Carey: Psyc 7291: Multivariate Statistics

Multivariate Statistics Home Page

Professor:

Greg Carey
Mail: Dept. of Psychology, University of Colorado, Boulder CO 80309-0345 USA
Office: D261 Muenzinger
Phone: 303-492-1658
Fax:303-492-2967
Email: gregory.carey@colorado.edu
Home Page: http://ibgwww.colorado.edu/~carey

Overview:

This course will emphasize computer approaches to multivariate statistical analysis. We will discuss the three major goals of multivariate analysis and their associated techniques: (1) data reduction (principal components, factor analysis, and cluster analysis); (2) discrimination and classification (cluster analysis, discriminant analysis); and (3) hypothesis testing (multivariate regression, multivariate analysis of variance, logistic regression). Special topics of interest to a sufficient number of students may also be taught.

 

Requirements:

You will be required to turn in reports on problem sets over the semester. In the problem sets, you will be given data and must perform one or more analyses on them. You should write the reports as if they were the Methods and the Results section of a journal article, including relevant tables. Avoid jargon from the statistical package that you used (e.g., "used the MANOVA option of the SAS GLM procedure") and in general write about 2 double-spaced pages, excluding tables and figures.

In lieu of a final exam, you will be required to do a major analysis of a data set of your choice and write the appropriate report. There will be no tests or exams. All reports and papers should be written in the style of a major journal in your field.

Finally, you must memorize all the material at the bottom of this web page.

 

Statistical Packages:

Course examples and illustrations will primarily use SAS, the Statistical Analysis System. You are free to use any statistical package that you choose to perform the analyses. You may also use any computer that you choose (as long as you are responsible for the computing expenses). The documentation for statistical packages is voluminous and can be very expensive. For that reason, we will rely on class notes and the online documentation to learn about the major features of the statistical packages.

CU has site licenses for three statistical packages--SAS, SPSS, and STATA. Go to http://www.colorado.edu/its/tpsitelic/ for more information.

 

Textbook:

Because students take this course for very different reasons, there is no mandatory textbook. Instead, you should purchase a text that suits your needs (e.g., practical application versus mathematical statistics). Recommended texts are:

Johnson, Dallas E. (1998). Applied multivariate methods for data analysis. Pacific Grove, CA: Duxbury Press. Good balance between theory and practice.

Tabachnick, B.G. & Fidell,L.S. (2000). Using Multivariate Statistics, 4th Ed. New York: Allyn & Bacon. A traditional and popular text that focuses on practical applications.

There is a reading list that contains more texts, including specialized ones. If you are interested in the mathematics behind multivariate statistics, I suggest that you examine this list, peruse the books in the library, and then look for a used copy, Unlike computer applications, the mathematics remains constant, so old texts can still be useful.

 

Required Memorization: The major reason to study statistics

A mathematician, an engineer, and a statistician were hunting big game on the plains of Africa. They sighted a very large rhinoceros who lifted his head, caught the hunters' scent, and immediately charged the trio.

The engineer--ever the practical one--was the first to lift his rifle and shoot. The bullet grazed the left ear of the rhinoceros.

Never to be outdone by a mere engineer, the mathematician immediately raised his gun and fired. The bullet grazed the right ear of the rhinoceros.

The statistician threw his gun to the ground, raised his arms in the air, and shouted triumphantly, "Got him!"

 

Moral of the story:

Naturally, the rhinoceros, angered at having his ears nicked, trampled the trio to death. Hence, the statistician saved a valuable member of an endangered species. If you too want to save our planet, then become a statistician. (Remember, September 17 is kiss a rhinoceros day.)

 

Other required memorization:

"There are three kinds of lies--lies, damn lies, and statistics." Variously attributed to Benjamin Disraeli and Samuel Clemens.

"There are liars, outliers, and out-and-out liars." Robert Dawson

"A single death is a tragedy. A million deaths is a statistic." Joseph Stalin.

"Think about how stupid the average person is. Now realize that half of them are dumber than that." George Carlin

"Most people, when asked if they would ever date a statistician, respond, 'Probably.'" ANON

"A statistician calculated that the probability of one person carrying a bomb onto an airplane was 1 in 350,000. The probability of two people carrying a bomb onto the same place was 1.6E-11. The next time the statistician had to fly, he immediately made a bomb." ANON

"Statistics means never having to say you're certain." ANON

"If you want three different opinions, ask two statisticians." ANON

"My life is an experiment I never had the chance to properly design." Diane Ballard

"Oneway ANOVAs are always jealous of twoway ANOVAs because they can have interactions with their variables." ANON

"Arguing with a statistician is like wrestling with a pig. After a few hours, you realize that the pig likes it." Steve Carlson

"A lottery is a tax on people who don't understand statistics." Alvaro Montenegro Garcia

"The most famous statistician is George Washington. He claims that he never told a lie and got away with it." ANON

"The best reason for teaching statistics to medical students is that it keeps the dumbies out of the profession." FGC

"Smoking is the major cause of statistics." ANON

"Statistics say that if your parents don't have any children, you won't either." Hugh W. Graham

" A statistician is a professional who diligently collects facts and data and then carefully draws confusions about them. " ANON

"42.7 percent of all statistics are made up on the spot." Steven Wright (although the actual quoted percent has a large sd)

 

Addendum (the class motto):

Statistica, praeter facetiam.

 

Addendum ad addendum:

Quantum materiae materietur marmota monax si marmota monax materiam possit materiari?