Structural Equation Modeling Seminar
BUS 620
Fall 2004
CLASS PERIOD: Thursday 2:30 - 5:00 pm, Holman 132
INSTRUCTOR: Dr. Douglas Vorhies, Ph.D.
North Holman 327
915-1351
OFFICE HOURS: MW 3:00 – 4:00 pm. (An appointment is recommended). Other times can be obtained by appointment.
WEB PAGES: Vorhies: http://faculty.bus.olemiss.edu/dvorhies
SEMNET: http://www.gsu.edu/~mkteer/semnet.html
TEXTS: Bollen, Kenneth (1989). Structural equations with latent variables. New York: Wiley.
Byrne, Barbara, (2001) Structural Equation Modeling With Amos: Basic Concepts, Applications, and Programming, LEA press.
COURSE DESCRIPTION:
This seminar provides instruction in the use of Structural Equation Modeling (SEM) and the statistical software to estimate such models. SEM is a common tool in business and social science research. In general SEM combines factor analysis and multiple regression analysis to produce models capable of simultaneously assessing the measurement properties of latent variables and their indicators and the relationships among those latent variables. Prerequisites: Students should have experience with matrix algebra, factor analysis and regression analysis.
COURSE OBJECTIVES:
The primary goal of this course is to provide students with an understanding of SEM theory, methodology and application. To do this, the course is structured to help students achieve the following objectives:
1. To develop a broad understanding of the theoretical foundations of SEM and the tools and techniques used in SEM research.
2. To develop student’s ability to apply SEM theory and method via commonly used SEM software examples and homework.
3. To provide each student with the skills to be able to use commonly available SEM software to solve problems in their discipline (e.g., AMOS, SAS – Calis, EQS, and Lisrel).
4. To develop an understanding of theory and application via in class discussion of assigned readings.
5. To integrate theory, method and applications via the preparation of an applied research article of sufficient theoretical and methodological interest to be submitted to an appropriate journal in the student’s area of expertise. This will require obtaining a dataset, developing an appropriate theoretical model with appropriate literature support; performing the appropriate SEM techniques to test the conceptual model and writing up the results of the modeling.
GRADING:
Grades for the course will be determined based on your performance in the following areas:
Final Exam / 15 %Homework / 10 %
Final Paper / 50 %
Class Participation. / 25 %
Final letter grades will be given according to the following points schedule:
A = / 90%B = / 80%
C = / 70%
D = / 60%
F = / Below 60%
EXPLANATION OF ASSIGNMENTS AND CLASS POLICIES:
1. Final Exam: The final examination will contain questions similar to those included on doctoral comprehensive exams. The questions will require that you integrate and synthesize the material we have discussed. The exam will contain 4-6 questions and you will have some discretion concerning which questions you answer.
2. Homework: Throughout the course of the semester I will assign homework problems for you to prepare using either AMOS or SAS – Proc Calis. The datasets for these homework assignments will be made available through my website for the class. Details on how to do each homework assignment will be given in class.
3. Class Participation: Class participation is a very important part of the SEM seminar and all seminar students are expected to play an active role in regularly discussing the assigned readings. The evaluation of class participation will be based on your level of preparation and the strength of your contributions to our discussion. Please feel free to discuss your class participation with me at any time during the semester.
Weekly Readings Summaries: Each week (after session one), the seminar team will divide up the next week’s readings for the purpose of determining who will lead the discussion of each article or chapter from the text. Each student will prepare written summaries of each of the readings and chapters for each session. These summaries must not exceed two typed pages (single spaced) and may be collected to check quality and the understanding each student is getting from the articles. To maintain consistency, each summary should have the format outlined below.
4. Final Paper: 50% of your grade in this class is determined by the class paper.
5. Attendance: Attendance is required for all students. If you must miss class, you do so at your own risk. Missed in-class assignments, homework and participation cannot be made up.
6. Other Assignments: I also reserve the right to give you other assignments throughout the course of the semester. These may include case reports, homework, and any other assignments I feel are necessary. As assignments are added, the total number of points for the class will be increased.
7. School of Business Administration Statement of Academic Integrity: The School of Business Administration upholds honor and academic integrity in all of its teaching, research, and service activities.All business faculty, staff, and students are charged with the responsibility to behave with personal and professional integrity and to refrain from dishonorable conduct.
Article Summaries Format
Chapters in the Textbook:
Chapter citation: in bold, at the top of the page. For example, Bollen, (1989) Chapter 1, p. 1-9.
Summarized by: Your name here.
Purpose of the Article: This section should describe why the purpose for the chapter.
Major Issues discussed: This section should summarize the methodological issues discussed in the chapter and give insight regarding key points.
Critique: A brief critique of the chapter. What do you agree with and what do you disagree with? What methodological questions are left unanswered? What could be done in future research to fill these gaps? Last, but not least, what didn’t you understand? What could be made more clear to everyone?
Research Articles:
Complete citation: in bold, at the top of the page (following JM format).
Summarized by: Your name here.
Purpose of the Article: this section should describe why the article was written, how this reading fits in a broader stream of research, and what the authors’ purpose was in writing the piece. Identify and label the Research Questions (RQ).
Research Questions: Identify and list the research questions.
Theoretical foundation: This section should summarize the theoretical arguments made by the authors. This would include the theoretical concepts, paradigms and assumptions used by the researchers. Identify and summarize any hypotheses or research propositions proposed in the article. Identify and label these.
Hypotheses/Propositions: Summarize the hypotheses or propositions.
Methodology: this section should summarize the research design and methodology (if appropriate).
Results and conclusions: this section should summarize the results and major discussion points.
