1. Happiness comes not from material wealth but less desire.

2. Inferences Comparing Two Population Means
Paired samples
Independent samples
Estimation
Tests- z test and t test

3. Paired Samples
If the same set of sources are used to obtain data representing two populations, the two samples are called paired. The data might be paired:
As a result of the data from certain “before” and “after” studies
From matching two subjects to form “matched pairs”

4. Tests for Paired Samples
Calculate the pair differences
Proceed as in one sample case
Example: Figure 11.1

7. Independent Samples The two samples are unrelated
Example: Chapter 11, problem 15 (heart.mpj)

8. Estimation for m1-m2 Point estimator Confidence interval
Normal populations with known s1,s2, or two large samples (n1,n2>30): Z interval
Normal populations with unknown s1,s2: t interval
s1=s2: pooled t interval
s1=s2: approximate t interval
At least one nonnormal population and at least one small sample: out of our scope

9. Two Populations

10. Tests for m1-m2 = d0
Normal populations with known s1,s2, or two large samples (n1,n2>30): Z test
Normal populations with unknown s1,s2: t test
s1=s2: pooled t test
s1=s2: approximate t test
At least one nonnormal population and at least one small sample: out of our scope

11. Two Populations

12. Minitab: stat>>basic statistics>>2 sample t …

13. Results for: HEART.MTW
Two-Sample T-Test and CI: pdi, trtment
Two-sample T for pdi
trtment N Mean StDev SE Mean
Difference = mu (0) - mu (1)
Estimate for difference:
95% CI for difference: (-11.02, -0.68)
T-Test of difference = 0 (vs not =): T-Value = P-Value = DF = 141
Both use Pooled StDev =
95% CI for difference: (-11.01, -0.70)
T-Test of difference = 0 (vs not =): T-Value = P-Value = DF = 140