•3/7/11
•1
Lecture 13: March 7, 2011
•
Exam 1 will be returned during discussion this Friday.
•
Homework 9: onefactor ANOVA
– One problem: Number 11.2, page 449
– It is a hypothesis test.
It has several parts.
– Original closing: Wednesday, March 9 at 6:00 p.m.
– Closing is changed to Thursday, March 10 at 9:00 p.m.
•
This class: finish developing the tedious F test statistic
•
Next
class:
–
Detailed ANOVA calculations by hand
and
the hypothesis test itself
–
Automation using Minitab
•
Readings for March 21:
– Multiple Comparisons: pages 450452
– ANOVA assumptions: 452455
– TwoFactor ANOVA:pages 456462
Revisit page 444 of the text
•
This has all of the bits and pieces – every
tediouslooking equation – needed to construct
the F test statistic.
•
Note: the test statistic has a numerator and a
denominator
– just like z and t.
•
Last class we developed the numerator.
•
This class we develop the denominator and put
everything together.
•
Now, here’s page 444 again.
Table 11.2 on page 444 of the text
•
The first row contains the calculations that we did so far.
•
It shows how to get the variance for the treatment means:
–
= SSA/(c1)
–
also called the Mean Square due to Treatments:
MSA
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
•3/7/11
•2
Treatment Means vs. Grand Mean
•
We have three treatments and their
corresponding observations.
•
Compute the treatment means for all
treatments.
•
Also, compute the grand mean.
• A
1
:
13, 14, 9, 8
→
• A
2
:
20, 23, 17, 16
→
• A
3
:
8, 4, 9, 3
→
•
Grand Mean:
•
Given all of this information, let’s compute SSA
and MSA.
1
y
11
2
y
19
3
y
6
y
12
We concluded last class with several questions
about this “new” variance.
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '08
 KOUZEHKANANI
 Standard Deviation, Summation, i=1

Click to edit the document details