test whether the two variables relig16 and letdie1 are independent or if the belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family significantly depend on their religion.

SOCI332Statistics_Agni_7thNovForum6.docx

Home >Social Science homework help >Sociology homework help >test whether the two variables relig16 and letdie1 are independent or if the belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family significantly depend on their religion.

Association between Religion and Belief on Euthanasia

We are testing the strength of the correlation between religion (relig16, RELIGION IN WHICH RAISED) and the belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family (letdie1, ALLOW INCURABLE PATIENTS TO DIE).

The research problem is to test whether the two variables relig16 and letdie1 are independent or if the belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family significantly depend on their religion.

Thus, the appropriate hypotheses to be tested is:

Null hypothesis H0: The belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family (letdie1) is independent of their religion (relig16).

Alternative hypothesis H1: The belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family (letdie1) significantly depend on their religion (relig16).

We use chi-square test statistic to test this hypothesis at 0.05 level of significance.

*Religion in which raised Allow incurable patients to die Crosstabulation**
Count
	Allow incurable patients to die	Total
	YES	NO
Religion in which raised	PROTESTANT	424	210	634
	CATHOLIC	279	135	414
	JEWISH	16	2	18
	NONE	74	27	101
	OTHER	5	0	5
	BUDDHISM	2	0	2
	HINDUISM	7	2	9
	MOSLEM/ISLAM	6	1	7
	ORTHODOX-CHRISTIAN	5	4	9
	CHRISTIAN	25	16	41
	NATIVE AMERICAN	1	1	2
Total	844	398	1242

Chi-Square Tests
	Value	df	Asymp. Sig. (2-sided)
Pearson Chi-Square	11.892a	10	.292
Likelihood Ratio	14.875	10	.137
Linear-by-Linear Association	.010	1	.920
N of Valid Cases	1242
a. 10 cells (45.5%) have expected count less than 5. The minimum expected count is .64.

We observe that if we apply a chi-square test without any re-coding, 45.5% cells have expected count less than 5 (as is mentioned in the chi-square output table above).

Thus, the chi-square test fails to meet the desired requirement and so, we use re-coding to combine certain categories of religion (with expected counts less than 5).

*Religion in which raised Allow incurable patients to die Crosstabulation**
Count
	Allow incurable patients to die	Total
	YES	NO
Religion in which raised	PROTESTANT	424	210	634
	CATHOLIC	279	135	414
	JEWISH	16	2	18
	NONE	74	27	101
	OTHER	21	4	25
	CHRISTIAN	30	20	50
Total	844	398	1242

Then, we apply chi-square test of independence to obtain the following result:

Chi-Square Tests
	Value	df	Asymp. Sig. (2-sided)
Pearson Chi-Square	9.739a	5	.083
Likelihood Ratio	10.851	5	.054
Linear-by-Linear Association	.005	1	.943
N of Valid Cases	1242
a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 5.77.

The p-value of the 2-sided chi-square test of independence is 0.083

p-value (0.083) > level of significance (0.05)

Since p-value > level of significance, we fail to reject the null hypothesis (H0).

Thus, we conclude that the belief that doctors should be allowed to euthanize a patient with an incurable disease on request of the patient and the patient’s family (letdie1) is independent of their religion (relig16).