Table 5.

Parameter Estimates (and Standard Errors) of the Random-Effects Models^a Assessing the Relationship of Diagnostic Group and Chronologic Age with Baseline Level and Rates of Change Over Time in Sleepiness and Performance and Behavior

Model variable	Sleepiness and performance and behavior variable
	ESS	DS	CBCL–I	CBCL–Eb	Peg rate	PEP-R P	PEP-R FM	PEP-R EH
Effects on initial level of variable
Intercept (TD)c	6.08 (.30)***	11.83 (.28)***	49.96 (1.31)***	55.22 (2.53)***	.48 (.02)***	11.66 (.13)***	12.22 (.18)***	8.77 (.21)***
Autism diagnosis	.87 (.50)†	.46 (.42)	10.73 (1.62)***	8.41 (1.86)***	−.19 (.03)***	−2.70 (.35)***	−3.39 (.42)***	−3.20 (.41)***
Intellectual disability diagnosis	2.12 (.50)***	.69 (.45)	9.42 (1.82)***	8.30 (1.98)***	−.22 (.03)***	−2.20 (.39)***	−3.52 (.47)***	−3.83 (.42)***
Aged (at baseline)	−.12 (.06)*	.11 (.05)*	.72 (.18)***	−.24 (.40)	.04 (.00)***	.15 (.03)***	.45 (.05)***	.68 (.05)***
Age × Autism	NS	ns	ns	.18 (.51)	−.03 (.01)***	ns	−.27 (.12)*	−.40 (.11)***
Age × Intellectual disability	NS	ns	ns	1.14 (.52)*	−.03 (.01)***	ns	−.02 (.12)	−.26 (.11)*
Effects on rate of change during study
Timed	−.27 (.10)**	ns	−.62 (.25)*	−.69 (.24)**	.05 (.01)***	.43 (.07)***	.63 (.06)***	.69 (.07)***
Autism diagnosis	ns	ns	ns	ns	−.03 (.01)*	ns	ns	ns
Intellectual disability diagnosis	ns	ns	ns	ns	−.03 (.01)*	ns	ns	ns

Note. The reported models contain only significant effects, except for some of the terms reflecting the interaction between diagnosis and age. Those estimates without a symbol were not significant, but the corresponding terms had to be included in the model to preserve the interpretability of the interaction with diagnosis.

ESS = Epworth Sleepiness Scale; DS = Daytime Sleepiness subscale of the Children’s Sleep Habits Questionnaire; CBCL–I = Child Behavior Checklist—Internalizing Problems subscale; CBCL–E = Child Behavior Checklist—Externalizing Problems subscale; PEP-R P = Psychoeducational Profile—Revised (PEP-R) Perception Scale; FM = Fine Motor Scale; EH = Eye–Hand Coordination Scale; TD = typically developing.

^aModels included and tested terms for demographic characteristics (ethnicity, mother’s age, marital status, and education).

^bMother’s marital status was a significant predictor; children with unmarried mothers scored significantly higher than those with married mothers.

^cAge at baseline was centered at 44 months, so the intercept can be interpreted as the average response at the first recording time for a 44-month-old TD child.

^dEffects of age, time, and their interaction with diagnosis are reported for 3-month increments.

^†p < .10.

^*p < .05.

^**p < .01.

^***p < .001.