Dr Jouni Kuha COL8.04

This course is compulsory on the MPhil/PhD in Philosophy of the Social Sciences. This course is available on the MPhil/PhD in Data, Networks and Society, MPhil/PhD in Environmental Economics, MPhil/PhD in International Relations, MPhil/PhD in Media and Communications, MPhil/PhD in Social Policy, MPhil/PhD in Social Research Methods, MRes/PhD in Management (Marketing) and MRes/PhD in Management (Organisational Behaviour). This course is available as an outside option to students on other programmes where regulations permit.

The course is available to all research students.

The course will assume a knowledge of standard linear regression models, to the level covered in MY452/MY552 (Applied Regression Analysis)

An introduction to the application of modern multivariate methods used in the social sciences, with particular focus on latent variable models for continuous observed variables, and their application to questions of measurement in the social sciences. At least the following topics will be covered: principal components analysis, exploratory factor analysis, confirmatory factor analysis and structural equation models. In addition, a selection from the following topics will be covered: cluster analysis, correspondence analysis, multidimensional scaling, latent class models, latent trait models.

20 hours of lectures and 8 hours of computer workshops in the LT.

Exercises from the computer classes can be submitted for marking.

D J Bartholomew, F Steele, I Moustaki ; J I Galbraith (2008) Analysis of Multivariate Social Science Data (Second Edition).

Research project (100%).

Research paper 25-30 pages, including tables and figures, or approximately 3,000-4,000 words, answering a set of questions by applying multivariate analyses to some data (100%).