Last Time: The t-distribution; practice
with type I and type II errors
The t-distribution is discussed in section 8.5; you can
see a comparison of the normal and the t on page 253.
Major issue: we want to use the sample standard
deviation, rather than the population standard
deviation, which is generally unknown.
Major twist: degrees of freedom is now required.
As in the case of the Z-value, we need the underlying distribution
to be normal for this to be exactly right
If the underlying distribution is roughly normal, then we can
proceed with confidence if the sample size is big enough
(p. 361). You need to address this issue when you solve
problems with t-statistics!
Notes:
Recall that we have a test coming up next week.
Collect chapter 11 homework (both assignments).
Today: Finish t-distribution: the t-distribution replaces the Z
(normal) distribution (Sections 12.1-2, w/ Minitab)
Example problems:
12.24 and 12.25 (Calc -> Probability Distributions)
#12.32 (test of hypothesis, using Minitab: Stat -> Basic Statistics -> 1-sample t)
Compare versus a mu of 18 as well (use table).
#12.44
(constructing a confidence interval; by hand, and using
Minitab)