PSYC 711 - DATA ANALYTIC PROCEDURES II
Spring 1998

Marilyn E. Demorest
Office: MP 320
Email: demorest@umbc2.umbc.edu
Hours: Tues. & Thurs. 10:00-11:30
           Thurs. 2:30-4:00, & by appointment
Stephanie Eischen
Office: MP009E
Email: eischen@umbc2.umbc.edu
Hours: To be arranged & by appointment

Primary Text:

Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York: Harper Collins.

Supplementary Texts:

Grimm, L. G., & Yarnold, P. R. (1994). Reading and understanding multivariate statistics. Washington, DC: American Psychological Association.

Harris, R. J. (1985). A primer of multivariate statistics (2nd ed.). Orlando, FL: Academic Press.

Stevens, J. (1996). Applied multivariate statistics for the social sciences (3rd ed.). Mahwah, NJ: Erlbaum.

Course Objectives. The course is organized around two broad objectives. The first is to understand general principles and theory underlying multivariate analysis, including both the statistical concepts and the interrelationships among different procedures. The second objective is to be able to select, implement, and interpret the results of multivariate analyses.

Texts. The primary text by Tabachnick and Fidell is a well-organized and well written introduction to multivariate analysis, which covers the conceptual basis for the various procedures, assumptions underlying their use, and annotated examples of statistical programs and their output. The text by Grimm and Yarnold gives an excellent conceptual introduction to various methods of analysis with simple, clear examples. Although the chapters are not explicitly assigned, reading appropriate chapters of this book before and after the assigned readings may help you to integrate the material and to find the forest after seeing the trees! It is available from APA in paperback. The text by Harris has an excellent introductory chapter that both summarizes univariate statistical methods and provides an overview of their multivariate counterparts. It is assigned reading and will be distributed to you. The text by Stevens is recommended both for supplementary reading during the course and as a reference on multivariate analysis for the future. You are strongly encouraged to include the chapters from Stevens in your reading so that you can identify recurrent themes and benefit from another perspective on important issues. All reading assignments are shown in the Course Calendar.

Quizzes. Ten brief quizzes will be administered during the semester. The quizzes will be based on readings, lecture material, and handouts, and will utilize a variety of question formats such as multiple-choice, true-false, fill-in-the-blank, matching, and so on. They will emphasize theory and understanding of concepts. Quizzes will be administered using an elimination procedure that permits you to receive partial credit if you don't choose the correct answer, but can eliminate one or more of the incorrect answers.

There will be no make-ups for quizzes that are missed and no additional time will be given to students who arrive late for a quiz. However, only the best eight quiz scores will count toward the course grade; the lowest two quiz scores will be dropped. Thus, if you miss a quiz, for any reason, you can simply let that be one of the quizzes you drop.

Exams. There will also be four in-class examinations, which will be administered during the laboratory. The exams will not be cumulative, except to the extent that the material itself is. Each exam will comprise two parts, one consisting of multiple-choice questions and the other requiring several short-answer essays. The elimination procedure will be used with the multiple-choice questions. The essays will require you to define and discuss specific concepts or to contrast and compare related concepts. A list of key terms from which the essay items will be drawn will be distributed before each exam as a study guide.

Exercises. Five sets of exercises will be assigned to facilitate your mastery of the reading material and to give you experience conducting multivariate analyses and interpreting the results. Exercises will consist of study questions, computational problems, and computer assignments. Students are free to consult with one another and are encouraged to do so, but each student will receive a different set of data and must hand in his or her own complete set of exercises. Exercises are due at the beginning of class on the dates shown in the course calendar.

Laboratory. All quizzes will be administered at the beginning of the laboratory session. On some occasions, there will be a lecture during laboratory time. Other laboratory sessions will be devoted to a variety of activities including review of material covered in lectures, illustration of computer-based multivariate analyses, discussion of assigned reading from the texts, and so on. Laboratory sessions will be conducted by the teaching assistant, Stephanie Eischen.

Grading System. The course grade will be based on quizzes, exams, and exercises. Each of these components will be graded on a percentage basis, and an average will then be taken to obtain an overall percentage for the course. Together, quizzes and exams will count 75% toward the course grade; the exercises will count 25%. However, the relative weight of quizzes and exams will be determined for each student: whichever average is higher will receive double weight. For example, a student with a quiz average of 87%, an exam average of 78%, and an exercise average of 94% would receive a grade of [2(87%) + 78% + 94%]/4 = 86.5%.

The criteria for letter grades are as follows: At most, 90% will be required for a grade of A, 80% for a grade of B, 70% for a grade of C, and 60% for a grade of D. These cutoffs may be lowered at the end of the semester; they will not be raised.

