NMST 570 Statistical methods in psychometrics
Syllabus (last update: October 10, 2022)Course description in English and in Czech.
Pages of a related course Seminar in Psychometrics.
Pages of last years' course: Winter 2021
Course schedule
| Lecture: | Tuesday, 3:40-4:25pm, K9, Sokolovská 83, Praha 8 - Karlín. |
| Lab session: | Tuesday, 4:25-5:10pm, K9, Sokolovská 83, Praha 8 - Karlín. |
| Online: | Link for ZOOM sessions will be sent to registered students via e-mail. |
News
| (Sep 26, 2022) | Course starts on October 11, 2022. |
Course description
Psychometrics uses statistical models for analysis of educational, psychological, or patient-reported measurements. This course covers computational aspects of main topics in psychometrics including reliability and validity of measurement, traditional item analysis, use of regression models for item description, item response theory (IRT) models, differential item functioning (DIF), computerized adaptive testing (CAT), and an overview of further topics. Methods are demonstrated using data of behavioral measurements from different areas. Exercises are prepared in freely available statistical software R, other psychometric software is also introduced.
Tentative course plan
| Welcome! (4.10.2022) | (No meeting) Welcome message via e-mail |
| Perusall R code on GitHub Project | |
| Lesson 1 (11.10.2022) | Introduction to measurement data |
| Lesson 2 (18.10.2022) | Validity of measurement |
| Lesson 3 (25.10.2022) | (No meeting) Internal structure |
| Lesson 4 (1.11.2022) | (No meeting) Part I of Project due |
| Lesson 5 (8.11.2022) | Reliability and measurement error |
| Lesson 6 (15.11.2022) | Item analysis |
| Lesson 7 (22.11.2022) | Item analysis with regression models |
| Lesson 8 (29.11.2022) | Item response theory (IRT) models |
| Lesson 9 (6.12.2022) | Polytomous and multidimensional IRT models |
| Lesson 10 (13.12.2022) | Differential item functioning (DIF) |
| Lesson 11 (20.12.2022) | Invited talk. |
| Lesson 12 (3.1.2023) | Computerized adaptive testing (CAT), further topics. |
Grading policy
Each week, students are expected to be actively present in lecture (45 minutes), and lab session (45 minutes). Lecture may take form of a Zoom meeting and/or video presentation and/or individual work on assignment. Lab session provides hints and solutions for homework assignments which will involve calculations, software implementation, and reading. Part of the assignments will ask students to annotate readings using perusall.com.
Course credit requirements
The credit for the exercise class will be awarded to the student who hands in satisfactory solutions to homework assignments (requiring 60% of total points) by the prescribed deadline. It is possible to skip up to 4 assignments and to provide satisfactory feedback (at least 10 relevant annotations) to readings instead. Homework will be assigned during lab sessions and will be due by the end of the week.
Exam and grade
Final project will be assigned during the course. Students can work in teams of size 2 or 3, multidisciplinary teams are preferred. Teams are welcome and encouraged to use their own data for the project in lieu of the data assigned to the class. In such a case, teams are expected to prepare written project proposal and submit it to the lecturer during the first month of the course. Final grade will be assigned during oral examination, which will take into account project (40%), homework assignments (40%, one HW with lowest grade is being dropped), and answers to follow-up question(s) (20%). Project report needs to be submitted at least 2 days before oral exam, one feedback is provided to projects sent to the instructor at least two weeks before the oral exam.
Course texts
Computational aspects of psychometric methods. With R. (Book in preparation)
Rao, C. R. & Sinharay S. (2006). Handbook of statistics. Volume 26: Psychometrics. Amsterdam, NL: Elsevier.
van der Linden, W. J. (2016). Handbook of item response theory: Models, statistical tools, and applications (Vols.1-3). Boca Raton, FL: Chapman & Hall/CRC.