Statistics 655 Official Web Page
Analysis of Variance/Experimental Design -- Spring
2004
Section 001 Thursday 7:20-10:00 PM Innovation Hall--Room 137
Professor:
John J. Miller
Office:
Sci-Tech II--Room 147
Phone:
(703) 993 - 1690 (You can leave a message on my voice mail if I am not in the
office. I will get back to you as soon as possible.) You may also use email.
My email address is jmiller@gmu.edu.
Office Hours:
By mutually convenient appointment. Regular hours may be scheduled.
Links to Syllabi with Schedule of Classes
Text:
Hicks, Fundamental Concepts in the Design of Experiments, 5 Edition.
Potential Supplements:
(Note: You should not buy all (or perhaps any) of these books. None (except
The Little SAS Book) have been ordered at the bookstore. You may consider them
if you feel the material handed out and in the required text does not fulfill
your needs. Consult with Prof. Miller before purchasing supplemental books.)
Box, Hunter, and Hunter, Statistics for Experimenters.
Edwards, Experimental Design in Psychological Research.
DiIorio, SAS Applications Programming, A Gentle Introduction.
Delwiche and Slaughter, The Little SAS Book.
Course Objectives:
After you successfully complete this course, you should be able to analyze data
from any balanced experimental design, no matter how complicated. You should
also be able to design a good experiment to fit most situations. You will be
ready to learn how to analyze unbalanced data because you completely understand
the concepts required to design experiments and analyze experimental data. You
will understand completely randomized designs, including the use of multiple
comparisons and orthogonal contrasts. You will be able to extend that knowledge
to factorial experiments. You will thoroughly understand the concept of interaction.
You will understand the concept of blocking and be able to use it to create
more efficient designs. You will be able to analyze complex partial hierarchical
designs, including the use of multiple comparisons and orthogonal contrasts.
You will understand the methodology of 2 designs, including block confounded
and partial replicate designs.
The emphasis in this course will be on application. Theory will not be featured prominently. Most topics will be covered in two parts: “What to Do” and “Why It Works.” You will be expected to thoroughly understand the “What to Do” parts. Much of the “Why It Works” information will be available for you but not required for tests. The most important goal of this course is to enable you to design and analyze experiments in real life situations.
Tests and Grading:
There will be four tests, each with a take-home part and an in-class part. You
will have one week to complete the take-home parts of Tests 1 and 3 and two
weeks to complete the take-home parts of Tests 2 and 4. Test 4 is, in effect,
your final exam. There will be homeworks assigned every week (including those
weeks when tests are also extant). Some homeworks may be graded. There will
also be in class quizzes. Quizzes are not scheduled on the syllabus, but I will
give warning the week before each quiz. Your course grade will be determined
by your scores on the four tests plus, quizzes, plus whatever homeworks I grade.
If you have any questions about your grade or course standing as the semester
progresses, feel free to ask me about your situation. I will be most happy to
discuss it with you. (Final grade is based on the total points of all graded
work. I reserve the right to modify values for test totals. However, tentative
values are: T1IC, 35; T1TH, 75; T2IC, 35; T2TH, 100; T3IC, 15; T3TH, 30; T4IC,
60; T4TH, 120. Quizzes will be 10-15. I anticipate between 2 to 4 quizzes.)
Computing:
We will use the SAS® System to perform the analyses required for this course.
There are two options available to you to access SAS. The first option is to
use SAS on any of several PCs at GMU. Several GMU labs have SAS installed (several
don't). Of course, to use this option, you must be physically present on campus.
The second option is to obtain a copy of SAS to use on your owncomputer. You work through Patriot Computers (the GMU Computer Store in the Johnson Center). There is a modest fee for this privilege, but may be well worth the investment in terms of time savings for you.
Make sure you allow enough time to get your assignments done. If you wait until the last moment to begin your assignments, you may very likely not have enough time to finish. The learning experience in this course depends strongly on the fact that you do the homework from week to week.
Having an email Address:
You must use the GMU email system and actually read the email frequently. I
will be sending you at least one email each week containing the handouts for
the next week in .pdf format. You need to have a .pdf reader which you can download
for free using this link: Link
to Adobe Download Site
This page was last modified January 14, 2008 .