Course Calendar
Date Lecture Topic Reading
Tuesday, 1/27 Course Overview
Introduction to Multivariate Analysis
Harris, Chapter 1, pp. 1-9
Thursday, 1/29 Multivariate Data and Descriptive Statistics Tabachnick & Fidell, Chapter 1
Friday, 1/30 DIAGNOSTIC QUIZ
Review of Univariate Statistics

Tabachnick & Fidell, Chapter 3

Harris, Chapter 1, pp. 9-28

Tuesday, 2/3 Linear Combinations of Variables  
Thursday, 2/5 Overview of Multivariate Techniques Tabachnick & Fidell, Chapter 2
Friday, 2/6 QUIZ #1 (thru 2/3)
Matrix Algebra: Basic Operations
Tabachnick & Fidell, Appendix A, pp. 821-824
Stevens, Chapter 2, pp. 41-45
Tuesday, 2/10 Overview of Multivariate Techniques Harris, Chapter 1, pp. 28-38
Thursday, 2/12 Data Screening Tabachnick & Fidell, Chapter 4, pp. 57-87
Friday, 2/13 QUIZ #2 (thru 2/10)
Matrix Algebra: Scalar Products & Matrix Multiplication
Tabachnick & Fidell, Appendix A, pp. 824-826
Stevens, Chapter 2, pp. 45-49
Tuesday, 2/17 EXERCISE #1 due
Data Screening
Tabachnick & Fidell, Chapter 4, pp. 88-125
Thursday, 2/19 Multiple Regression Analysis Tabachnick & Fidell, Chapter 5, pp. 127-165
Friday, 2/20 QUIZ #3 (thru 2/17)
Matrix Algebra: Determinants

Stevens, Chapter 2, pp. 49-56
Tuesday, 2/24 Multiple Regression Analysis  
Thursday, 2/26 Matrix Algebra: Matrix Inversion Tabachnick & Fidell, Appendix A, pp. 826-828
Stevens, Chapter 2, pp. 56-60
Friday, 2/27 EXAM #1 (thru 2/24)  
Tuesday, 3/3 EXERCISE #2 due
Matrix Algebra: Characteristic Equations, Eigenvalues, & Eigenvectors
Tabachnick & Fidell, Appendix A, pp. 828-831
Stevens, Chapter 2, pp. 60-62
Thursday, 3/5 Multivariate Multiple Regression Analysis Stevens, Chapter 3, Sections 3.14, pp. 107-115; Sections 3.19-3.20
Friday, 3/6 QUIZ #4 (thru (3/3)
Illustration: Mutivariate Multiple Regession Analysis
 
Tuesday, 3/10 Analysis of Covariance (ANCOVA) Tabachnick & Fidell, Chapter 8, pp. 321-352
Thursday, 3/12 Canonical Correlation Analysis Tabachnick & Fidell, Chapter 6, pp. 195-222
Friday, 3/13 QUIZ #5 (thru 3/10)
Illustration: Canonical Correlation Analysis
 
Tuesday, 3/17 Canonical Correlation Analysis  
Thursday, 3/19 Oneway Multivariate Analysis of Variance and Covariance (MANOVA & MANCOVA) Tabachnick & Fidell, Chapter 9, pp. 375-407
Friday, 3/20 QUIZ #6 (thru 3/18)
Illustration: MANOVA & MANCOVA
 
Tuesday, 3/24
Thursday, 3/26
Friday, 3/27
SPRING BREAK  
Tuesday, 3/31

EXERCISE #3 due

Factorial MANOVA & MANCOVA

 
Thursday, 4/2 Repeated Measures ANOVA
Profile Analysis
Tabachnick & Fidell, Chapter 10, pp. 441-459
Stevens, Chapter 13, pp. 450-461; 466-468; 497-500
Friday, 4/3 QUIZ #7 (thru 3/31)
Illustration: Repeated Measures ANOVA
 
Tuesday, 4/7 Repeated Measures MANOVA and MANCOVA Tabachnick & Fidell, Chapter 10, pp. 459-483
Stevens, Chapter 13, pp. 500-512
Thursday, 4/9 Discriminant Function Analysis Tabachnick & Fidell, Chapter 11, pp. 507-546
Friday, 4/10 EXAM #2 (thru 4/7)  
Tuesday, 4/14 Discriminant Function Analysis  
Thursday, 4/16 Logistic Regression Tabachnick & Fidell, Chapter 12, pp. 575-609
Friday, 4/17 QUIZ #8 (thru 4/14)
Illustration: Discriminant Function Analysis
 
Tuesday, 4/21 EXERCISE #4 due
Logistic Regression
 
Thursday, 4/23 Loglinear Analysis Tabachnick & Fidell, Chapter 7, pp. 239-285
Friday, 4/24 QUIZ #9 (thru 4/21)
Illustration: Logistic Regression
 
Tuesday, 4/28 Loglinear Analysis Stevens, Chapter 14, Sections 14.1-14.11, 14.13, 14.14
Thursday, 4/30 PCA & FA Tabachnick & Fidell, Chapter 13, pp. 635-679
Friday, 5/1 EXAM #3 (thru 4/28)  
Tuesday, 5/5 PCA & FA Stevens, Chapter 11, pp. 362-388
Thursday, 5/7 Structural Equation Modeling Tabachnick & Fidell, Chapter 14, pp. 709-767
Friday, 5/8 QUIZ #10 (thru 5/5)
Illustration: Principal Components Analysis
 
Tuesday, 5/12 EXERCISE #5 due
Structural Equation Modeling
Stevens, Chapter 11, pp. 389-410, 415-418
Friday, 5/15 EXAM 4 (thru 5/12), 8:00-10:00 AM