Dozent: Ronny Scherer (Centre for Educational Measurement at the University of Oslo)

Datum: March 22-23, 2022 (Time: 4 am – 9 am EST/ Time: 9 am – 2 pm CET)

Ort: Virtual (Zoom)

Abstract: Structural equation modeling (SEM) represents a statistical approach to disentangle the relationships among latent and/or manifest variables, across groups, over time, and at different analytical levels. The potential of SEM has been recognized in many areas, including educational sciences, sociology, psychology, and business. This workshop provides an introduction to the principles and procedures of basic and more advanced SEM in the context of the international large-scale assessment PIAAC. Specifically, the following topics are covered: (a) Principles of structural equation modeling (model specification, identification, estimation, and evaluation), (b) Measurement models and confirmatory factor analysis, (c) Measurement invariance testing with few and many groups (including multi-group CFA, multilevel CFA, and the alignment method), and (d) Structural regression and indirect effects models (including multi-group and multilevel SEM). Participants can also present their own research or research ideas using PIAAC data and receive feedback on how to improve the analysis (optional).

Daten: PIAAC Public Use Files

Software: SPSS and Mplus

Agenda: Agenda (266 kB)

Literaturempfehlung: Maehler, D. & Rammstedt, B. (2020). Large-scale cognitive assessment: Analyzing PIAAC data. Series: Methodology of Educational Measurement and Assessment (MEMA). Springer: Cham. https://link.springer.com/book/10.1007/978-3-030-47515-4

Dozenten: Paul Bailey¹, Ting Zhang¹, Saida Mamedova¹, Emily Pawlowski¹, Emmanuel Sikali², Michael Lee¹, Eric Buehler¹, & Sinan Yavuz¹ (¹American Institutes for Research; ²National Center for Education Statistics)

Datum: March 22, 2022 (Time: 10 am – 2 pm EST/ 3 pm – 7 pm CET)

Ort: Virtual (Zoom)

Abstract: This course will provide an overview of the PIAAC study and guidance in data analysis strategies, including the selection and use of appropriate plausible values, sampling weights, and variance estimation procedures. The course will train participants in the analysis of PIAAC data files using the R package EdSurvey, which was developed specifically to analyze large-scale assessment data with complex psychometric and sampling designs. Participants will learn how to

• perform data processing and manipulation,
• produce descriptive statistics;
• generate cross-tabulations and plausible value means; and
• perform linear and logistic regressions.

This course is designed for researchers and policy analysts across various sectors and organizations who are interested in learning how to analyze PIAAC data. Participants should have at least a basic knowledge of R software (e.g., have taken entry-level training in R programming), as well as of statistical techniques, including statistical inference and multiple regression. A working knowledge of plausible values and sampling theory would be helpful but is not required. Participants should bring a computer preloaded with the latest version of the R and RStudio software so that they can practice the analytical techniques covered in the lesson plan.

Daten: PIAAC Public Use Files

Software: R

Agenda: Agenda (145 kB)

Literaturempfehlung: Maehler, D. & Rammstedt, B. (2020). Large-scale cognitive assessment: Analyzing PIAAC data. Series: Methodology of Educational Measurement and Assessment (MEMA). Springer: Cham. https://link.springer.com/book/10.1007/978-3-030-47515-4