This preview shows page 1. Sign up to view the full content.
Unformatted text preview: are paired by Test Condition
(“position on the original cylinder”?) The scatterplot seem s to
indicate that the pairing variable will not prove very useful.
b. A 95% confidence interval is given by Normal
4.34 B ox Plot of multiple variables
Normal and High Strength Concrete Peak Stress
100 90 80 70 . A normal probability plot 60 50 of the differences indicates the validity of using the t distributi on.
The huge distance of this 95% confidence interval from zero
indicates that there is a big difference between the mean peak stress for the normal versus high stress mixes.
40 Normal High 43. The statistics are in the table. The standard error is 5.98.
a. The normal probability plot does not exhibit noteworthy curvature, so there is no
reason to doubt that the data are not normal.
b. The lower confidence bound uses t.05;14 = 1.761, and is given by -38.6 - 1.761(5.98) = -38.6 - 10.5 = -49.1. We are 95%
confidence that the mean number of months between the onset of Cushing’s disease and its diagnosis is no greater than
View Full Document
This note was uploaded on 09/17/2012 for the course STAT 312 taught by Professor Chung during the Spring '04 term at University of Wisconsin.
- Spring '04
- Standard Error