Links under topics below are .pdf files for the lecture materials.
Versions of the .pdf files including the answers will be available after each class under answers.
MP4 files were recorded from the class lecture (right-click, use save target/link as).
Week |
Date |
Course Materials |
Readings |
1 | 1/15 | Course Introduction Lecture 1: Review of MLM for Longitudinal Data Example 1: Unconditional Models of Change Electronic materials for Example 1 are available in my 2013 ICPSR materials as Example 3 |
Hoffman ch. 1-7 |
1/17 | HOMEWORK #0 DUE BY 11:59 PM | ||
2 | 1/22 | NO CLASS (or office hours this week) | |
3 | 1/28 | HOMEWORK #1 EFFORT DRAFT DUE BY 11:59 PM: Download HW1 | |
1/29 | In-Class Q & A (bring HW1 responses with you) Lecture 2: Time-Varying Predictors for Within-Person Fluctuation Example 2: Predicting Within-Person Fluctuation and Heterogeneity |
Hoffman ch. 8 Hedeker & Mermelstein (2012) |
|
1/31 | By now you will need to have successfully registered for Tusker and submitted SAS and Mplus scripts for analysis. Do so by getting your faculty advisor to sign up for a free HCC account, and then create a sub-account of your own. Follow the instructions given above to do so. Submit proof of your successful run (the SAS log and the Mplus output file with your Tusker directory path as a title) as an assignment in Blackboard (worth 4 HW points). | ||
4 | 2/5 | NO CLASS (or office hours) | Hoffman ch. 10 Sections 1-2 Hoffman (2012) |
2/7 | HOMEWORK #1 ACCURACY DRAFT DUE BY 11:59 PM | ||
5 | 2/12 | Lecture 2, continued Lecture 3: Accelerated Longitudinal Models Example 3: Models of Accelerated Time (completed version) (spreadsheet) Lecture 4: Review of Multilevel Models for Clustered Data Example 4: Two-Level Models for Persons in Groups (spreadsheet) |
Raudenbush & Bryk (2002) ch. 5 |
6 | 2/17 | HOMEWORK #2 NOW DUE BY 11:59 PM ON MONDAY 2/17 | |
2/19 | Lecture 2 and Example 2, continued |
||
7 | 2/26 | Lecture 5: Multivariate Longitudinal Models in SAS and Mplus Example 5a: Cross-Sectional Difference Score Models Example 5b: Longitudinal Difference Score Models Example 5c: Multivariate Within-Person Fluctuation |
|
8 | 3/5 | Example 5d: Multivariate Within-Family Change Example 5e: Mediation of Within-Person Fluctuation |
Hoffman ch. 9 |
9 | 3/12 | Lecture 5 and Examples, continued Example 5f: 7 Ways of Estimating Multivariate Change in SAS and Mplus Class ends at 10:50 |
Lüdtke et al. (2008) Preacher et al. (2010; 2011) |
10 | 3/17 | HOMEWORK #3 OUTLINE NOW DUE BY 11:59 PM ON MONDAY 3/17: Download HW3 | |
3/19 | NO CLASS (or office hours this week) | ||
11 | 3/24 | NO CLASS (or office hours this week) | |
12 | 4/2 | Lecture 5 and Example 5f, continued |
Bauer & Curran (2011) Curran et al. (2012) Bauer (2003) |
13 | 4/7 | HOMEWORK #3 NOW DUE BY 11:59 PM ON MONDAY 4/7 | |
4/9 | Lecture 6: Generalized Multilevel Models Example 6a: Clustered Models with Binary Outcomes (spreadsheet) |
Hoffman ch. 13 sec. 2 Bauer (2009) Hox (2010) ch. 6-8 Hur et al. (2003) |
|
14 | 4/16 | Example 6a, continued Example 6c: Longitudinal Models for Percent Correct (new SAS macros for GLIMMIX) Example 6d: Longitudinal Models with Other Non-Normal Outcomes (spreadsheet) |
|
15 | 4/21 | HOMEWORK #4 NOW DUE BY 11:59 PM ON MONDAY 4/21: Download HW4 | |
4/23 | Lecture 6, continued Example 6b: Longitudinal Models with Ordinal Outcomes (spreadsheet) Lecture 7: Three-Level (and Crossed) Random Effects Models Example 7a: Clustered Longitudinal Models |
Hoffman ch. 10 sec. 3-6 Hoffman ch. 11 sec. 1-2 |
|
16 | 4/30 | Lecture 7 and Example 7a, continued Example 7b: Cross-Classified Models for Clustered Data Example 7c: Changes in Nesting in Groups over Time Course Evaluations |
Hoffman ch. 11 sec. 3-6 |
17 | 5/9 | ALL HOMEWORK REVISIONS DUE BY FRIDAY 5/9 AT 11:59 PM | |
This course will feature the advanced uses of the multilevel models (aka mixed models, hierarchical linear models) for complex data analysis. After reviewing two-level longitudinal models, the course will cover multiple extensions, including models for accelerated time, cross-classification, multivariate outcomes, three-level outcomes, and generalized outcomes. Class time will be devoted primarily to lectures and examples. Lecture materials in .pdf format will be available electronically at the course website above prior to each class. Audio/Video recordings of the class lectures in .mp4 format will also be posted online, but are not intended to take the place of class attendance. In addition to the primary textbook (see below), supplemental book chapters and journal articles will be assigned for each topic; the list of readings below will be updated as needed. All readings will be available electronically within the online homework portal.
