STAT 5303 --- Applied Statistics II --- Spring 2008
Basic Information
Course instructor:
Dr. Alex
Trindade, 211 Mathematics & Statistics Building.
E-mail: alex.trindade "at" ttu.edu; Phone: 742-2580 x 233.
Course Meets: 10:00-10:50am MWF, in Math 016.
Office Hours: 11:00-12:00 MWF.
Text Book
Required: An Introduction to Statistical Methods and Data Analysis, 5th edition (2001), R. Lyman Ott and Michael Longnecker, Duxbury-Thomson-Brooks/Cole, Belmont, CA [ISBN: 0-534-25122-6].
Also strongly recommended is the accompanying Student Solutions Manual containing solutions to selected exercises [ISBN: 0-534-37123-X].
Course Objectives and Syllabus
STAT 5302-5303 is an introductory sequence of courses for graduate students in the life, biological, agricultural, or social sciences, who have little or no background in statistics, yet who plan to use statistical techniques in their
research. The emphasis is on the analysis of data. Familiarity with the
material covered in STAT 5302 is assumed in STAT 5303. Primary objectives are to:
- Provide a foundation in basic statistical concepts, numerical and graphical data summaries.
- Introduce statistical inference based on the t, F and Chi Square tests.
- Introduce simple linear and multiple regression analysis.
- Introduce statistical aspects of experimental design and the
associated analysis of variance (ANOVA) and analysis of covariance (ANCOVA).
- Introduce statistical aspects of categorical data analysis.
- Introduce more advanced concepts and methods in: design of experiments;
mixed models; factorial experiments; repeated measures designs.
Secondary objectives: provide practice in using a statistical
computing package to perform the basic analyses covered; introduce a
few of the many "nonparametric" alternatives to the standard
"parametric" methods covered; discuss research data management to
facilitate data analysis.
Note that this sequence is not for mathematics, statistics, engineering, or physical science majors; these students should take STAT 5384-5385.
Lecture Notes and Syllabus
Notes are available here in ppt format for some Units. Topics without
pre-prepared slides will likely be covered on the board. You'll find it helpful to augment the lecture notes by reading the appropriate sections in the book.
Grading and Other Policies
- Your course grade will be based on a mix of written Assignments (70%), a
Midterm Test (10%), and a comprehensive Final Exam (20%). Although discussion with others in
broad conceptual terms is encouraged for the Assignments, your
submitted work must be your own.
- Course averages of at least 90% and 80% will guarantee the
passing grades of A and B, respectively. Course averages below 80% are
candidates for the failing grades of C, D, and F. If your course
average starts to fall in an undesirable (or catastrophic) category,
it is your responsibility to counsel with me about what your options are, and
what you might realistically be able to get. Once final grades have
been awarded there will be NO APPEALS!
Assignments
There will be 7 Assignments, each involving an extensive
statistical analysis of a dataset, and subsequent writing up of the results. Your
analysis should be typeset in the form of a report. More detailed instructions
will be given with each Assignment. The numbering scheme parallels that of the
Units in the Lecture List. Assignments and their due dates are as follows:
Submit a stapled hard copy by noon, either to me in class or to the secretary in the main office.
Tests
There will be a midterm test and a comprehensive final exam.
- Midterm Test (Fri Mar 14): Over Units 5-8.
- Final Exam (Tue May 6, 1:30-4:00pm).
Statistical Computing, Software, and Other Course Resources
Some of the more commonly used datasets in class examples, and tests, are available here.
This course will require extensive statistical computing work, most of
which you will have to pick up on your own. Since the software package you choose to do this with is up to you, the following statistical software overview may be useful. The course text shows output from MINITAB, SAS, JMP, STATA, and SPSS. The course notes have some example code and output from MINITAB, R, SAS, and SPSS.
You should probably select a software package that you already have access to (through your home department's computing system for example) and/or will most likely end up doing research with. Failing this, if you have no prior programming experience I suggest you use MINITAB. If you are
more ambitious, computer-savvy, plan on doing extensive exploratory data analysis in
the future and preparing statistical plots for publication, I would
encourage you to become acquainted with R; but be aware that it is a steep learning curve! Follow this
link to access some of my resources on statistical computing, particularly R.
top of page