CJT 765: SPECIAL TOPICS IN QUANTITATIVE COMMUNICATION
RESEARCH METHODS: STRUCTURAL EQUATION MODELING
Spring, 2005
Professor:
Office:
Phone:
Office hours:
Classroom:
Class meeting hours:
Teaching Assistant:
TA’s office:
TA’s phone number:
TA’s office hours:
Rick Zimmerman
245 Grehan
257-4099
Thurs. 3:30-4:50 and by appointment
B35 W.T. Young Library/223 Grehan Building
Thurs. 1-3:30; labs Tues. 4-5 pm, Thurs. 5-6 pm B35 WTY
Olga Dekhtyar
308 Breckinridge Hall
257-8133
Tues, 1:30 – 3 p.m.
Course Philosophy
While for most students, some of the material will be review, the course will be taught at the
graduate student level. That is, students are expected to have a level of commitment to the course well
beyond that expected for undergraduates in their courses, as the material covered in this course may be of
use throughout the students' careers. It is expected that multiple readings of material will have been
undertaken before class, and that an all-out effort will be made to understand the material and to work on
assignments for the class.
The instructor will do his best to make sure that the information presented is understandable, but
expects that students will have first spent some time trying to assimilate the material on their own.
Quizzes in this class will be completed in-class, will be designed to be difficult, and will be graded on a
curve, so that students who have superior ability and/or have expended much effort will be able to
demonstrate these on the exam. Late assignments will not be accepted and make-up exams will not be
given, except for extenuating circumstances.
Each student must keep up with the material as we go along; statistical methods is typically not a
subject for which several weeks of material can be crammed into one's brain in several hours.
Laboratories to become familiar with computer programming, model testing, and completing exercises
will be required. Students are strongly urged to stop in during office hours with Dr. Zimmerman or Ms.
Dekhtyar if they have any questions.
Primary Course Objectives





to familiarize graduate students in the social and behavioral sciences with the language, logic, and
implementation of structural equation modeling;
to compare and contrast structural equation modeling with more commonly used statistical
strategies in the social and behavioral sciences such as multiple regression analysis and factor
analysis;
to teach the criteria associated with the decisions that must be made at each phase of a structural
equation modeling analysis;
to consider the philosophical and statistical criticisms of structural equation modeling as an
approach to research design and data analysis;
to provide firsthand experience reviewing research reports that feature structural equation
modeling and writing up and presenting orally the results of structural equation modeling
analyses.
Elements of the Course
Readings
There are two required books:
1. Kline, R.B. (2005). Principles of Structural Equation Modeling (2nd edition). New York:
Guilford.
2. Hoyle, R.H. (ed.) (1995). Structural Equation Modeling: Concepts, Issues, and Applications.
Thousand Oaks, CA: Sage.
Most weeks, additional readings are also required. More advanced topics are occasionally covered
in suggested (but not required) readings, marked by an asterisk (*). All additional readings (generally as
pdf files) will be posted on the course website by the first day of class.
Quizzes
Two quizzes will comprise the testing in the course. While we hope you will read and learn the
material just for learning sake, sometimes in the mix of other activities and coursework, it is easy to let
readings and mastery of the material go by the wayside. So, I think some grades related to mastery of the
material may help students keep on the top of the material, and have decided to include 2 quizzes as part
of your grade. Each will occupy about 30 minutes of class time on Feb. 17 and Mar. 24. Most will be
short answer or short essay but some writing of computer programming, simple calculations, path
diagrams, and /or interpreting output may also be included.
Homework Assignments
Three homework assignments will also be required. All will use the dataset we are using for the
course, a 3-wave, longitudinal sample of about 3000 rural high school students from the beginning of 9th
to the end of 10th grade. We will discuss the dataset and the codebook for the dataset in the first
laboratory section of the course, to be held on Jan. 20 and Jan. 25. Laboratory sessions will focus on
preparing students for these homework assignments, including practice questions.
Research Project
The major product of the course will be a written report of a structural equation modeling analysis
you conduct on data of your choosing. On Feb. 10 I will ask you to specify a dataset that you will
analyze and write up for the course. On Mar. 17 I will ask you to prepare a document in which you
specify the names and characteristics of the variables your analysis will include and the nature of the
model you plan to fit. All models should include both measurement and structural components for this
assignment. About two-third of the way through the course I will ask you to meet outside of class with
another member of the class to discuss your data and plan of analysis and to exchange feedback on your
projects. On Apr. 14 you will provide a copy of a draft to two members of the class, and you will receive
copies of two drafts from other class members. You will provide written reviews for the authors of the
two drafts you receive; you will receive two reviews of your draft. On Apr. 21 your reviews will be due.
