Introduction to linear models in R

Dr. Franklin Chang (School of Psychology, University of Liverpool)

June 25th and 26th (6 hrs per day)

Location: Keynes College, University of Kent, Canterbury, CT2 7NP

This workshop will introduce you to the most commonly used statistical methods (anova, regression) using the R language. It is designed for those who have taken stats before (e.g., UK Psychology undergraduate level) and want to have a deeper understanding of these methods as well as be able to do their analyses in R. In particular, newer methods such as logistic mixed effects models will discussed. Examples will include data from eye-tracking, experimental designs with factors with three or more levels, and categorical dependent variables.

Here is a list of topics:

R basics, t-tests, correlations, making graphs (e.g., multi-panel, heatmaps, fitted regression lines)

anova (contrasts, posthocs), ancova,

regression (linear, logistics, polynomial),

linear model assumptions,

coding variables (dummy, effect coding, residualization, centering)

power analysis using simulation

mixed effects models, model simplification (convergence) and model comparison