Previous Programs

1st Spring School
1st_springschool

The 1st Spring School, hosted by the Institute for School Development Research at TU Dortmund University, Germany, was on the topic of Combining Categorical and Continuous Latent Variable Modeling with Mplus. Empirical educational researchers frequently make a polar distinction between categorical variables (e.g. for cluster analysis) and continuous variables (e.g. for factor analyses). However, such a distinction is neither completely accurate nor is it useful since many phenomena fall under both categories at the same time, and data analysis strategies should take this into account. Recent developments in latent variable modeling and so-called hybrid or mixture modeling provide an appropriate approach in order to deal with continuous and categorical constructs at the same time. Katherine E. Masyn (Harvard University) delivered workshops on models which combine of continuous and categorical latent variables and hands-on training with Mplus in computer labs.

bmbf-logo-englisch

Year 2011
Topic Combining Categorical and Continuous Latent Variables Modeling with Mplus
Main Lecturer(s) Katherine E. Masyn (Harvard University)
Host Institute for School Development Research, Dortmund University, Germany
2nd Spring School
2nd_springschool

The 2nd Spring School on Advanced Methods in Educational Research has been conducted at the University of Gothenburg. The topic was Methods for Causal Inference from Observational Data. In educational research, experiments cannot be conducted for practical or ethical reasons. However, during the last decades powerful alternatives to the experimental ideal have been developed to draw causal inference from observational data. Such methods include fixed-effect models, differences-in-differences approaches, instrumental variables and regression discontinuity analyses. The spring introduced such methods to promising young researchers. Ludger Woessmann and Guido Schwerdt (ifo Institute, Munich) delivered workshops. Further trainings included computer laps in R.

Year 2012
Topic Methods for Causal Inference from Observational Data
Main Lecturer(s) Ludger Woessmann and Guido Schwerdt (ifo Institute, Munich)
Host University of Gothenburg, Sweden
3rd Spring School
3rd_springschool

The 3rd Spring School, hosted by the Institute for School Development Research at TU Dortmund University, Germany, was on the topic of Multilevel Models: The Next Generation. The lecturer Joop Hox (Utrecht University) introduced Multilevel Structure Equation Modeling and Mplus to the participants, and discussed similarities and differences of Multilevel regression and Multilevel Structure Equation. The analysis of longitudinal data by multilevel regression and latent growth approaches was another topic. Such techniques are very useful to analyze data from longitudinal studies. The Spring School ended with a lecture about multilevel non-normal data.

bmbf-logo-englisch

Year 2013
Topic Multilevel Models: The Next Generation
Main Lecturer(s) Joop Hox (Utrecht University)
Host Institute for School Development Research, Dortmund University, Germany
4th Spring School
4th_springschool The 4th Spring School, hosted by the Centre for Educational Measurement at the University of Oslo, Norway, was on the topics Propensity Score Matching, Instrumental Variables, and Design Issues in studies on causal analysis. The first lecturer, Daniel Caro (University of Oxford), introduced the software R, later giving insight into issues in the analysis of international assessment data. The next four days, Elisabeth A. Stuart (Johns Hopkins University, Baltimore) addressed the framework of causal inference with a focus on non-experimental study designs. Under this framework, Propensity Score Matching methods were discussed in detail and exemplified. Thereafter, Instrumental Variables and its applications into the framework of causal inference was the topic. All workshops covered hands-on experience and computer labs.
Year 2014
Topic Propensity Score Matching, Instrumental Variables, and Design Issues
Main Lecturer(s) Elisabeth A. Stuart (Johns Hopkins University Baltimore), Daniel Caro (Oxford University)
Host Centre for Educational Measurement at University of Oslo, Norway
5th Spring School
The 5th Spring School,  hosted by the Institute for School Development Research at TU Dortmund University, Germany, was on Measurement Theory and Models. The first lecturers, Mark Wilson and Karen Draney (UC Berkeley), reviewed foundations and introduced advances in Item Response Theory (IRT). The next two and a half days, Richard Shavelson (Stanford University) introduced Generalizability Theory (G Theory) to the participants. All workshops on IRT and G-Theory covered hands-on experience and computer labs.

bmbf-logo-englisch

Year 2015
Topic Measurement Theory and Models
Main Lecturer(s) Karen Draney & Mark Wilson (Berkeley University), Richard Shavelson (Stanford University)
Host Institute for School Development Research, Dortmund University, Germany
6th Spring School
The 6th EERA Spring School on Advanced Methods in Educational Research,  hosted by the Centre for Educational Measurement at the University of Oslo, Norway, was on Longitudinal Structural Equation Modeling. In the week-long intensive course, the lecturer Todd D. Little (Texas Tech University) reviewed both foundations of and recent advances in longitudinal research. He covered issues of design, issues of measurement, and issues of analysis. He discussed the three primary models for longitudinal data including the panel model, the growth model, and the intra-individual p-technique model. Students learned when each model is preferred and how to interpret the parameters of each model.

 

Year 2016
Topic Longitudinal Structural Equation Modeling
Main Lecturer(s) Todd D. Little (Texas Tech University)
Host Centre for Educational Measurement at University of Oslo, Norway