Instructor Contact Information: |
Professor Lesa Hoffman (she/her—you can call me Lesa) Department of Education and Human Development Email: LesaH@Clemson.edu (preferred mode of contact) Office: 213D Tillman Hall (infrequently used) | Zoom Link for Class and Office Hours: | https://clemson.zoom.us/j/8236434097 Meeting ID: 823 643 4097; Mobile Access: +19292056099 (please use your real name as your account name to be admitted) |
| Course Location and Time: |
206 Tillman Hall or via zoom Thursdays 5:00–7:45 PM Eastern | Zoom-Only Office Hours | Mondays 10–11 AM and Tuesdays 12–1 PM Eastern in a group format (first-come, first-served, no appointments) or individually by request |
| Coursework- Related Sites: |
Canvas Online Homework System (now available!) YouTube Playlist of Course Recordings |
R Package Dcoumentation: |
TeachingDemos,
readxl,
pacman,
lm,
multcomp,
nlme,
lme4,
performance,
lmerTest,
psychometric,
r2mlm,
prediction,
emmeans Class R Functions (updated 11/6/26): R Source File and Documentation (pdf) (html) List of R packages hosted on CRAN |
Week |
Class / Due Date |
Materials |
Resources |
|
|---|---|---|---|---|
| 1 | Due by 11:59 PM | 8/19 | No Formative Assessments (FA), Homework (HW), Activity Answers, or Project Components Due | Hoffman & Walters (2022) |
| During Class | 8/20 | MEET ON ZOOM ONLY Lecture 1: Introduction to this Course and to Multilevel Models (MLMs) Video Part 1: Lecture 1 slides 1-13 (in class) Video Part 2: Lecture 1 slides 14-25 (out of class) |
||
| 2 | Due by 11:59 PM | 8/26 | Watch Lecture 1: Video Part 2 (slides 14-15) Watch Lecture 2: Video Part 1 (slides 1-19) and Video Part 2 (slides 20-39) FA1 (Quiz in Canvas) Install R software and then install RStudio software for use in class |
|
| During Class | 8/27 | Discussion of HW and FA1: Video Lecture 2 and Example 2: From Empty Models to Level-2 Predictors (corrected 9/10/26) Activity 1 (updated 8/27/26): Summary Video Unit 2 R Source Files (updated 8/27/26) Video Part 1: Example 2 pages 1-3 |
McNeish (2017) | |
| 3 | Due by 11:59 PM | 9/2 | Watch Lecture 3: Video Part 1 (slides 1-19) FA2 (Quiz in Canvas) HW0 (online, for 2 points extra credit over the syllabus): Video Demo Activity 1 (Assignment in Canvas) |
|
| During Class | 9/3 | Video: Discussion of course requirements and Activity 1 Video Part 2: Example 2 all Video: Discussion of FA2 (terms spreadsheet) Lecture 3 and Example 3: Fixed Slopes of Level-1 Predictors Unit 3 R Source Files Activity 2 (updated 9/2/26) |
Rights & Sterba (2019, 2020) Yaremych et al. (2023) |
|
| 4 | Due by 11:59 PM | 9/9 | Watch Lecture 3: Video Part 2 (slides 18-42) FA3 (Quiz in Canvas) Activity 2 (Assignment in Canvas) |
|
| During Class | 9/10 | Discussion of FA3 and Activity 2 Lecture 3 and Example 3, continued Video Part 1: Example 3 pages 1-6 Video Part 2: Example 3 pages 6-8 Activity 3 (updated 9/10/26) |
Hamaker & Muthén (2020) McNeish (2026) |
|
| 5 | Due by 11:59 PM | 9/16 | Watch Lecture Reviewing Interactions: Video and
Slides HW1 (online, based on Example 2) Activity 3 (Assignment in Canvas) |
|
| During Class | 9/17 | MEET ON ZOOM ONLY Discussion of Activity 3 Lecture 3 and Example 3, continued Video Part 3: Discussion of Activity 3; Example 3 pages 5-10 Video Part 4: Example 3 pages 11-16; Discussion of Projects |
||
| 6 | Due by 11:59 PM | 9/23 | Watch Lecture 4: Video Part 1 (slides 1-15) Project Plan (Assignment in Canvas): Document to Complete (corrected 9/4/26) |
|
