Group project 2

Example of Binary Logistic Regression in SPSS with one independent and one dependent variable

Research Question

Our research question is: to what extent does political party affiliation predict the respondents’ having gun in home

Hypotheses

Null Hypothesis (H0): There is no statistically significant relationship between political party affiliation and having gun in home?

Alternative Hypothesis (HA): There is statistically significant relationship between political party affiliation and having gun in home?

Variables

Independent Variables (IV):

partyid — political party affiliation, measured as categorical, where 0 = Strong Democrat, 1 = Not Strong Democrat, 2 = Independent Near Democrat, 3 = Independent, 4 = Independent Near Republican, 5 = Not Strong Republican’ 6 = Strong Republican, and 7 = Other Party.

Dependent Variables (DV):

Owngun — have gun in home, where 1 = Yes; 2 = No. Prior to conducting the analysis, the variable “owngun” was recoded as 0 = No and 1 = Yes.

Results

The average age of participants is 49.37 ( SD = 19.143) (Table 1).

Table 1

Descriptives Statistics for Age of Respondents

	Statistic	Std. Error
AGE OF RESPONDENT	Mean	49.37	.662
	95% Confidence Interval for Mean	Lower Bound	48.07
		Upper Bound	50.67
	5% Trimmed Mean	49.03
	Median	49.00
	Variance	293.879
	Std. Deviation	17.143
	Minimum	18
	Maximum	89
	Range	71
	Interquartile Range	27
	Skewness	.226	.094
	Kurtosis	-.781	.188

Regression results

At the model level, the results showed the Nagelkerke R square to be .111, which means that 11.1% of the variability in having gun in home is explained by political party affiliation (See Table 2). The analysis showed that political party affiliation was significantly associated with having gun in home ꭕ2(7) = 36.010, p < .001 (See Table 3).

Table 2

Model Summary

Step	-2 Log likelihood	Cox & Snell R Square	Nagelkerke R Square
1	501.181a	.077	.111
a. Estimation terminated at iteration number 4 because parameter estimates changed by less than .001.

Table 3

Omnibus Tests of Model Coefficients

	Chi-square	df	Sig.
Step 1	Step	36.010	7	.000
	Block	36.010	7	.000
	Model	36.010	7	.000

At the variable level, the results showed that the odds of having gun at home are 2.6 times higher for “Independent Near Republican” political party affiliation than “Strong Democrat” political party affiliation. The confidence interval showed that we are 95% confident that the odds-ratio in the population falls between 1.151 and 5.876. Likewise, the odds of having gun at home are 6.627 times higher for “Strong Republican” political party affiliation than “Strong Democrat” political party affiliation. The confidence interval showed that we are 95% confident that the odds-ratio in the population falls between 2.890 and 15.199 (See Table 4).

Table 4

Variables in the Equation

	B	S.E.	Wald	df	Sig.	Exp(B)	95% C.I.for EXP(B)
							Lower	Upper
Step 1a	POLITICAL PARTY AFFILIATION			34.213	7	.000
	POLITICAL PARTY AFFILIATION(1)	-.059	.379	.024	1	.876	.943	.449	1.981
	POLITICAL PARTY AFFILIATION(2)	.438	.410	1.144	1	.285	1.550	.694	3.461
	POLITICAL PARTY AFFILIATION(3)	-.076	.390	.038	1	.846	.927	.432	1.990
	POLITICAL PARTY AFFILIATION(4)	.956	.416	5.277	1	.022	2.600	1.151	5.876
	POLITICAL PARTY AFFILIATION(5)	.617	.409	2.281	1	.131	1.854	.832	4.130
	POLITICAL PARTY AFFILIATION(6)	1.891	.423	19.944	1	.000	6.627	2.890	15.199
	POLITICAL PARTY AFFILIATION(7)	1.341	.758	3.133	1	.077	3.824	.866	16.884
	Constant	-1.341	.272	24.239	1	.000	.262
a. Variable(s) entered on step 1: POLITICAL PARTY AFFILIATION.