Discussion Week 5_ HYPOTHESIS TEST FOR A MEAN

Response 1: Instructions

· Change the alpha value to 0.10. Notice that this change will not alter any of the test data in Excel.

· Using alpha=0.10, which hypothesis is supported? Write a conclusion statement, comparing appropriate Excel data to alpha, to justify your choice.

· Is this the same choice your peer made when alpha=0.05 was used?

Janene Turner

WEEK 5 DISCUSSION

In the first test, I used a t-Test: Two-Sample Assuming Unequal Variances, because the samples were not equal and also because the variances were unknown. This provided me with the variance I needed to perform the z-test. This would be a two tailed test; it is two tailed because we are looking for a change in the parameters that would indicate statistical significance, for this test, we are looking to see if the BMI of smokers have a greater BMI than nonsmokers.   We use the two tailed test when the null hypothesis should be rejected if the test value dips into a critical area on either side of the mean.

 

t-test (to obtain the variances needed for the z test)

 

t-Test: Two-Sample Assuming Unequal Variances
	SMOKER	NON-SMOKERS
Mean	27.9969697	29.56071429
Variance	28.789678	35.87876984
Observations	33	28
Hypothesized Mean Difference	0
df	55
t Stat	-1.065524
P(T<=t) one-tail	0.14564711
t Critical one-tail	1.67303397
P(T<=t) two-tail	0.29129423
t Critical two-tail	2.00404478

 

z- test ( to determine statistical significance)

z-Test: Two Sample for Means
	SMOKER	NON-SMOKERS
Mean	27.9969697	29.5607143
Known Variance	28.789678	35.8787698
Observations	33	28
Hypothesized Mean Difference	0
z	-1.065524
P(Z<=z) one-tail	0.14331944
z Critical one-tail	1.64485363
P(Z<=z) two-tail	0.28663889
z Critical two-tail	1.95996398

 

We will use the critical number of 2 (because 1.9 is the higher of the two critical numbers) to compare with the t stat value of  -1.06. Because this number is less than 2, we can accept the hypothesis.  P values are 0.14564711, and 0. 29129423.Both numbers are greater than our alpha number of 0.05, meaning we can accept the hypothesis. Due to these findings, we can say that smokers do not have a BMI that is greater than non-smokers.