| During Class | 9/24 | Lecture 4 (updated 9/25/26): Random Slopes and Cross-Level Interactions Example 4a: Random Slopes and Cross-Level Interactions of Cluster-Mean-Centered Level-1 Predictors Video Part 1: Example 4a pages 1-3 Video Part 2: Example 4a pages 4-11 Unit 4 R Source Files Activity 4 |
Rights & Sterba (2023) Rights (2023) |
|
| 7 | Due by 11:59 PM | 9/30 | Watch Lecture 4: Video Part 2 (slides 15-34) FA4 (Quiz in Canvas) Activity 4 (Assignment in Canvas) |
|
| During Class | 10/1 | MEET ON ZOOM ONLY Video Part 1: Discussion of FA4 and Activity 4 (FA4 models) Video Part 2: Centering and Interpretation (excel smush chart) |
||
| 8 | Due by 11:59 PM | 10/7 | Optional —
Watch Lecture 4: Video Part 3 (slides 35-61) Project Plan Revisions (if needed, Assignment in Canvas) HW2 (online, based on Example 3) |
Enders (2010) ch. 3–5 Hoffman ch. 5 |
| During Class | 10/8 | Lecture 4, continued Example 4b: Random Slopes and Cross-Level Interactions of Hybrid Level-1 Predictors Example 4b R Source Files (no data) Video: Example 4b all (excel predictor coding) Bonus Twin Example using lme Activity 5 (updated 10/8) |
||
| 9 | Due by 11:59 PM | 10/14 | Watch Lecture 5: Video Part 1 (slides 1-28) FA5 (Quiz in Canvas) Activity 5 (Assignment in Canvas) |
|
| During Class | 10/15 | Discussion of FA5 and Activity 5 Lecture 5: Longitudinal Models Example 5a: Time-Varying Predictors of Within-Person Fluctuation Example 5a R Source Files Video Part 1: Discussion of Activity 5 and FA 5; Example 5a pages 1-7 Video Part 2: Example 5a pages 8-13 |
Hoffman (2015) ch. 1, 6–9 | |
| 10 | Due by 11:59 PM | 10/21 | HW3 (online, based on Example 4b) | |
| During Class | 10/22 | NO CLASS Watch Lecture 5: Video Part 2 (slides 28-58) |
Padilla et al. (2026) | |
| 11 | Due by 11:59 PM | 10/28 | Project Check-In (Assignment in Canvas): Document to Complete (updated 10/7) Project Rubric; Peer-Review Document to Complete |
|
| During Class | 10/29 | Lecture 5, continued Video: Discussion of Projects and Lecture 5 Review Example 5b: Time-Invariant Predictors of Within-Person Change Example 5b R Source Files Video Part 1: Example 5b pages 1-7 Video Part 2: Example 5b pages 7-15 |
Walters & Hoffman (2017) Hoffman (2025) |
|
| 12 | Due by 11:59 PM | 11/4 | Watch Lecture 6: Video FA6 (Quiz in Canvas) |
|
| During Class | 11/5 | Discussion of FA6 Lecture 6: Two-Level Cross-Classified Designs Example 6a: Crossed Primary and Secondary Schools Example 6a R Source Files Video: Discussion of FA6 and Example 6a (all) Example 6b: Subjects Crossed with Items Example 6b R Source Files Video: Example 6b all (out of class) |
Guo et al. (2024) Hoffman (2015) ch. 12 ten Hove et al. (2022) |
|
| 13 | Due by 11:59 PM | 11/11 | Project Check-In Revisions (if needed, Assignment in Canvas) | |
| During Class | 11/12 | MEET ON ZOOM ONLY Student Project Presentations |
||
| 14 | Due by 11:59 PM | 11/18 | Nothing Due | |
| During Class | 11/19 | MEET ON ZOOM ONLY Student Project Presentations |
||
| 15 | Due by 11:59 PM | 11/25 | Nothing Due | |
| During Class | 11/26 | NO CLASS | ||
| 16 | Due by 11:59 PM | 12/2 | Project Peer Reviews (via email) FA7 (Quiz in Canvas) |
|
| During Class | 12/3 | MEET ON ZOOM ONLY Discussion of FA7 Study Planning; TBD |
Arend & Schäfer (2019) Hoffman (2015) ch. 13 |
|
| 17 | Due by 11:59 PM | 12/11 | Project Presentation Revisions and All Other Work | |
Introduction to multilevel modeling, focusing on designs common in educational research, such as students nested within classrooms or schools. Covers the specification, estimation, and interpretation of random intercept and random slope models, with attention to analyzing grouped data and contextual effects.
