RR765 – Applied Multivariate Analysis
Semester: Fall – 2007Course Instructor: Dr. Jerry J. Vaske 491-2360 email: [email protected] Office: 244 Forestry Bldg.
Office Hours: Tuesday/Thursdays 1:00 – 2:00
Feel free to arrange other times by appointment Prerequisites: RR665 and ST312 (or equivalent)
Location: CNR lab
Time: Tuesday/Thursday 8:00 – 9:40 Course Description and Goals
RR665 (Survey Research and Analysis) and RR765 (Applied Multivariate Analysis) are designed to prepare students for their masters and / or dissertation research. Both courses are predicated on the assumption that the best way to learn research methodology and statistics is to become directly involved in the conduct of scientific inquiry. Consequently, a considerable amount of time is devoted to
analyzing data from actual research projects. The projects were selected from on-going research here at CSU, as well as other studies that have been conducted across the United States and internationally. For each study examined, we will: (1) discuss why the project was initiated; (2) critically evaluate the overall theoretical / research design, sampling procedures, and survey instrument; (3) analyze the data using SPSS; and (4) interpret the findings from both a managerial and a theoretical perspective. Neither RR665 nor RR765 is intended to be substitutes for formal statistics courses. Rather, emphasis is placed on understanding data manipulation techniques and what statistics are appropriate for addressing specific methodological problems.
In taking this applied approach, the goal is to achieve the following objectives:
1) To provide an overview of the major statistical techniques used by human dimensions in natural resources researchers. The specific data analysis techniques include:
• data manipulation • analysis of variance • handling missing data • cluster analysis
• effect size estimation • principal component analysis • correlation and regression • exploratory factor analysis • mediation and moderation models • confirmatory factor analysis
• discriminant analysis • structural equation modeling (SEM) • logistic regression
2) To provide guidelines for understanding what types of statistical techniques are appropriate for analyzing selected types of research questions.
3) To provide experience analyzing data and interpreting computer printouts from selected statistical software packages (e.g., SPSS and AMOS).
4) To provide assistance and experience in critically evaluating statistical analyses presented in published articles in the human dimensions of natural resources literature.
5) To provide experience in preparing written papers and presentations that address research questions using multivariate analysis techniques. This includes the use of computer printouts to construct data tables for use in journal articles.
Readings
Selected chapters from:
Vaske, J. J., & Graefe, A. R. (In preparation). Understanding Multivariate Statistics. Other required readings will be announced in class and provided for certain topics. Some of these readings will provide background information on specific techniques, while others will provide examples of their application in the human dimensions field. Optional readings will also be mentioned and/or provided in certain areas for those who want to learn more about a specific technique described in class.
All of the readings, my lecture notes (PowerPoint slides), assignments, exams, and databases that will be used in this course can be found in the following directories:
H:\RR765\1 – Readings\
H:\RR765\2 – In-class exercises\ H:\RR765\3 – Assignments\ H:\RR765\4 – Exams\ Course Policy
1. Attendance – This course is a graduate level seminar. Your participation in class is an essential element of this course. You are expected to do the required readings ahead of time and come to class prepared for discussion. If you need to miss a class, it is your responsibility to notify me and obtain information about announcements, assignments, and course content covered during a missed class.
2. Written Assignments and Exams – All assignments (excluding in-class exercises) for this class must be typed. Pay close attention to spelling and grammar, as these will count toward your grade on written assignments. APA (American Psychological Association, 5th Edition) format must be used for all written work in this class (e.g., in referencing, creation of tables, and formatting headers for paper sections).
All assignments are due no later than the beginning of class on the designated date. Late assignments will not be accepted.
3. NRRT Ph.D. students who receive a letter grade of “A” in RR765 do not have to take the Department’s Ph.D. quantitative screening exam.