The final draft of the research project is due by May 6th at 10 a.m. (the university-assigned final exam
date and time). A 10-minute oral presentation will also be given on either April 21 or April 28.
Attendance and Participation
Students are expected to attend all class sessions, as both hearing about statistics material and
reading it as important elements to learning it. Attendance is also required at laboratory sessions (1 per
week), as doing statistics is probably the most important learning component of all. I also expect
students’ participation in class; both the quality and quantity of student’s participation will be considered
in their evaluation. A
Published Article Presentation
Each student will give a 5-minute presentation in which he/she describes and evaluates a published
study in which the data are analyzed using structural equation modeling. Students can choose from a list
of recently published articles in top-tier journals in their field of study; references and abstracts for
psychology, communication, and business/economics/marketing will be available on the course website
by January 31, two months before the presentations begin. Presentations will take place on March 31st
and April 7th. Details about the selection of an article and the contents of the presentation will be
provided around January 31st as well.
Grading
Three computer assignments, an oral presentation about a published article using structural
equation modeling, a research project (both an oral presentation and a written paper), and a midterm exam
will comprise the grading in this course. The total grade will be distributed as follows:
Homework Assignments
Published Article Presentation
Research Project—written component
Research Project—oral presentation
Quizzes (2 @ 10%)
Attendance/participation
24% (3 @8% each)
10%
26%
10%
20%
10%
Everyone should receive an “A” or a “B” barring poor attendance or not doing the work, so that
students can spend more time and energy on learning the material rather than on their grade.
Course Website
The website for the course is at www.uky.edu/centers/hiv/cjt765/cjt765.html. The course
syllabus, assignments, dataset to be used throughout the course (in SPSS format), additional readings (in
PDF files), articles for the published article presentation, datasets, and a variety of other materials will be
available on the course website.
Acknowledgment
I would like to acknowledge the following faculty members, whose syllabi helped provide some
suggestions for assignments, readings, or course organization. I either spoke to these faculty members
and/or their syllabi were available through publicly accessible websites. Copies of my syllabus have been
shared with them.
Rick Hoyle, Duke University, Psychology/Sociology 779, Structural Equation Modeling, taught Fall,
2000 at UK. (I have especially drawn on this syllabus for readings and assignments.)
Robert Hauser, University of Wisconsin, Sociology 952, Mathematical and Statistical Applications in
Sociology, Topic: Path Analysis and Structural Equation Models, taught Spring, 2004.
Stephen West, Cathy Cottrell, & Oi-Man Kwok, Arizona State University, Psychology 533,
Structural Equation Modeling, taught Spring, 2004.
Gregory Elliott, Brown University, Sociology 226, Structural Equation Models in the Social
Sciences, taught Fall, 2004.
Course Outline
Jan. 13
Introduction to Structural Equation Modeling:
Ancestry, History, and Philosophy of Science
Kline: 1
Hoyle: 7
Berk (1988)
Hershberger (2003)
Freedman (1991a)
Blalock (1991)
*Wolfle (2003)
Jan. 20
Review of correlation and regression
Kline: 2
Cohen et al. (2003), Chapter 2
Cohen et al. (2003), Chapter 3
Jan. 27
Review of data preparation, screening,
measurement issues
Kline: 3
Cohen et al. (2003) pp.225-251
DeVellis (1991), pp. 1-41
Schaefer & Graham (2002)
Feb . 3
Overview of SEM notation, path diagrams,
programs; Homework 1 due
Kline: 4; Hoyle: 1, 2, 8
Byrne (2001), Chapter 1
Byrne (2001), Chapter 2
*Byrne (1994), Chapters 1 & 2
*Kelloway (1998),Ch. 4-7
Feb. 10
Path Analysis 1: Basic theorems, mediation, coefficients, Kline: 5 pp. 93-105
Choose dataset
Hoyle: 3
Kenny (1979), Chapters 3-4
Alwin (1988)
Baron & Kenny (1986)
MacKinnon et al. (2002)
*Shrout & Bolger (2002)
Feb. 17
Path Analysis 2: decomposing a correlation, direct and
indirect effects, identification; Quiz 1
Kline: 5 pp. 105-122
Alwin & Hauser (1975)
Holbert & Stephenson (2001)
Pedhazur (1982), pp. 614-628
Fox (1980)
Feb. 24
Path Analysis 3: fitting a model, fit indices, comparing
models, statistical power; Homework 2 due
Kline: 6; Hoyle 3, 5
Hayduk et al., (2003)
Bollen & Long (1993)
Tanaka (1993)
*Jöreskog (1993)
*Meehl & Waller (2002)
*Reichardt (2002)
*Dormann, 2001
*Muthén & Muthén (2002)
Course Outline (cont.)