Because the course will have an applied focus using SAS and Mplus software, portions of the class and instructor office hours will be held in the 234 Burnett computer lab to provide opportunities for in-class practice and to work on homework assignments and receive immediate assistance. Participants will need access to SAS and Mplus software, both of which are available in rooms 234, 227, and 230 Burnett and online through the Tusker computing cluster (see the course webpage for access instructions). Finally, participants should be familiar with the general linear models and multivariate modeling using maximum likelihood (as covered in PSYC 941, 942, and 943 or the equivalent) prior to enrolling in this course. This course is intended to have PSYC 944 (or equivalent) as a pre-requisite as well.
As a reminder, the University has a policy on academic honesty (see the Graduate Studies Bulletin). All course assignments should be done individually.
Students with disabilities are encouraged to contact the instructor for a confidential discussion of their individual needs for academic accommodation. It is the policy of UNL to provide flexible and individualized accommodation to students with documented disabilities that may affect their ability to fully participate in course activities or to meet course requirements. To receive accommodation services, students must be registered with the Services for Students with Disabilities (SSD) office, 132 Canfield Administration, 472-3787 voice or TTY.
Course performance will be evaluated using 4 homework assignments designed to give participants hands-on practice applying the techniques discussed in class and will be due as listed on the online syllabus. Assignments will include data analysis, results interpretations, and other questions. Successfully running SAS software through the HCC will also be required (worth 4 homework points). In addition, “Homework 0” will be worth 3 bonus points to familiarize participants with the online homework system.
Some of the homework assignments will feature common data and have two parts. The computational portion of the homework assignments will be administered and submitted using the online homework portal above. The written results sections that accompany them will be submitted through UNL Blackboard and returned using the online homework portal. Other homework assignments will be based on independent data, and will not have an onlinecomputational portion. Any written assignment must be at least 3/4 complete in order to be accepted and may be revised ONCE to earn the maximum points (due by the end of semester, submitted via Blackboard). Assignments should be submitted as a Microsoft Word document using this naming convention: 945HW#_Firstname_Lastname (adding an “r” to the end of the # for a revision). Please use the track changes function in Microsoft Word and leave in all previous instructor comments when revising assignments.
In order to be able to provide the entire class with prompt feedback, late homework assignments will incur a 3-point penalty. However, extensions will be granted as needed for extenuating circumstances (e.g., conferences, family obligations) if requested at least three weeks in advance of the due date. Finally, a grade of “incomplete” will only be given in the event of extremely dire circumstances.
=97 = A+ 93-96 = A 90–92 = A- 87-89 = B+ 83-86 = B 80-82 = B- < 80 = C or no pass
Hoffman, L. (forthcoming). Longitudinal analysis: Modeling within-person fluctuation and change. NY, NY: Routledge Academic.
Bauer, D. J. (2003). Estimating multilevel linear models as structural equation models. Journal of Educational and Behavioral Statistics, 28 (2), 135-167.
Bauer, D. J. (2009). A note on comparing the estimates of models for cluster-correlated or longitudinal data with binary or ordinal outcomes. Psychometrika, 74 , 97-105.
Curran, P. J., & Bauer, D. J. (2011). The disaggregation of within-person and between-person effects in longidutinal models of change. Annual Review of Psychology, 62, 583-619.
Curran, P. J., Lee, T., Howard, A. L., Lane, S., & MacCallum, R. C. (2012). Disaggregating within-person and between-person effects in multilevel and structural equation growth models. In G. Hancock & J. Harring (Eds.), Advances in Longitudinal Methods in the Social and Behavioral Sciences (pp. 217-253). Charlotte, NC: Information Age Publishing
Hedeker, D., & Mermelstein, R. J. (2012). Mood changes associated with smoking in adolescents: An application of a mixed-effects location scale model for longitudinal ecological momentary assessment (EMA) data. In G. Hancock & J. Harring (Eds.), Advances in Longitudinal Methods in the Social and Behavioral Sciences (pp. 59-79). Charlotte, NC: Information Age Publishing.
Hoffman, L. (2012). Considering alternative metrics of time: Does anybody really know what “time” is? In J. Harring & G. Hancock (Eds.), Advances in Longitudinal Methods in the Social and Behavioral Sciences (pp. 255-287). Charlotte, NC: Information Age Publishing.
Hox, J. (2010). Multilevel analysis: Techniques and applications (2nd ed). NY, NY: Routledge Academic.
Hur, K., Hedeker, D., Henderson, W., Khuri, S., & Daley, S. (2003). Modeling clustered count data with excess zeros in health care outcomes research. Health Services and Outcomes Research Methodology, 3 , 5-20.
Lüdtke, O., Marsh, H. W., Robitzsch, A., Trautwein, U., Asparouhov, T., & Muthén, B. (2008). The multilevel latent covariate model: A new, more reliable approach to group-level effects in contextual studies. Psychological Methods, 13 (3), 203-229.
Preacher, K. J., Zhang, Z., & Zyphur, M. J. (2011). Alternative methods for assessing mediation in multilevel data: The advantages of multilevel SEM. Structural Equation Modeling, 18 , 161-182.
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for assessing multilevel mediation. Psychological Methods, 15 (3), 209-233.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage.