This preview shows page 1. Sign up to view the full content.
Unformatted text preview: he following test.
H0: C - S = 0 versus Ha: C - S > 0
The test statistic is
The individual standard errors are n
15 and , so = 25 degrees of freedom will be used. The p- The degrees of freedom are value is P(t25 > 2.10) = .023. Since there was no significance level given, the common default is to use = .05. In that case,
our p-value is less than .05, so we reject the null hypothesis. The data support the beli ef that the cola cans have greater
mean compressive strength than the strawberry drink cans.
30. The data is summarized in the table.
a. The estimated standard error is
individual standard errors are The
4.3 . The degrees of freedom are given by . So the degrees of freedom are = 37. A 99% confidence interval for - will use the critical value t.005;37 = 2.716, and is given by (42.8 – 33.4) 2.716(.947) = 9.4 2.57 = (6.83, 11.97). Since this interval
does not include zero, we are 99% confident that the carbon beams are stronger than the fiberglass beams.
b. No. A 99% confidence upper bound for
will use the critical value t.01;37 = 2.432, and is given by...
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 Wisconsin.
- Spring '04
- Standard Error