Mar. 3
Measurement Models and Confirmatory
Factory Analysis
Kline: 7; Hoyle: 10, 12
Lance et al., 2002
Noar, 2003
Mar. 17
Putting it all together: Structural and measurement
components in SEM; Turn in dataset description;
Homework 3 due
Kline: 8
Holbert & Stephenson (2002)
McDonald & Ho (2002)
Stephenson & Holbert (2003)
MacCallum & Austin (2000)
Hayduk & Glaser (2000)
Anderson & Gerbing (1988)
Bentler (2000)
Bollen (2000)
Mar. 24
Nonrecursive structural models; Quiz 2
Kline: 9
James & Singh, 1978
*Berry, 1984
Mar. 31
Advanced topics: Multi-group SEM, Latent Growth
Models; Published article presentations 1
Kline: 10, 11
Hoyle: 11, 13
Apr. 7
Pitfalls in using SEM;
Published article presentations 2
Kline: 12
Gonzalez & Griffin (2001)
Apr. 14
Critique of SEM and future directions
Rough draft of research report due
Kline: 13
Bentler & Dudgeon (1996)
McCallum et al. (1993)
Raykov & Marcoulides (2001)
Schumacker (2002)
deJong (1999)
Apr. 21
Research report presentations 1
Reviews of research reports due
Apr. 28
Research report presentations 2
May 6
Research reports due, 10 a.m.
Additional Readings
January 13
Berk, R.A. (1988). Causal inference for sociological data. In N.J. Smelser (Ed.), Handbook of
Sociology (pp. 155-172). Newbury Park: Sage.
Hershberger, S. L. (2003). The growth of structural equation modeling: 1994-2001. Structural
Equation Modeling, 10, 35-46.
Blalock, H. M., Jr. (1991). Are there really any constructive alternatives to casual modeling? P. V.
Marsden (Editor), Sociological Methodology, 21, 325-335. Cambridge, MA: Basil Blackwell.
Freedman, D. A. (1991). Statistical models and shoe leather. P. Marsden (Editor), Sociological
Methodology, 21, 291-313. Cambridge, MA: Basil Blackwell.
January 20
Cohen, J., Cohen, P., West, S. G., & Aiken, L.S. (2003). Applied multiple regression/correlation analysis
for the behavioral sciences (3rd Ed.). Hillsdale, NJ: Erlbaum.
January 27
Cohen, J., Cohen, P., West, S. G., & Aiken, L.S. (2003). Applied multiple regression/correlation analysis
for the behavioral sciences (3rd Ed.). Hillsdale, NJ: Erlbaum.
DeVellis, R. F. (1991). Scale Development: Theory and Applications. Newbury Park, CA: Sage.
Schafer, J.L., & Graham, J.W. (2002). Missing data: our view of the state of the art. Psychological
Methods, 7, 147-177.
February 3
Byrne, B. M. (1994). Structural Equation Modeling with EQS and EQS/Windows: Basic Concepts,
Applications, and Programming. Thousand Oaks, CA: Sage.
Byrne, B. M. (2001). Structural Equation Modeling with AMOS: Basic Concepts, Applications, and
Programming. Mahwah, NJ: Erlbaum.
Kelloway, E. K. (1998). Using LISREL for Structural Equation Modeling: A Researcher’s Guide.
Thousand Oaks, CA: Sage.
February 10
Kenny, D. A. (1979). Correlation and Causality. New York, NY: John Wiley.
Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 173-182.
MacKinnon, D. P., Lockwook, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A comparison
of methods to test the significance of the mediation and intervening variable effects.
Psychological Methods, 7, 83-104.
Alwin, 1988
*Shrout, P.E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New
procedures and recommendations. Psychological Methods, 7, 422-445.
Additional Readings (cont.)
February 17
Alwin, D.F., and Hauser, R.M. (1975). The decomposition of effects in path analysis. American
Sociological Review, 40, 37-47.
Pedhazur, E.J. (1982). Multiple Regression in Behavioral Research, Chapter 15, pp. 577-588. New
York, NY: Holt.
Fox, J. (1980). Effect analysis in structural equation models. Sociological Methods and Research, 9,
3-26.
