. Author manuscript; available in PMC: 2012 Aug 1.

Published in final edited form as: Dev Psychopathol. 2011 Aug;23(3):831–844. doi: 10.1017/S0954579411000332

Table 3.

Parameter estimates for three-class model

	Moderate Arousal Class	Hyperarousal Class	Hypoarousal Class
Sample size	32	42	79
Avg. probability of class membership	.92	.97	.95
Results
Latent variable means
Intercept mean, g0	2.64^**	3.39^**	2.14^*
Slope mean, g1	2.67^*_a	2.92^**_a	1.43^**_a
Slope loadings
Episode 1	= 0 fixed	= 0 fixed	= 0 fixed
Episode 2	.39 (.07)	.49 (.07)	.32 (.10)
Episode 3	.79 (.14)	.94 (.14)	.72 (.14)
Episode 4	.20 (.06)	.08 (.06)	.23 (.08)
Episode 5	= 1 fixed	= 1 fixed	= 1 fixed
Episode 6	.61 (.13)	.59 (.13)	.69 (.15)
Episode 7	.39 (.10)	.59 (.12)	.36 (.09)
Residual variance	0.34 (0.17)	2.53 (0.33)	1.41 (0.15)
Insensitivity average	2.03_c	2.92_c	4.20_c
Disengagement average	1.25_d	1.56_d	2.06_d
Intrusive average	2.12_ef	2.87_e	2.79_f
Harsh/hostile average	1.20_g	1.85_g	1.45_g

Note: Values with common subscripts denote significant between-group differences as examined by theWald test of significant differences at p < .05.

p < .05.

^**

p < .01.