Critique: A brief critique about the article. What do you agree with and what do you disagree with? What theoretical questions are left unanswered? What could be done in future research to fill these gaps?
Please note: I reserve the right to collect your summaries periodically to check for quality and to see how well you are understanding the text and readings. Incomplete or poorly prepared summaries will result in a reduction in your participation score.
STRUCTURAL EQUATION MODELING SEMINAR (BUS 620) TENTATIVE COURSE OUTLINE
Date / Topic / Readings and Assignments8-26 / Introduction to the course
How to read and prepare to discuss a research article.
Why SEM?
9-2 / Introduction to SEM
Causal or just correlational? / 1. Bollen 1, 2, 3, 4 and Appendix A (if your Matrix Algebra is a bit rusty!).
9-9 / Review of Measurement Theory:
- Measurement Theory and Application
- Measurement Error
- Reflective vs Formative Indicators
- Parceling (summating scales). / 1. Bollen Chp 5 and 6
2. Churchill Gilbert A. (1979) “A Paradigm for Developing Better Measures of Marketing Constructs.” Journal of Marketing Research. A6 (February) 64-73.
3. Diamantopolous Adamantous and Heidi M. Winklhofer (2001), “Index Construction with Formative Indicators: An alternative to Scale Development. Journal of Marketing Research. 38 (May) 269-277.
4. Jarvis et al. (2003) “A Critical Review of Construct Indicators and Measurement Model Misspecification in Marketing and Consumer Research,” Journal of Consumer Research, 30 (September).
5. Little, Todd D, William A Cunningham, Golan Shahar, Keith F. Widaman. (2002) “To Parcel or Not to Parcel: Exploring the Question, Weighing the Merits.” Structural Equation Modeling, 2002, 9(2), 151.
9-16 / Applications of SEM in Business Research:
- Systems of Linear Equations
- / 1. Torkzadeh, Gholamreza, Xenophon Koufteros, and Kurt Pflughoeft (2003) “Confirmatory Analysis of Computer Self-Efficacy.” Structural Equation Modeling, 2003, 10(2), 263.
2. Hult, Tomas M., Robert F. Hurley, Larry C. Giunipero, Ernest L. Nichols, Jr. (2000) “Organizational Learning in Global Purchasing: A Model and Test of Internal Users and Corporate Buyers,” Decision Sciences. 21(2): 293.
3. Li, Tiger and Roger J. Calantone (1999) “The Impact of Market Knowledge Competence on New Product Advantage: Conceptualization and Empirical Examination,” Journal of Marketing. 62 (October), 13-29.
4. Maurer, Todd J. and Heather R. Pierce (1998) “A Comparison of Likert Scale and Traditional Measures of Self Efficacy,” Journal of Applied Psychology. 83(2), 324-329.
5. Homburg, Christian and Christian Pflesser (2000) “A Multiple Layer Model of Market-Oriented Organizational Culture: Measurement Issues and Performance Outcomes.” Journal of Marketing Research. 37 (November). 449-462.
6. HW1 and HW2 assigned – Due 9-23.
9-23 / Confirmatory Factor Models (CFA)
- Using AMOS and SAS Calis
- Basic CFA, Item Analysis / 1. Bollen Chp 7
2. Byrne Chp 1, 2, 3, 4
3. Get AMOS Class Demo Program 1 and run it on your PC (this is a two factor CFA).
4. Get HW 1 from website. Using the Strat1 dataset, set up a your first 3 factor model (CFA 1) using both AMOS graphics and AMOS basic. Run both and compare results. Bring output from both to class.
5. HW2 - Item analysis: Using the Strat1 dataset, determine where the problems with CFA 1 are and using the other items in the dataset, determine what improvements you can make by substituting items. Write up a short discussion of what you did and report results (Chi square, GFI, AGFI, CFI should be reported). We will see who has the best fitting model for class.
6. One page description of project.
9-30 / Confirmatory Factor Models (CFA)
- Discriminant and Convergent Validity
- Average Variance Extracted and
Composite Reliability
/ 1.
2. Anderson and Gerbing (1988)
10-7 / Confirmatory Factor Models (CFA)
- Applications / 1.
10-14 / Confirmatory Factor Models (CFA)
- Second order factor models / 1. Byrne Chp 5
2.
10-21 / Structural Models
- Path Analysis
- Full Information Models / 1. Bollen Chp 8
10-28 / Structural Models / 1.
11-4 / Structural Models
- Full Information Models
- Common Methods Variance Tests / 1. Netemeyer
2. Hewett and Beardenz
3. Piercy, Cravens, Lane and Vorhies ???
11-11 / Structural Models – Additional Issues
- Using Second order factors in SEMs
- Outliers and Influence Points
/ 1. Vorhies and Morgan (Forthcoming), Benchmarking paper Journal of Marketing.
11-18 / Structural Models – Multi-Group Models
- Measure Invariance / 1. Vandenberg and Vance (2000), “A Review and Synthesis of the Measurement Invariance Literature: Suggestions, Practices, and Recommendations for Organizational Research. Organizational Research Methods, Vol. 3 No. 1, January 2000 4-70.
2. Baumgartner and Steenkamp (1988) Journal of Consumer Research
3. Baumgartner and Steenkamp (2001) Journal of Consumer Research
4. Netemeyer et al (1991)
5. Morgan, Zou, Vorhies and Katseikas (2003)
Decision Science.
11-25 / Thanksgiving Holiday
12-2 / Structural Models
- Testing Mediation and Moderation
(Interaction) Models in SEM / 1. Bollen Chp 9
2. Kenny and Judd
3. Joerskog and Yang
4. Ping
5. Cadogan
6. Vorhies et. al.
12-9 / Finals / Final Exam
Papers Due