Last updated November 5, 2026
Each course family is outlined by topic using its most recent version (the Clemson course where one exists). Under each family, Also covered in older versions lists topics that appear only in earlier courses, tagged with the course and term. Topics are drawn from the lecture and example titles on each course page; all materials are linked from the course directory.
| Family | Anchor (most recent) | Older versions |
|---|---|---|
| Intro GLM / regression | EDF 9770, Clemson, Spring 2026 | PSQF 6243 (Fall 2025), PSQF 6242 (Spring 2021), SPLH 861 (Fall 2014), PSYC 943 (Fall 2012) |
| Clustered multilevel models | EDF 9850, Clemson, Fall 2026 | PSQF 6272 (Fall 2025) |
| Longitudinal multilevel models | PSQF 6271, Iowa, Fall 2024 | CLDP 944 (Fall 2017), PSYC 944 (Spring 2013), 2014 summer workshop |
| Advanced multilevel / longitudinal | PSQF 7375, Iowa, Spring 2025 | CLDP 945 (Spring 2018), PSYC 945 (Spring 2014) |
| Generalized linear models | PSQF 6270, Iowa, Spring 2025 | PSYC 930 Generalized (Summer 2013), PSYC 943 (Fall 2012) |
| Factor analysis / SEM / measurement | PSQF 6249, Iowa, Spring 2024 | CLDP 948 (Fall 2018), PSYC 948 (Spring 2014) |
| Software | PSYC 930 Make Friends with SAS, Summer 2013 | none |
Anchor: EDF 9770, Spring 2026 (R). PSQF 6243 (Fall 2025) follows the same six-unit sequence; PSQF 6242 split some units differently.
Class videos: EDF 9770 YouTube playlist
| Topic | EDF 9770 materials |
|---|---|
| Course introduction | Lecture 0 |
| Univariate data description | Lecture 1 |
| GLMs with single predictors (incl. power tables) | Lecture 2, Example 2 |
| GLMs with multiple slopes for one predictor | Lecture 3, Example 3 |
| GLMs with multiple conceptual predictors | Lecture 4, Example 4 |
| GLMs with interactions | Lecture 5, Example 5 |
| Factorial coding (bonus) | Bonus example |
| Caveats and next steps | Lecture 6 |
| Topic | Where |
|---|---|
| Review and multiple-predictor GLM (extra worked example) | PSQF 6243 (2025), Example 4b |
| Bivariate association and significance testing (stand-alone unit) | PSQF 6242 (2021), Lecture/Example 2 |
| Univariate inference (paired with description) | PSQF 6242 (2021), Lecture 1 |
| GLMs with multiple-slope interaction effects | PSQF 6242 (2021), Lecture/Example 7 (bonus) |
| Interactions among continuous predictors; among categorical predictors | SPLH 861 (2014), Lectures 2–3 |
| "Other" kinds of effects in GLMs | SPLH 861 (2014), Lecture 4 |
| Intro to multivariate and repeated measures models (piecewise age effects) | SPLH 861 (2014), Lecture/Example 5 |
| Simple, marginal, and interaction effects in GLMs | PSYC 943 (2012), Lectures 2–3 |
| Univariate normal distribution; least squares vs. maximum likelihood estimation | PSYC 943 (2012), Lectures 4–6 |
| Matrix algebra and PROC IML; mean vectors, covariance matrices, multivariate normal | PSYC 943 (2012), Lectures 10–12 |
| Multivariate models vs. classical MANOVA; multivariate regression; difference scores | PSYC 943 (2012), Lectures 13–15 |
| Repeated measures and dyadic/family data as multivariate models | PSYC 943 (2012), Lecture 16 |
| Bayesian and MCMC estimation | PSYC 943 (2012), Lecture 17 |
| Missing data: ML and multiple imputation | PSYC 943 (2012), Lecture 18 |
| Power analysis via PROC POWER and Monte Carlo simulation | PSYC 943 (2012), Lectures 21–22 |
| Mixture and latent class models | PSYC 943 (2012), Lecture 23 |
| Principal components and exploratory factor analysis | PSYC 943 (2012), Lecture 24 |
SPLH 861 also introduced crossed random effects, clustered data, and generalized models; those are listed under their own families.
Anchor: EDF 9850, Fall 2026 (R), in progress. It covers two-level nesting, adds a longitudinal unit, and ends with cross-classified designs. The Fall 2023 version of PSQF 6272 went further into generalized, multivariate, and three-level models.
