Module Description

This module introduces students to a selected set of advanced statistical methods that are commonly used in quantitative social research.

You will cover three advanced methods such as Bayesian analysis, multilevel modelling, and agent-based simulation. These advanced methods allow you to answer questions such as how does the social or political context influe nce how people behave? How can we combine results from previous research in a principled way? How can we identify causal effects?

To gain hands-on experience with answering these questions with social science data of varying complexity, you will apply these methods to existing small- and large-scale data sets. The expectation is that by the end of the module you will understand the basic principles of the advanced statistical methods covered, appreciate the context in which the methods are best applied, and have had practical experience of applying these methods to real-world data.

Module Objectives

To familiarise students with the key issues in the craft of applied work so that they become careful, considered and thoughtful researchers in quantitative social sciences.

Learning Outcomes

By the end of the course, students will:

• understand the basic principles of the advanced statistical methods covered;
• appreciate the context in which the methods are best applied, and;
• have had practical experience of applying these methods to real-world data, using R language and environment for statistical computing or equivalent statistical software.

Structure

1 hour lecture + 2 hour computer laboratory.

Assessment

The assessments consists of 2 technical reports of 2,000 words.

Illustrative Bibliography

• Faraway, J. (2016) Extending the Linear Model with R. Chapman & Hall/CRC.
• Gelman, A., ed. (2009) A Quantitative Tour of the Social Sciences. Cambridge University Press.
• Gelman, A. and Hill, J. (2007) Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press.
• Morgan, S.L. and Winship, C. (2014) Counterfactuals and Causal Inferences. Cambridge University Press.
• Snijders, T.A.B. and Bosker, R.J. (2012) Multilevel Analysis. Sage.

Indicative Syllabus

• Week 1: Introduction to the module, data and software.
• Weeks 2-3: Generalised linear models.
• Weeks 4-5: Multilevel models.
• Week 6: Reading week.
• Weeks 7-8: Causal inference.
• Weeks 9-10: Revision and data analysis workshops.