This course will use R software for statistical analysis, which is free online. Participants will need to first install R software and then install RStudio software. Alternatively, participants can access R through CU APPS. For help getting started with R, Watch this R Demo Video based on Lecture 1 Files from EDF 9770.
- The planned schedule of topics and events given here may need to be adjusted during the course. The schedule above will always have the most current schedule of events and due dates.
- The instructor will provide announcements and reminders via email; larger changes will also be announced through Canvas. Emails will be answered within 1–2 business days barring unforeseen circumstances.
- Instructor office hours will be held only on zoom in a group format (first-come, first-served), in which multiple participants can receive assistance in succession (or simultaneously) without an appointment. Individual zoom meetings with the instructor are also available upon request.
- This “HyFlex” course will meet synchronously in person in 206 Tillman Hall and in the instructor’s zoom room Thursdays from 5:00 PM to 7:45 PM at the latest. Participants may change their modality—in person as a “roomer” or online as a “zoomer”—without notice each week as desired.
- Specific class sessions will be held only on zoom for logistical reasons, as noted in the online syllabus.
- Attendance is expected and strongly encouraged but is not formally required. Auditors and visitors are always welcome. No required sessions will be held outside the regular class time (i.e., no exam sessions).
- If you are seriously ill, please DO NOT attend class in person! Similarly, if the instructor has been exposed to illness or the weather prohibits safe travel to class, course meetings will move to a zoom-only format.
- This course will illustrate the uses of multilevel models (aka, general linear mixed-effect models, hierarchical linear models), primarily for the analysis of clustered data (i.e., persons nested in naturally-occurring groups), and potentially for longitudinal data (i.e., occasions nested in persons), time permitting.
- The overall course learning objective is for participants to be able to complete all the necessary steps in a multilevel analysis : deciding which type of model is appropriate; organizing the data for analysis and creating predictor variables; testing fixed and random effects; predicting multiple sources of variation; and interpreting and presenting one's own empirical findings and those reported by others.
- Participants should know how to use at least one statistical software package and be comfortable with general linear models (e.g., regression, ANOVA). Review materials are available within EDF 9770.
- Class time will be devoted primarily to lectures, examples, discussions, and in-class activities, the materials for which will be available at the table above. Canvas will only be used for announcements, collecting work, and access to readings (as suggested for each unit and which may be updated later).
- Video recordings of portions of each class will be hosted on a class-specific YouTube playlist so that searchable closed captioning will be provided, and out-of-class recordings will be posted to the playlist as well. Only the screen share and classroom/instructor audio will be captured on these video recordings.
- The instructor wants all participants to feel welcome and encouraged to actively participate in this course. There is no such thing as a “stupid” question (or answer). All course participants—enrolled students and auditing visitors—should always feel welcome to ask whatever questions will be helpful to understand the course content. Questions are welcome at any point: during class (aloud or using the zoom chat window, either to everyone or a direct message to the instructor), in office hours, over email, or in individual appointments with the instructor (by request). Students with special needs are encouraged to contact the instructor for a confidential discussion of their individual considerations for academic accommodation.
- To facilitate communication and connection across modalities, both roomers and zoomers should be logged into zoom with their cameras on (with audio muted for roomers). Please use the name you wish for us to call you so that it appears on your zoom window while in use. Please do not log into zoom while driving or any other potentially hazardous locations. During zoom-only classes, students may join zoom from the in-person classroom if desired.
- Rather than missing class, it is preferable to join late, leave early, or re-join during class as needed.
- It is expected that the use of cell phones, non-class-related applications, and other distractions are kept to a minimum during class in either modality.
