Bayesian Psychometric Modeling (Fall 2024)

Course Information

Instructor Jonathan Templin
Email jonathan-templin@uiowa.edu
Office S300A Lindquist Center
Office Phone 319-335-6429
Classroom Zoom Only (meeting link)
Meeting Time T & Th 9:30am–10:45am
Office Hours W 10am–12pm on Zoom (link)
Course YouTube Playlist Playlist
IDAS Notebooks Site notebooks.hpc.uiowa.edu/fall2024-psqf-7375-0exw
GitHub Repository jonathantemplin/Bayesian-Psychometric-Modeling-Course-Fall-2024
Syllabus bpm24_syllabus.pdf
Teaching Assistant Bladimir Padilla
TA Email geraldo-padilla@uiowa.edu
TA Office Hours W 9am–10am and F 10am–12pm in N476LC and on Zoom (link)

Course Objectives, and Prerequisites

In this course, a unified Bayesian modeling approach will be presented across traditionally separate families of psychometric models. Focusing more directly how to use Bayesian methods in psychometrics, this course will to cover Bayesian theory along with applied treatments of popular psychometric models, including confirmatory factor analysis (CFA), item response theory (IRT), latent class analysis, diagnostic classification models, and Bayesian networks. The focus of this course will be on model building directly in Bayesian programs (i.e., Stan and JAGS) rather than the use of packages that build such code automatically.

Time permitting, multilevel models and multilevel psychometric models will be presented.

Course Schedule and Content

27 Aug — Course Introduction

29 Aug — Introduction to Bayesian Concepts (Lecture 01)

2 Sep

  • Formative Assessment: FA 1 (via ICON)

3 & 5 Sep — Introduction to Psychometric Models (Lecture 02)

9 Sep

  • Formative Assessment: FA 2 (via ICON)

10, 12 & 17 Sep — MCMC and Stan (Lecture 03a)

17 Sep — Example Bayesian Linear Model (Lecture 03a Example)

19 Sep — No live class (JT at conference)

Listen to the Learning Bayesian Statistics Podcast episode on which I appeared — link on my main website.

24 & 26 Sep — No live class (JT at conference)

Work on HW 2.

1 & 3 Oct — Example Bayesian Linear Model (continued)

3 Oct — Efficient Stan Code and Generated Quantities (Lecture 03b)

8 & 10 Oct — Bayesian Model Fit and Comparisons (Lecture 03c)

Note: No live lecture on 8 Oct.

15 & 17 Oct — Generalized Measurement Models (Lecture 04a)

17 & 22 Oct — Modeling Observed Data (Lecture 04b)

22 Oct

24, 29 & 31 Oct — Modeling Observed Dichotomous Data (Lecture 04c)

29 Oct

  • Formative Assessment: FA 7 (via ICON)

5 & 7 Nov — Modeling Observed Polytomous Data (Lecture 04d)

Note: 7 Nov has no “live” lecture — see the Nov 7 video above for the end of the polytomous data lecture.

12 & 19 Nov — Modeling Multidimensional Latent Variables (Lecture 04e)

14 Nov — No class

19 & 21 Nov — Bayesian Psychometric Model Fit (Lecture 04f)

21 Nov — Missing Data (Lecture 04g)

3 Dec — Empirical Priors (Lecture 04h)

3 & 5 Dec — Scale Identification (Lecture 04i)

References

Levy, R., & Mislevy, R. J. (2016). Bayesian Psychometric Modeling (1st ed.). Chapman and Hall/CRC. https://doi.org/10.1201/9781315374604

McDonald, R. P. (1999). Test theory: A unified treatment. Erlbaum.