NRRT Ph.D. students who receive less than a letter grade of “A” in RR765 have the option of: • re-taking RR765 or
• taking the Department’s Ph.D. quantitative screening exam. Course Requirements
1. In-Class Exercises – Throughout the semester, you will be required to participate in in-class exercises that provide experience in the application of statistical techniques learned in this course. These exercises will not count toward your final grade but are intended to supplement in-class lectures and assist with your understanding of course content.
2. Review of Research Articles – For many of the statistical techniques covered in class, you will be asked to locate a journal article in the human dimensions field that applies that technique. Articles you select cannot be among those I provide as required or optional readings. You will need to come to class prepared to share a summary and critique of the statistical procedures and research
question(s) addressed in the article. You will also be asked to turn in a copy of the article. Specific due dates are listed on the course schedule provided below. Note that certain statistical procedures we cover in class are excluded from this exercise.
3. Homework assignments – Assignments to be completed outside of class will occur throughout the semester on selected topics. The assignments are designed to help students gain experience in the use and interpretation of statistics and to facilitate learning about the techniques covered in class. Each assignment involves:
a) conducting and interpreting statistical analysis using SPSS. b) summarizing the findings in a tables (APA journal format).
The following book is an excellent resource for understanding how to present statistical information from a variety of techniques in tables:
Nicol, A. A. M., & Pexman, P. M. (1999). Presenting your findings: A practical guide for creating tables. Washington, DC: American Psychological Association.
c) summarizing the tables in text. For each assignment, you should turn in:
a) the computer printouts including the SPSS commands (e.g., any RECODE, IF, COMPUTE, FREQUENCIES, REGRESSION statements) used to generate the printout.
b) typed copies of the data tables and written summary.
The following table details the topics covered in each assignment, as well as their points and due dates.
Assignment Topic Points Date
1 Data Manipulation 50 Sept. 6
2 Regression Assignment 50 Sept. 25
3 Cluster Analysis 50 Nov. 1
4 Principal Components Analysis 50 Nov. 13
More details and guidelines for these assignments will be discussed in class before they are due. All assignments are due no later than the beginning of class on the designated date.
Assignments that are turned in late will not be accepted. 4. Exams – There will be 2 exams covering the course material.
Exam 1 contains two parts: (1) in-class and (2) take-home. The in-class portion of the exam will contain short answer questions and questions focused on interpretation of computer printout. The take-home portion will involve conducting analyses, interpreting and summarizing output in an in-depth write-up.
Exam 2 is a take-home exam. Similar to the take-home portion of Exam 1, Exam 2 will involve an in-depth write-up of the analyses that you will be conducting, interpreting, and summarizing. Dates for exams, including due dates for take-home portions, are provided below.
5. Final Project – The final project for this course involves the analysis and interpretation of data using SPSS. Although you may and should discuss your ideas about this project with your fellow students and myself, group projects are not allowed (note: A group is defined as 2 or more people). For the final project, you may use:
• One of my data sets. All available data sets can be found in the directory: H:\rr665\4 – Final Project Data\
The grading criteria that will be used to evaluate these projects are also included in this directory.
Overview of the Final Project: In the final project you will be asked to:
a) Select a data set for analysis.
b) Generate specific hypotheses regarding variable relationships in that data set. These hypotheses should be based on past research (a minimum of 10 citations). The directory:
“H:\RR Reference Articles” contains over 8,000 articles sorted by topic to get you started. c) Test those hypotheses using one or more of the analysis strategies that will be discussed in class
(e.g., Regression, Discriminant, Logistic, ANOVA, Cluster, Factor analysis, SEM). d) Prepare a 10-15 minute presentation summarizing your project.
e) Prepare a journal-style manuscript suitable for submission to a scientific journal. The format of this manuscript will vary depending on your project. However, the typical contents of a journal article include: Title page, Abstract, Introduction, Methods, Results, Discussion, and References sections. The American Psychological Association (APA) formatting style should be used throughout the paper.
f) You should turn in:
• the typed manuscript, including data tables / figures.
