Categorical Data Analysis with R

Novice and advanced R users from all professional groups.

• CHF 600.- for members of UZH/ETH and associated institutes
• CHF 800.- for alumni of UZH/ETH, members of other universities, the public sector and non-profit organizations
• CHF 1200.- for companies

This course provides an introduction to categorical data analysis in R. The course involves both theoretical and practical sessions. The most important R packages for categorical data analysis are explained. While most of the applications come from the Social Sciences or the Humanities, both theory and practice are relevant to other scientific disciplines as well.

On the first day, the fundamental concepts of categorical data are introduced, including the standard tests for association/independence (e.g., Fisher exact, Pearson chi squared). Mosaic plots are introduced as a means to visualize the associations in a frequency table in a concise way. Log-linear analysis will be explained as the technique for modelling associations or (conditional) independence in multiway tables. The hierarchical principle in modelling will be explained, along with commonly used statistics for model selection such as the likelihood-ratio chi square (G squared), BIC and AIC. Extensions of the model to the analysis of ordinal variables will be discussed, for instance linear-by-linear association. The second day will continue with log-linear analysis and other generalized linear models for categorical data. Predictive models such as Poisson regression and logistic regression, but also negative binomial regression for over-dispersed data will be covered. When time allows, we will also introduce latent class analysis as the categorical alternative for structural equations models.

May 31 - June 1, 2018