Participants can earn up to 100 total points by completing course requirements as follows:
- Up to 10 points can be earned from submitting answers to group-based in-class activities (5 planned initially) using Canvas. These are designed to reinforce lecture concepts through visualization and applied practice. Participants may work through the questions alone or in small groups of their choosing, and then responses will be discussed as a class. For each, 2 points will be given automatically for reasonable answers (generated without the use of AI) to the activity's questions (to be submitted using Canvas before the next class as an individual or group, as was selected during class). Participants who miss class may still complete the activity outside of class and submit their individual answers separately to receive credit.
- Up to 14 points can be earned from submitting formative assessments (FAs; 7 planned initially) using Canvas. These are designed to provide a comprehension check, especially over material covered in videos watched outside of class. They should be completed individually without the use of AI (which defeats their purpose). For each, 2 points will be given automatically for brief but thoughtful completion regardless of accuracy, and points may be reduced for AI-generated responses. No individual feedback will generally be given, but responses will be summarized and discussed during the next class.
- Up to 32 points can be earned from submitting homework assignments (HWs; 3 planned initially) through the instructor's custom online system. These are designed to provide scaffolded practice doing data analysis, should be completed individually, and will be graded for accuracy. The computational questions allow infinite attempts, whereas the multiple-choice results questions allow one attempt. Complete individual feedback will be available after the due date passes. HWs will be based directly on examples given in class, such that no AI use should be needed. However, using AI to better understand the syntax or how to interact with RStudio is permitted. To ensure participants are building their own examples of syntax and output for their Future Selves, the syntax and output used to complete each HW must also be submitted individually using Canvas before any homework points will be given. Nearly-identical files submitted by different students may result in a loss of HW points and/or disciplinary action.
- Up to 44 points may be earned from completing a data analytic project alone or in pairs. Each student or pair will give a short presentation on zoom designed to mimic a presentation at an academic conference (25 points). Written components include a plan (5 points; describing the sample, variables, and research questions), check-in (10 points; documenting the analyses and results to be presented), and a peer review (4 points; providing feedback for another presentation). AI may be used to assist in data management, analysis, and creation of graphics or tables, but not for peer reviews. Each student must be able to explain and defend their work, as well as provide syntax and output documenting their process. Presentations may be revised once via a submitted recording. Details and a rubric will be shared later.
- Up to 2 points of extra credit can be earned through HW 0; there may be other extra credit opportunities at the instructor's discretion.
- Revisions to the planned course schedule and/or content may result in fewer requirements (and thus fewer total points) at the instructor's discretion. If so, this syllabus will be updated to reflect the new point totals. Further, the due dates currently listed above may be pushed later if needed, but never sooner.
Because the material for this course builds cumulatively, the following policies are designed to reinforce the need for participants to complete coursework in a consistent and timely manner.
- Participants may submit coursework up to 2 weeks late with these point penalties: in-class activity (−1), FA (−1), HW (−3), project plan (−2), project check-in (−4), presentation (−10), and peer review (−2) penalty.
- Extensions without penalty will be granted as needed for extenuating circumstances (e.g., conferences, comprehensive exams, family obligations) if requested in writing at least two weeks in advance of the due date.
- A final grade of "incomplete" will only be given in unforeseen dire circumstances for participants who have otherwise kept up with the coursework and entirely at the instructor's discretion. All work must be submitted by 11:59 PM on Friday, December 11, 2026, to be included in the course grade.
- Final grades will be determined by the percentage earned out of the total possible points:
≥93 = A, 90–92% = A−, 87–89% = B+, 83–86% = B, 80–82% = B−, 77–79% = C+, 73–76% = C, 70–72% = C− (PASS), 67–69% = D+, 63–66% = D, 60–62% = D−, <60% = FThe instructor realizes that this course is not your only obligation in your work or your life. While class attendance in real time is not strictly mandatory, it is strongly encouraged because frequent review of the material will be your best strategy for success in this course. However, if work or life events may compromise your ability to succeed, please contact the instructor for a confidential discussion so that we can work together to make a plan for your success. Please do not wait until you are too far behind to try to catch up!
Arend, M. G., & Schäfer, T. (2019). Statistical power in two-level models: A tutorial based on Monte Carlo simulation. Psychological Methods, 24(1), 1–19. https://doi.org/10.1037/met0000195
Enders, C. K. (2010; chapters 3–5). Applied missing data analysis (1st ed). Guilford.
