Skip to main content

Table 3 Receptive and expressive language models

From: Language delay aggregates in toddler siblings of children with autism spectrum disorder

  Unstandardized coefficients Standardized coefficients T Sig. 95.0% confidence interval for B
B Std. error Beta Lower Upper
Receptive language model
 (Constant) 51.19 2.75   18.65 < .001 45.79 56.59
Sex 3.57 1.05 .18 3.39 .001 1.50 5.65
 Maternal education 2.08 1.27 .092 1.63 .103 − .43 4.59
Income 3.27 1.13 .16 2.89 .004 1.05 5.49
 Race − 2.83 1.55 − .094 − 1.82 .069 − 5.88 .23
ASD risk status − 4.69 1.13 − .22 − 4.15 .000 − 6.91 − 2.47
Expressive language model
 (Constant) 43.13 3.05   14.13 < .001 37.12 49.13
Sex 3.50 1.17 .16 2.98 .003 1.19 5.80
Maternal education 3.72 1.42 .15 2.63 .009 .94 6.51
Income 2.55 1.26 .11 2.02 .044 .072 5.02
 Race − .043 1.73 − .0010 − .025 .98 − 3.44 3.35
ASD risk status − 3.19 1.26 − .14 − 2.54 .011 − 5.66 − .72
  1. In these models, language (either receptive or expressive) is the dependent variable. ASD-risk status [0 = low-risk sibling without ASD, 1 = high-risk sibling without ASD] is the independent variable. Covariates of sex (0 = male, 1 = female), maternal education (0 = no college, 1 = college or greater), income (0 = <$75,000/year and 1 = ≥$75,000/year), and race (0 = Caucasian, 1 = non-Caucasian) are entered first into the model, followed by risk status. Bolded variables demonstrate a significant relationship with language. For receptive language, sex, income, and ASD risk are significant, with females and higher family incomes associated with higher scores. ASD risk status shows the greatest impact, and greater ASD risk is associated with lower language scores. For expressive language, all variables except race are significant and show similar relative impact. Female sex, higher maternal education, and higher family income are associated with higher scores, while greater ASD risk is associated with lower expressive language scores. ASD autism spectrum disorder, std. standard, sig. significance