Chapter# 8 Exercises:

1.a: Matched Pairs are more appropriate. Measures are repeated over time

1.b: The matched pairs report for June 1999- March 1999 shows a difference as seen below of 0.05:
[pic] [pic]

1.c: The matched pairs reports for Aug-June and Aug-March show no significant differences:
[pic]

2-a The grouped means method would not be the appropriate test since statistical assumptions for the t test for groups are not satisfied with correlated data. The difference can be detected much better with a paired t-test and you may be overstating the significance if you used a group t-test rather than a paired t-test.

2-b MATCHED PAIRS: There is a small ratio of 0.1218 difference Kiln compared to regular.
[pic]

3-a There is an outlier in sales for 63,438 computer sales.

[pic]

3-b grouped means.

3-c: |t| 0.8874 (very small difference)

[pic]

3-d Outliers: It would not be appropriate to remove the outlier because the difference in the analysis would be unfavorable due to the sale being such a drastic amount and the difference reported in question “c” being already reported as a small difference.

4-a: Yes, there is a significant difference between 2 door and 4 door impact on left legs.

[pic]

4-b: The difference on compression on right legs between 2 and 4 door vehicles is not significantly different.
[pic]

4-c The difference is very small on Head IC between 2 and 4 door cars.

[pic]

5-a We will be using a matched set of variables to determine if their difference is zero or different than zero.
5-b and 5-c

[pic]

The matched pairs t test analysis Prob>|t| 0.1107, is insignificant. In the t test above you will see that we get the same response: t ratio 1.65, and the