AQM: An Introduction to Latent Variable Modelling in R

Advanced Quantitative Methods (AQM) Training

Latent variable modelling at its most basic involves taking original variables from data and transforming them in some way to give new variables with useful properties. Sometimes this can be used to create uncorrelated variables that can allow dimension reduction (like principal component analysis). Other times the new variables are chosen to best explain the relationships between original variables in our data and to represent underlying, unobservable constructs (e.g. intelligence or political opinion) as is the case with factor analysis.

These methods will be introduced in general and then students will be given the chance to learn how to apply them to data in R using a variety of commands and packages. If there is time, we may discuss some basic item response theory models as well.

Course Leader: Nema Dean

Nema graduated from the Department of Mathematics in Trinity College Dublin, Ireland, with a BA (Hons) and Gold Medal in Mathematics in 2002. In summer 2006 she received her PhD. in Statistics from the Department of Statistics in the University of Washington, Seattle, USA. Nema currently lectures in the Department of Statistics in the University of Glasgow.

Course prerequisites:

Some working knowledge of R is essential for this course (e.g. using basic commands, plotting, etc).

Students are responsible for arranging travel to and from this AQM training session. The WRDTP cannot reimburse travel costs in this instance.

This is an advanced Quantitative Methods (AQM) training event, open to ESRC and non-ESRC students aligned with any of the WRDTP Pathways.