• the computer printout including the SPSS commands used to generate the printout. g) To facilitate your efforts, feel free to discuss your final project with me.
Course Grading
Grading Summary Percent of Grade Total Points Review of research articles (8 @ 10 points each) 12% 80Assignments (4 @ 50 points each) 29% 200
Exams (2 @ 100 points each) 29% 200
Final project presentation 15% 100
Final project journal article 15% 100
100% 680
Grades will be based on the total points accumulated from the assignments, exams and final project.
Grade Percent Points
A 100 – 90% 680 – 612
B 89 – 80% 611 – 544
C 79 – 70% 543 – 476
D 69 – 60% 475 – 408
Course Content
Date Day Topic Assignment / Exam Dates
Aug. 21 T Class Outline Review Aug. 23 R Review of Concepts:
Level of Measurement: Once over again Selection of Analysis Strategies
Aug. 28 T Review of Concepts:
Reliability & Validity, Effect Size Aug. 30 R Data Manipulation Strategies using SPSS
Sept. 4 T Computing Psychological Indices
Sept. 6 R Treatment of Missing Data and Outliers Data Manipulation Assignment
Sept. 11 T Exam 1 – In-class In-class Exam
Sept. 13 R Correlation Take-home Exam
Sept. 18 T Regression
Sept. 20 R Regression Article – Student Regression Articles Sept. 25 T Mediation & Moderation Regression Regression Assignment
Sept. 27 R Mediation & Moderation Article – Student Mediation – Moderation Articles Oct. 2 T Discriminant Analysis
Oct. 4 R Discriminant Analysis Article – Student Discriminant Articles Oct. 9 T Logistic Regression
Oct. 11 R Logistic Regression Article – Student Logistic Articles
Oct. 16 T ANOVA Take-home Exam
Oct. 18 R ANOVA Article – Student ANOVA Articles
Oct. 23 T Cluster Analysis – Introduction Oct. 25 R Ipsative Cluster Analysis
Oct. 30 T Cluster Analysis Article – Student Cluster Analysis Articles Nov. 1 R Principle Components Analysis (PCA) Cluster Analysis Assignment Nov. 6 T Exploratory Factor Analysis (EFA)
Nov. 8 R PCA / EFA Article – Student PCA / EFA Articles
Nov. 13 T Confirmatory Factor Analysis (CFA) PCA Assignment Nov. 15 R Structural Equation Modeling (SEM)
Nov. 20 – 22 Thanksgiving
Nov. 27 T CFA / SEM Article – Student CFA / SEM articles
Nov. 29 R Student Presentations Student Presentations
Dec. 4 T Student Presentations Student Presentations
Dec. 6 R Student Presentations Student Presentations
Additional Background Material 1) Supplemental Readings
The course readings contain information I believe useful for understanding the statistical topic. Additional statistical and methodological readings can be found in the following directories:
H:\RR Reference Articles\29 – Methodology H:\RR Reference Articles|30 – Monitoring Use H:\RR Reference Articles\31 – Measurement H:\RR Reference Articles\32 – Survey Research H:\RR Reference Articles\33 – Statistics – General H:\RR Reference Articles\34 – Statistics – Techniques H:\RR Reference Articles\35 – Significance – Statistical H:\RR Reference Articles\36 – Significance – Practical H:\RR Reference Articles\37 – Significance – Editorial Policy
For purposes of this course, if I did not discuss a specific issue mentioned in one of the references, the issue will not be included in the assignments or exams.
2) Potential File Types Used in Course File Extension Type of File
doc MS Word ppt MS Powerpoint pdf Adobe Acrobat sav SPSS data sps SPSS syntax spo SPSS output amw AMOS spl LISREL syntax
out LISREL output
Fit LISREL goodness-of-fit statistics Pth LISREL path diagram
dsf msf
pjf
Miscellaneous files created by LISREL (not needed for interpreting the findings)