0.98 | 0.06 (0.05) | 1.06 | 1.13 (1.31) | 3.10 | −0.27 (0.22) | 0.76 |
| Word decoding at age 12e (102) | — c | −0.87 (0.99) | 0.42 | −0.11 (0.17) | 0.89 | 0.00 (0.08) | 1.00 | — c | 0.00 (0.41) | 1.00 | ||
| Reading comprehension at age 12f (122) | 0.70 (0.63) | 2.0 | −0.13 (0.37) | 0.88 | 0.13 (0.10) | 1.14 | 0.03 (0.05) | 1.03 | — c | −0.07 (0.17) | 0.94 | |
Note. Analyses are based on one randomly chosen twin per pair. Voc-2 = age 2 vocabulary (severity of early delay); Voc-4 = age 4 vocabulary; Gram-4 = Grammar score at 4; Abstr-4 = abstract language use at 4; OR = odds ratio.
a
Cox and Snell R2 = .077; Nagelkerke R2 = .139; model ×2(6) = 12.09,p = .06.
b
Coxand Snell R2 = .077; Nagelkerke R2 = .188; model ×2(6) = 12.02, p = .06.
c
Parameter cannot be estimated due to a low cell frequency or high multicollinearity
d
Cox and Snell R2 = .156; Nagelkerke R2 = .319; model ×2(6) = 17.70, p = .007.
e
Cox and Snell R2 = .043; Nagelkerke R2 = .185; model ×2(6) = 4.50, p = .61
f
Cox and Snell R2 = .048; Nagelkerke R2 = .102; model ×2(6) = 6.06, p = .42.
*
p < .05.