. Author manuscript; available in PMC: 2015 Apr 1.

Published in final edited form as: J Am Assoc Nurse Pract. 2013 Jun 18;26(4):187–193. doi: 10.1002/2327-6924.12051

Table 2.

Hierarchical logistic regression of intention to complete an AD decision on age, race, AD knowledge, and group assignment (n=41)

	Predictor	β	SE β	Exp β	p
Step 1	Age (0=50–64 yrs., 1=≥65 yrs.)	−.306	.749	.736	.736
	Race (0=white, 1=nonwhite)	1.091	.755	2.98	.678
	AD knowledge	−.658	.315	.518	.279
Step 2	Age	.775	.930	2.171	.405
	Race	1.984	1.006	7.275	.049^*
	AD knowledge	−.918	.448	.399	.040^*
	Group assignment (0=control, 1=experimental)	3.206	1.608	24.690	.004^**

Note. AD: advance directive. Yrs.: years. Dependent variable: intention to complete an AD.

Model statistics for step 1: R²= .151 (Cox & Snell), .203 (Nagelkerke). Model χ²=6.697, df=3, p=.082. Hosmer & Lemeshow test (χ²=12.408, df=8, p=.137).

Model statistics for step 2: ΔR²= .221 (Cox & Snell), .298 (Nagelkerke). R²= .372 (Cox & Snell), .501 (Nagelkerke). Model χ²=19.074, df=4, p<.0. Hosmer & Lemshow test (χ²=2.748, df=7, p=.907). Overall correct classification was 79% for this model.

p<.05,

^**

p<.01.