Introductory description

The module presents an overview of some regression-based methods widely used in political science. The module emphasises modules where the assumptions of ordinary least-squares regression are violated, primarily because the dependent variable is non-continuous such as counts, categories, or durations.

Module aims

Outline syllabus

This is an indicative module outline only to give an indication of the sort of topics that may be covered. Actual sessions held may differ.

1. Introduction to probability models
2. Estimation
3. Generalised linear models
4. Binary choice models
5. Multinomial choice models
6. Reading week
7. Ordered choice models
8. Event count or duration models I
9. Event count or duration models II
10. Model selection and robustness

Learning outcomes

By the end of the module, students should be able to:

• Understand how maximum likelihood models can be used to answer research questions.
• Identify which parametric model best fits their research question and the underlying data-generating process.
• Interpret a variety of maximum likelihood models and present the results using prediction, substantive interpretation, marginal effects, and simulation.
• Apply maximum likelihood estimation to their own research.

Indicative reading list

Reading lists can be found in Talis

Research element

Students will be conducting their own research using the techniques learned in the module and write it up in their technical report.

International

Examples will be taken from across the world.

Subject specific skills

• Have an increased understanding of the technical and theoretical/conceptual dimensions of maximum likelihood estimation in quantitative data analysis.
• Develop substantial competency in identifying the statistical modules suited to the nature data and in using these models to answer a research question.
• Understand the value and gain practical experience in applying maximum likelihood estimation to political science data sets.
• Present and interpret the results of quantitative statistical analyses appropriately when using maximum likelihood estimation to fit models to data.

Transferable skills

• Understand the impact of the ways in which quantitative data are generated, manipulated and analysed on the validity and usefulness of research findings.
• Manipulate and analyse existing data and present the results of these analyses appropriately using statistical software
• Produce original empirical research and develop skills in written communication.
• Interpret and critique published quantitative research more accurately.