48th GESIS Spring Seminar (2019)

Bayesian Modelling in the Social Sciences

 

March 11 - March 29, 2016, Cologne, Germany

 

The GESIS Spring Seminar comprises three training courses for social scientists interested in advanced techniques of data analysis and in the application of these techniques to data. Each course comprises lectures and exercises using personal computers. While in the lectures the logic of models and the corresponding analysis strategies are explained, during the exercises  participants are given the opportunity to apply these methods to data. Different types of data are investigated in each course. The courses can be booked either separately or as a block.

Each year, the focus of the Spring Seminar is on another key theme. In 2019, it was “Bayesian Modelling in the Social Sciences”.

March 11 - 15, 2019

Lecturers:

Prof. Dr. Susumu Shikano

Dr. Taehee Kim

 

 

Short Course Description:

Social scientists increasingly apply the Bayesian approach to diverse kinds of research topics. To motivate further political scientists to use this approach, this course provides participants the following three points: First, the course provides a conceptual background for Bayesian inference. Second, participants will be guided how to read the literature using Bayesian statistics and interpret the results. Third, this course introduces to a software for Bayesian analysis with political science examples. The course consists of lectures (morning) and lab sessions (afternoon). The lecture deals with relevant background knowledge as well as specific skills for Bayesian analysis. In lab sessions, these skills are applied to political and social science data. Hence, course participants also learn the basic knowledge of JAGS, which is needed to conduct Bayesian estimation.

March 18 - 22, 2019

Lecturers:

Prof. Dr. Rens van de Schoot

Dr. Milica Miočević

 

Short Course Description:

During this course students will be introduced to philosophical underpinnings of Bayesian statistics and will learn how to fit regression, mediation, CFA, and longitudinal growth models in the Bayesian framework. Students will learn the steps in conducting Bayesian analyses and will be able to understand articles that examine and apply Bayesian SEM. The course is highly interactive, and the afternoons will be dedicated to implementing and practicing the material using the participant’s software of choice (Mplus or R in tandem with JAGS or stan). We highly recommend bringing your own data for Day 5 of the course; however, the instructors will have example data sets for participants who do not have their own data.

March 25 – March 29, 2016

Lecturers:

Dr. Mark Andrews

Dr. Jens Roeser

 

 

Short Course Description:

This course provides a general introduction to Bayesian multilevel modelling. Throughout, we will extensively use R and the Bayesian probabilistic programming language Stan and its R based brms interface. The course begins by providing a solid introduction to all the fundamental principles and concepts of Bayesian data analysis: the likelihood function, prior distributions, posterior distributions, high posterior density intervals, posterior predictive distributions, marginal likelihoods, Bayes factors, etc. We will do this using some simple models that are easy to understand and easy to work with. We then turn to providing a solid introduction to multilevel models, again beginning with conceptually and computationally simple multilevel models. We then proceed to more practically useful Bayesian multilevel models, and ones that necessarily require Monte Carlo methods, particularly multilevel general and generalized linear regression models. For these models and analyses, we will make extensive use of the R based brms interface to Stan, which allows general Bayesian multilevel regression to be done remarkably easily. As part of the introduction to multilevel regression, we will ensure that we first have a solid understanding of the (non-multilevel) general and generalized linear models. In final part of the course will cover multilevel models that are not regression models per se. In particular, we will explore multilevel probabilistic mixture models, especially Latent Dirichlet Allocation and the Hierarchical Dirichlet Process model, which have been shown to be particularly valuable in the modelling of text data. Throughout the course, we will have interludes where we address some major general issues in Bayesian data analysis, particularly Markov Chain Monte Carlo methods and Bayesian model evaluation (e.g., using cross-validation, WAIC, and Bayes factors, etc).