The University of Southampton
Courses

# STAT3004 Multivariate Data Analysis

## Module Overview

This module will introduce some common methods for analysing data involving two or more variables per observation. In this respect, the module covers three topics: linear regression, logistic regression, and principal components analysis.

### Aims and Objectives

#### Module Aims

• Demonstrate knowledge and understanding of the basic ideas behind several common statistical techniques for analysing multivariate data (linear regression analysis, logistic regression analysis and principal component analysis) • Analyse real data by applying these techniques using SPSS and interpret the resulting output • Write short statistical reports based on these analyses

#### Learning Outcomes

##### Learning Outcomes

Having successfully completed this module you will be able to:

• Demonstrate knowledge and understanding of the basic ideas behind several common statistical techniques for analysing multivariate data (linear regression analysis, logistic regression analysis and principal component analysis
• Analyse real data by applying these techniques using SPSS and interpret the resulting output
• Write short statistical reports based on these analyses

### Syllabus

The module is split into three parts: Part 1: Linear regression: simple linear regression, least squares estimation, interpreting regression coefficients, checking the regression assumptions, hypothesis testing, confidence intervals, multiple linear regression, model selection, handling of categorical explanatory variables, interactions, variable transformations. Part 2: Logistic regression: modelling dichotomous outcome variables, probabilities and odds, the logistic regression model, how good is the model, model interpretation, model selection, diagnostic methods, grouped data. Part 3: Principal component analysis: definition, construction of principal components from the covariance matrix, principal components based on the correlation matrix, idea of dimension reduction, steps in a principal components analysis.

### Learning and Teaching

TypeHours
Independent Study120
Teaching30
Total study time150

### Assessment

#### Summative

MethodPercentage contribution
Coursework 30%
Exam  (2 hours) 70%

#### Referral

MethodPercentage contribution
Coursework assignment(s) 30%
Exam 70%

Pre-requisites - Any equivalent modules can be accepted with permission from the Programme coordinator

#### Pre-requisites

To study this module, you will need to have studied the following module(s):

CodeModule
STAT2009Research Methods in The Social Sciences