Class videos: EDF 9850 YouTube playlist
| Topic | EDF 9850 materials |
|---|---|
| Introduction to the course and to MLMs | Lecture 1 |
| From empty models to level-2 predictors | Lecture 2, Example 2 |
| Fixed slopes of level-1 predictors | Lecture 3, Example 3 |
| Review of interactions | Slides |
| Random slopes and cross-level interactions: cluster-mean-centered level-1 predictors | Lecture 4, Example 4a |
| Random slopes and cross-level interactions: hybrid level-1 predictors | Example 4b |
| Centering and interpretation; predictor coding | Smush chart, predictor coding |
| Twin data using lme (bonus) | Bonus files |
| Longitudinal models: time-varying predictors of within-person fluctuation | Lecture 5, Example 5a |
| Longitudinal models: time-invariant predictors of within-person change | Example 5b |
| Two-level cross-classified designs: crossed primary and secondary schools; subjects crossed with items | Lecture 6, Example 6a, Example 6b |
| Topic | Where |
|---|---|
| Generalized MLMs for two-level nested data: categorical outcomes | PSQF 6272 (2023), Lecture 6, Example 6a |
| Generalized MLMs for two-level nested data: count outcomes | PSQF 6272 (2023), Example 6b |
| Generalized MLMs for persons crossed with items (explanatory IRT) | PSQF 6272 (2023), Lecture/Example 7 |
| Multivariate multilevel models (in Mplus) | PSQF 6272 (2023), Lecture 8, Example 8a |
| Three-level nested models | PSQF 6272 (2023), Example 8b |
| Changes in nesting over time (bonus) | PSQF 6272 (2023), bonus example |
| Location-scale models; power analysis for MLMs (bonus readings) | PSQF 6272 (2023) |
| Intro to clustered data: students within schools | SPLH 861 (2014), Lecture/Example 7; PSYC 944 (2013), Lecture/Example 10 |
EDF 9850 materials are listed through Week 12 (as of November 5, 2026).
Anchor: PSQF 6271, Fall 2024 (SAS, Stata, or R). It follows the Longitudinal Analysis textbook through chapter 7 (time-invariant predictors). Time-varying predictors appear in EDF 9850 Lecture 5 and in the advanced family.
Class videos: PSQF 6271 YouTube playlist
| Topic | PSQF 6271 materials |
|---|---|
| Introduction to the course and to MLMs for longitudinal data | Lecture 1 |
| Review of general linear models | Self-review of PSQF 6270 Lecture 1 and Example 1 |
| Within-person analysis and RM ANOVA: between- vs. within-person models; kinds of ANOVA | Lecture 3, Example 3a, Example 3b |
| Within-person fluctuation via alternative covariance structure (ACS) models; continuous-time AR1 (bonus) | Lecture 4, Example 4, bonus slides |
| Random effects of time and model estimation (incl. longitudinal models as single-level SEMs) | Lecture 5, Example 5 |
| Describing within-person change: polynomial trends | Lecture 6, Example 6a |
| Piecewise trends | Example 6b |
| Truly exponential (latent basis) models | Example 6c |
| Log time to approximate exponential trends | Example 6d |
| Review of unconditional models of time | Lecture 7a |
| Time-invariant predictors in longitudinal models | Lecture 7b, Example 7b |
| Topic | Where |
|---|---|
| Getting started in SAS | CLDP 944 (2017), Make Friends with SAS (see Software) |
| GLM review in SAS MIXED; interactions among continuous and categorical predictors | CLDP 944 (2017), Lectures/Examples 2a–2c; PSYC 944 (2013), Lecture 2 |
| Exponential models of change via NLMIXED | PSYC 944 (2013), Example 6c |
| Decomposing interactions of time-invariant predictors in models of change | PSYC 944 (2013), Example 7b |
| Repeated measures designs not involving time: crossed persons and words; eye movements | PSYC 944 (2013), Lecture 8, Examples 8a–8b (now EDF 9850 Lecture 6) |
| Time-varying predictors of within-person fluctuation: lagged effects; between- vs. within-person effects | PSYC 944 (2013), Lecture 9, Examples 9a–9b (now EDF 9850 Lecture 5) |
| MLMs for clustered data: children within schools | PSYC 944 (2013), Lecture/Example 10 (see Clustered family) |
| Alternative metrics of time | PSYC 944 (2013), Lecture/Example 11 |
The 2014 summer workshop condensed the intro, change, time-invariant predictor, and crossed subjects-and-items topics into two days, with examples in SPSS, SAS, and Stata.
Anchor: PSQF 7375, Spring 2025 (SAS, Stata, or R; Mplus). It centers on time-varying predictors, comparing univariate MLM with SEM and multilevel SEM (M-SEM). The older Advanced MLM courses ranged more widely, into generalized, three-level, and crossed designs.
