. Author manuscript; available in PMC: 2010 Jan 1.

Published in final edited form as: Int J Methods Psychiatr Res. 2009 Jun;18(2):69–83. doi: 10.1002/mpr.279

Table 5.

Chi-square values of the main effects and interactions of design effects and socio-demographic variables in predicting lifetime and 12-month disorder in the NCS-A (n=10,123)^a

	Lifetime Disorders					Twelve-Month Disorders
	Mood ^b	Anxiety	Impulse-control^b	Substance	Any^b	Mood^b	Anxiety	Impulse-control^b	Substance	Any^b	df
I. Main effects of clusters and weights
Clusters	140.3^*	260.7^*	157.8^*	356.1^*	186.9^*	126.3^*	210.3^*	144.5^*	262.9^*	175.2^*	83
Weights	0.1	1.7	0.4	0.1	3.7	0.3	2.9	2.0	0.0	3.4	1
II. Main effects of demographics
Age	79.8^*	10.5^*	12.8^*	589.7^*	79.5^*	53.7^*	12.4^*	1.2	363.0^*	54.2^*	1
Sex	52.4^*	140.8^*	68.0^*	32.6^*	2.0	62.1^*	179.6^*	32.9^*	30.9^*	11.6^*	1
Race-ethnicity	18.7^*	50.5^*	18.9^*	79.7^*	26.9^*	24.1^*	49.8^*	12.4^*	62.4^*	19.3^*	3
Parent education^c	8.3^*	57.9^*	76.9^*	23.9^*	63.6^*	9.6^*	61.9^*	64.4^*	13.8^*	59.9^*	3
III. Interactions involving weights
Age	0.0	4.3^*	0.1	4.9^*	0.6	0.4	7.0^*	0.5	4.5^*	3.1	1
Sex	0.3	0.1	2.3	3.9^*	2.1	0.9	0.2	1.1	4.0^*	0.3	1
Race-ethnicity	3.0	3.2	1.7	0.5	0.4	1.1	2.0	1.7	1.5	0.8	3
Parent education	5.4	5.9	1.6	1.3	0.5	3.0	10.6^*	0.5	1.8	0.2	3
IV. Interactions involving clusters
Age	198.2^*^d	94.1	114.3^*	777.2^*^d	99.0	168.1^*^d	91.1	147.5^*^d	522.4^*^d	86.2	83
Sex	193.7^*^d	99.4	75.3	380.2^*^d	71.3	188.5^*^d	82.7	80.1	287.4^*^d	87.3	83
Race-ethnicity	284.9^d	447.3^*^d	307.8^*^d	298.5^*^d	320.5^*^d	285.1^d	386.7^*^d	274.9^d	178.2^d	327.9^*^d	249
Parent education	352.2^*^d	254.9	367.7^*^d	162.2^d	430.2^*^d	291.9^*^d	479.6^*^d	326.5^*^d	347.5^*^d	419.2^*^d	249

Significant at 0.05 Level

There are n=10,123 observations with weights and design effects. Each effect is looked at through a bivariate model.

These disorders use information parent SAQs and consequently are limited to the 6483 cases with completed PSAQs

Parent education is defined as the maximum number of years of education of either parent. Categories are: 0–11 years, 12 years, 13–15 years, 16+ years

Due to sparse data, some models were run with a forward regression logistic model that stepped in two-way interactions included at an entry criterion of p = .95 (i.e., even very small coefficients were entered). The number of degrees of freedom in these models is lower than in the saturated model.