Guo, Y., Dhaliwal, J., & Rights, J. D. (2024). Disaggregating level-specific effects in cross-classified multilevel models. Behavior Research Methods, 56, 3023–3057. https://doi.org/10.3758/s13428-023-02238-7
Hamaker, E. L., & Muthén, B. (2020). The fixed versus random effects debate and how it relates to centering in multilevel modeling. Psychological Methods, 25(3), 365–379. https://doi.org/10.1037/met0000239
Hoffman, L. (2015). Longitudinal analysis: Modeling within-person fluctuation and change. Routledge/Taylor & Francis. Available for free through the Clemson University library at this link.
Hoffman, L. (2025). Disaggregating associations of between-person differences in change over time from within-person associations in longitudinal data. Multivariate Behavioral Research, 60(6), 1312–1330. https://doi.org/10.1080/00273171.2025.2519348
Hoffman, L., & Walters, R. W. (2022). Catching up on multilevel modeling. Annual Review of Psychology, 73, 629–658. https://doi.org/10.1146/annurev-psych-020821-103525
McNeish, D. (2017). Small sample methods for multilevel modeling: A colloquial elucidation of REML and the Kenward-Roger correction. Multivariate Behavioral Research, 52(5), 661–670. https://doi.org/10.1080/00273171.2017.1344538
McNeish, D. (2026). A practical guide to selecting and blending approaches for clustered data: Clustered errors, multilevel models, and fixed-effect models. Psychological Methods, 31(2), 225–251. https://doi.org/10.1037/met0000620
Padilla, B., Dorman, E., & Hoffman, L. (2026). Multilevel analysis for communication scientists. In L. Shen (Ed.) Quantitative research methods in communication science: Data analysis (pp. 95–122). De Gruyter Brill.
Rights, J. D. (2023). Aberrant distortion of variance components in multilevel models under conflation of level-specific effects. Psychological Methods, 28(5), 1154–1177. https://doi.org/10.1037/met0000514
Rights, J. D., & Sterba, S. K. (2019). Quantifying explained variance in multilevel models: An integrative framework for defining R-squared measures. Psychological Methods, 24(3), 309–338. https://doi.org/10.1037/met0000184
Rights, J. D., & Sterba, S. K. (2020). New recommendations on the use of R-squared differences in multilevel model comparisons. Multivariate Behavioral Research, 55(4), 568–599. https://doi.org/10.1080/00273171.2019.1660605
Rights, J. D., & Sterba, S. K. (2023). On the common but problematic specification of conflated random slopes in multilevel models. Multivariate Behavioral Research, 58(6), 1106–1133. https://doi.org/10.1080/00273171.2023.2174490
ten Hove, D., Jorgensen, T. D., & van der Ark, L. A. (2022). Interrater reliability for multilevel data: A generalizability theory approach. Psychological Methods, 27(4), 650–666. https://doi.org/10.1037/met0000391
Walters, R. W., & Hoffman, L. (2017). Applying the hierarchical linear model to longitudinal data. Culture and Education, 29(3), 666–701. https://doi.org/10.1080/11356405.2017.1367168
Yaremych, H. E., Preacher, K. J., & Hedeker, D. (2023). Centering categorical predictors in multilevel models: Best practices and interpretation. Psychological Methods, 28(3), 613–630. http://dx.doi.org/10.1037/met0000434
Darlington, R. B., & Hayes, A. F. (2016). Regression analysis and linear models: Concepts, applications, and implementation. Guilford. Available electronically from the Clemson library.
O’Connell, A. A., McCoach, D. B., & Bell, B. A. (Eds.). (2022). Multilevel modeling methods with introductory and advanced applications. Information Age Publishing.
Hox, J. J., Moerbeek, M., & van de Schoot, R. (2017). Multilevel analysis: Techniques and applications (3rd ed.). Routledge. https://doi.org/10.4324/9781315650982
S & B: Snijders, T. A. B., & Bosker, R. J. (2012). Multilevel analysis: An introduction to basic and advanced multilevel modeling. Sage.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Sage Publications.