Class videos: PSQF 7375 YouTube playlist
| Topic | PSQF 7375 materials |
|---|---|
| Course introduction | Lecture 0 |
| Review of longitudinal multilevel models | Lecture 1 |
| Time-varying predictors of within-person fluctuation in univariate MLM | Lecture 2, Example 2 |
| Alternative metrics of time in accelerated longitudinal designs (univariate MLM) | Lecture 3, Example 3 |
| Longitudinal analysis via SEM and M-SEM; alternative metrics of time in SEM and M-SEM | Lecture 4, Example 4a |
| Latent factors for change over time in SEM | Example 4b |
| Time-varying predictors in SEM and M-SEM: within-person fluctuation and mediation | Lecture 5, Example 5a |
| Multivariate change in SEM and M-SEM | Example 5b |
| Time-smushed effects (bonus talk); intensive longitudinal data and location-scale models (bonus resource list) | Talk slides, resource list |
| Topic | Where |
|---|---|
| Model assumptions; predicting heterogeneity of variance (location-scale models) | CLDP 945 (2018), Lecture 4, Example 4b; PSYC 945 (2014), Example 2 |
| Models for twin analysis | CLDP 945 (2018), Example 4a (bonus twin files now in EDF 9850) |
| Crossed subjects and items; eye movements | CLDP 945 (2018), Lecture 5, Examples 5a–5b (now EDF 9850 Lecture 6) |
| Two-level clustered, cross-classified, and changes-in-nesting models | CLDP 945 (2018), Lecture 6, Examples 6a–6c; PSYC 945 (2014), Lecture 4, Examples 7b–7c |
| Generalized MLMs: clustered binary outcomes | CLDP 945 (2018), Example 7a; PSYC 945 (2014), Example 6a |
| Generalized longitudinal models: ordinal outcomes; binomial percent correct | CLDP 945 (2018), Examples 7b–7c; PSYC 945 (2014), Examples 6b–6c |
| Longitudinal models with other non-normal outcomes (Mplus) | PSYC 945 (2014), Example 6d |
| Explanatory IRT models as crossed random effects (SAS GLIMMIX) | CLDP 945 (2018), Example 7d (also PSQF 6272 (2023) Lecture 7) |
| Three-level models: longitudinal twin data; intensive longitudinal (EMA) designs; clustered longitudinal data | CLDP 945 (2018), Lecture 8, Examples 8a–8b; PSYC 945 (2014), Lecture 7, Example 7a |
| Cross-sectional and longitudinal difference score models | PSYC 945 (2014), Examples 5a–5b |
| Multivariate within-family (sibling) change | PSYC 945 (2014), Example 5d |
| Ways of estimating multivariate change in MLM vs. SEM (3 and 7 approaches) | CLDP 945 (2018), Example 9c; PSYC 945 (2014), Example 5f |
| Story time with time-invariant predictors | CLDP 945 (2018), Example 1 |
Anchor: PSQF 6270, Spring 2025 (Stata or R; Mplus for path models). It runs from binary through count and skewed outcomes, then into multivariate models (path models are in section 7).
Class videos: PSQF 6270 YouTube playlist
| Topic | PSQF 6270 materials |
|---|---|
| Course introduction (incl. log-likelihood calculations) | Lecture 0 |
| Review of general linear models (reference categories, interactions) | Lecture 1 |
| Models for categorical outcomes: binary | Lecture 2, Example 2a |
| Models for ordinal and nominal outcomes | Example 2b |
| Models for count outcomes | Lecture 3, Example 3 |
| Models for other non-normal outcomes: ceiling or floor effects | Lecture 4, Example 4a |
| Skewed continuous outcomes (incl. quantile regression) | Example 4b |
| Multivariate models via univariate software: multivariate family outcomes | Lecture 5, Example 5 |
| Factorial design coding; repeated measures and difference score outcomes (bonus) | Factorial coding; RM and difference-score examples in the 2020 version |
| Topic | Where |
|---|---|
| Models for proportion outcomes; binomial repeated measures for percent correct | PSYC 930 Generalized (2013), Part 1; SPLH 861 (2014), Example 10a |
| Zero-inflated and "if and how much" outcomes | PSYC 930 Generalized (2013), Part 2; SPLH 861 (2014), Lecture 11 |
| Generalized multivariate and multilevel models (overview) | PSYC 930 Generalized (2013), Part 2 (multilevel versions also in PSQF 6272 (2023)) |
PSYC 943 (Lectures 7–9) and SPLH 861 (Lectures 9–11) covered binary, ordinal, nominal, and count models in SAS; those topics are all in PSQF 6270.
