Question

The data summarized in Table 23.6 (from a study by Goodwin, Schylsinger, Hermansen, Guze, & Winokur, 1973) show how “becoming a heavy smoker” is related to “having a biological parent who was diagnosed alcoholic.” Either enter these data into SPSS by hand or use the dataset goodwin.sav (available on the website for the textbook). Run a binary logistic regression to predict smoking status from parental alcohol diagnosis and write up your results, including tables that summarize all relevant whose biological parent is diagnosed with alcoholism have significantly higher odds of being heavy smokers?

parent not diagnosed with alcohol parent diagnosed with alcohol

child not a heavy smoker (y = 0) 41 6

child is a heavy smoker 37 49

Answer 1

The data summarized in Table 23.6 (from a study by Goodwin, Schylsinger, Hermansen, Guze, & Winokur, 1973) show how “becoming a heavy smoker” is related to “having a biological parent who was diagnosed alcoholic.” Either enter these data into SPSS by hand or use the dataset goodwin.sav (available on the website for the textbook). Run a binary logistic regression to predict smoking status from parental alcohol diagnosis and write up your results, including tables that summarize all relevant whose biological parent is diagnosed with alcoholism have significantly higher odds of being heavy smokers?

parent not diagnosed with alcohol parent diagnosed with alcohol

child not a heavy smoker (y = 0) 41 6

child is a heavy smoker 37 49

smoker	alcohol	freq
0	0	41
0	1	6
1	0	37
1	1	49

we entered the table data in spss as 3 columns. We weight the data by the variable freq.

spss command:

LOGISTIC REGRESSION VARIABLES smoker

/METHOD=ENTER alcohol

/CLASSPLOT

/PRINT=CI(95)

/CRITERIA=PIN(0.05) POUT(0.10) ITERATE(20) CUT(0.5).

SPSS output:

Model Summary
Step	-2 Log likelihood	Cox & Snell R Square	Nagelkerke R Square
1	145.833a	.183	.252
a. Estimation terminated at iteration number 5 because parameter estimates changed by less than .001.

Classification Tablea
	Observed		Predicted
			smoker		Percentage Correct
			No smoker	smoker
Step 1	smoker	No smoker	41	6	87.2
		smoker	37	49	57.0
	Overall Percentage				67.7
a. The cut value is .500

Variables in the Equation
		B	S.E.	Wald	df	Sig.	Exp(B)	95% C.I.for EXP(B)
								Lower	Upper
Step 1a	alcohol	2.203	.488	20.344	1	.000	9.050	3.475	23.568
	Constant	-.103	.227	.205	1	.651	.902
a. Variable(s) entered on step 1: alcohol.

Odds ratio = 9.05 and p value is 0.000 which is significant.

Biological parent is diagnosed with alcoholism have significantly 9.05 times more of being heavy smokers compared to non alcohol parents.