Cost and booking

This is a paid course costing £240. Please register your interest via EventBrite, and we will be in touch with details of how to book, and bursaries for PhD students.

Overview

This course uses the software package R in conjunction with the R-commander and the R-studio. R is an open-source programming language and software environment for performing statistical calculations and creating data visualisations.

Led by Dr Graeme Hutcheson, we will cover a relatively complete and practical introduction to data analysis in R. This includes:

• analyses for many different types of data using generalized regression techniques.

We will consider how to:

• define models, select an analytical technique and interpret the results using graphical and tabular output.

By the end of the course you will be able to perform contrast coding for categorical variables, interpreting and visualising interactions, regression diagnostics and data transformation and issues related to multicollinearity and variable selection.

Prerequisites

• No previous experience of using R is required for this course, nor is any previous experience of coding or using other statistical packages. However, it will be of most use to those who are interested in modelling social science datasets (survey andquasi-experimental) and applying graphics to interpret these.

• Participants are provided with instructions and examples for installing the software onto their own computers before the course starts.

Course timetable

Monday 05 July

• Morning: Introduction to the course and the software. Defining research questions.

• Afternoon: Selecting technique and interpreting output. Open discussion session (optional).

Tuesday 06 July

• Morning: A Regression for continuous data. Explanatory variables interactions and categorical variables.
• Afternoon: Poisson regression for count data. Open discussion session (optional).

Wednesday 07 July

• DAY OFF

Thursday 08 July

• Morning: Logistic regression for binary data. Multinomial logit models for unordered categorical data.
• Afternoon: Proportional-odds logit models for ordered categorical data. Open discussion session (optional).

Friday 09 July

• Morning: Regression diagnostics and data transformation. Regression diagnostics exercise.
• Afternoon: Variable selection andmulti-model inference. Open discussion session (optional).

Recommended reading

Agresti, A. (1996). An Introduction to Categorical Data Analysis. Wiley.

Fox, J. and Weisberg, S. (2011). An R companion to Applied Regression (second edition). SagePublications.

Harrell, F. E. (2001). Regression modelling strategies. Springer.

Hutcheson, G. & Sofroniou, N. (1999). The multivariate social scientist. Sage Publications.

Hutcheson, G. & Moutinho, L. (2008). Statistical modelling for management. Sage Publications.