Anchor: PSQF 6249, Spring 2024 (Mplus or R lavaan). The sequence has stayed stable since 2014: classical test theory, CFA, IRT/IFA, invariance, higher-order factors, then SEM.
Class videos: PSQF 6249 YouTube playlist
| Topic | PSQF 6249 materials |
|---|---|
| Introduction to the course and to latent trait measurement models | Lecture 1 |
| Exploratory factor analysis and PCA (bonus) | Lecture 2 from the Fall 2022 version |
| Classical test theory for scale reliability; classical item analysis (R, Stata, SPSS, SAS) | Lecture 3, Example 3 |
| Confirmatory factor analysis, parts 1–2 (Mplus, R, a little SAS MIXED) | Lecture 4a, Lecture 4b, Example 4 |
| Latent trait models for binary responses (IRT/IFA) | Lecture 5a, Lecture 5b, Example 5 |
| Latent trait models for other item responses: graded response models for ordinal items | Lecture 6, Example 6a |
| Semi-ordered (not applicable) responses; other non-normal outcomes (bonus) | CLDP 948, Example 6b, Example 6c |
| Measurement invariance in CFA and DIF in IRT/IFA: multiple-group and longitudinal CFA; multiple-group IFA (WLSMV and ML) | Lecture 7, Examples 7a, 7b, 7c, 7d |
| Higher-order and method factor models | Lecture 8, Example 8 |
| Structural equation modeling and alternatives | Lecture 9, Example 9 |
| Topic | Where |
|---|---|
| Exploratory factor analysis and PCA (full lecture) | CLDP 948 (2018), Lecture 2; PSYC 948 (2014), Lecture 2; PSYC 943 (2012), Lecture 24 |
| Introduction to generalized models (within a measurement course) | PSYC 948 (2014), Lecture 5 |
| IRT models in SAS NLMIXED | PSYC 948 (2014), Example 7c |
| Explanatory item response models in SAS GLIMMIX | PSYC 948 (2014), Lecture/Example 8 (also PSQF 6272 (2023) Lecture 7) |
| Mixture and latent class models | PSYC 943 (2012), Lecture 23 |
Most recent coverage: PSQF 6270 Lecture 6 (Spring 2025) for single-level path models, plus PSQF 7375 for longitudinal mediation in SEM and M-SEM. No course has a stand-alone unit; these topics are drawn from several families.
Class videos: PSQF 6270, PSQF 7375, and PSQF 6249 YouTube playlists
| Topic | Most recent materials |
|---|---|
| Multivariate models via path analysis | PSQF 6270, Lecture 6 |
| Path models for mediation with normal outcomes | PSQF 6270, Example 6a |
| Path models for mediation with binary outcomes | PSQF 6270, Example 6b |
| Path models for mediation with nominal outcomes | PSQF 6270, Example 6c |
| Time-varying predictors and mediation of within-person fluctuation in M-SEM and SEM | PSQF 7375, Lecture 5, Example 5a |
| Structural equation modeling (latent-variable extension of path models) | PSQF 6249, Lecture 9, Example 9 |
| Topic | Where |
|---|---|
| Introduction to Mplus and path analysis | PSYC 943 (2012), Lecture 19 |
| Mediation models | PSYC 943 (2012), Lecture 20 |
| Introduction to path analysis and mediation | SPLH 861 (2014), Lecture/Example 8 |
| Path models with generalized outcomes in Mplus | SPLH 861 (2014), Example 12 |
| Path modeling with mediation; path modeling with non-normal outcomes | CLDP 948 (2018), Lecture 9, Examples 9a–9b |
| Mediation of within-person fluctuation (SAS and Mplus) | CLDP 945 (2018), Example 9b; PSYC 945 (2014), Example 5e |
PSYC 930 Make Friends with SAS, Summer 2013, is the only stand-alone software course. Current courses teach R within the course (see the R demo video and Lecture 1 files in EDF 9770 and the class R functions in EDF 9850).
| Day | Topic |
|---|---|
| 1 | SAS data manipulation: import, export, merge, restructure |
| 2 | Working with variables in SAS |
| 3 | Increasing efficiency with arrays, DO loops, and macro programs |
| 4 | Exporting SAS output and using PROC GLM |
CLDP 944 (Fall 2017) reused these materials as a SAS primer. Syntax guides for SAS, Stata, SPSS, and Mplus are at PilesOfVariance.com.