Holbert, R. L., & Stephenson, M. T. (2001). The importance of indirect effects in media effects research:
Testing for mediation in structural equation modeling. Journal of Broadcasting and Electronic
Media, 47, 556-572.
February 24
Hayduk, L., Cummings, G., Stratkotter, R., Nimmo, M., Grygoryev, K., Dosman, D., Gillespie, M.,
Pazderka-Robinson, H., & Boadu, K. (2003). Pearl’s D-separation: One more step into causal
thinking. Structural Equation Modeling, 10, 289-311.
Bollen, K.A., & Long, J.S. (1993). Introduction. In K.A. Bollen & J.S. Long (Eds.), Testing Structural
Equation Models (pp.1-9). Newbury Park, CA: Sage.
Tanaka, J.S. (1993). Multifaceted conceptions of fit in structural equation models. In K.A. Bollen & J.S.
Long (Eds.), Testing Structural Equation Models (pp. 10-39). Newbury Park, CA: Sage
*Jöreskog (1993). A review of tests within LISREL. In K. A. Bollen & J.S. Long (Eds.), Testing
Structural Equation Models (pp. 130-145). Newbury Park, CA: Sage.
*Meehl, P. E., & Waller, N. G. (2002). The path analysis controversy: A new statistical approach to
strong appraisal of verisimilitude. Psychological Methods, 7, 283-300.
*Reichardt, C. S. (2002). The priority of just-identified, recursive models. Psychological Methods,
7, 307-315.
*Dormann, C. (2001). Modeling unmeasured third variables in longitudinal studies. Structural
Equation Modeling, 8, 575-598.
*Muthén, L. K., & Muthén, B. O. (2002). How to use a Monte Carlo study to decide on sample size and
determine power. Structural Equation Modeling, 9, 599-620.
March 3
Lance, C. E., & Noble, C. L. (2002). A critique of the correlated trait-correlated method and correlated
Uniqueness models for multitrait-multimethod data. Psychological Methods, 7, 228-244.
Noar, S. M. (2003). The role of structural equation modeling in scale development. Structural Equation
Modeling, 10, 622-647.
Additional Readings (cont.)
March 17
Holbert, R. L., & Stephenson, M. T. (2002). Structural equation modeling in the communication
sciences, 1995-2000. Human Communication Research, 28, 351-551.
McDonald, R.P., & Ho, M-H. R. (2002). Principles and practice in reporting structural equation analysis.
Psychological Methods, 7, 64-82.
Stephenson, M. T., & Holbert, R. L. (2003). A Monte Carlo simulation of observable latent variable
structural equation modeling techniques. Communication Research, 30, 332-354.
MacCallum, R.C., & Austin, J.T. (2000). Applications of structural equation modeling in psychological
research. Annual Review of Psychology, 51, 201-22.
Anderson, J. & Gerbing, D. (1988). Structural equation modeling in practice: A review and
recommended two-step procedure. Psychological Bulletin, 103, 411-423.
Hayduk, L. A., & Glaser, D. N. (2000). Jiving the four-step, waltzing around factor analysis, and other
serious fun. Structural Equation Modeling, 7, 1-35.
Bentler, P. M. (2000). Rites, wrongs, and gold in model testing. Structural Equation Modeling, 7, 82-91.
Bollen, K. A. (2000). Modeling strategies: In search of the holy grail. Structural Equation Modeling, 7,
74-81.
March 24
James, L.R., and B.K. Singh. (1978). An introduction to the logic, assumption, and basic analytical
procedures of two-stage least squares. Psychological Bulletin, 85, 1104-1122.
Berry, W. D. (1984). Nonrecursive Causal Models. Newbury Park, CA: Sage.
April 7
Gonzalez, R., & Griffin, D. (2001). Testing parameters in structural equation modeling: Every “one”
matters. Psychological Methods, 6, 258-269.
April 14
Bentler, P.M., & Dudgeon, P. (1996). Covariance structure analysis: Statistical practice, theory,
directions. Annual Review of Psychology, 47, 563-592.
MacCallum, R.C., Wegener, D.T., Uchino, B.N., & Fabrigar, L.R. (1993). The problem of equivalent
models in applications of covariance structure analysis. Psychological Bulletin, 114, 185-191.
Raykov, T., & Marcoulides, G. A. (2001) Can there be infinitely many models equivalent to a given
Covariance structure model? Strucutral Equation Modeling, 8, 142-149.
Schumacker, R. E. (2002). Latent variable interaction modeling. Structural Equation Modeling, 9, 4054.
deJong, P. F. (1999). Hierarchical regression analysis in structural equation modeling. Structural
Equation Modeling, 6, 187-211.