Module Directory

This module presents quantitative methods essential to test hypotheses in political science. After introducing the statistical computing environment R and associated document preparation software tools as well as bivariate hypothesis testing, linear regression using ordinary least squares estimation is covered in depth as the workhorse model for statistical inference in political science. The second half of the module will cover extensions for temporal and multilevel data and introduce methods for causal inference.

All models and methods are approached substantively, mathematically, and computationally (using R), with applications to political science research questions. Throughout the module, students will also familiarise themselves with the interpretation and presentation of empirical evidence in political science. The module will be particularly useful for students who aim to pursue careers in academia or in research-intensive environments, for example think tanks, research-related government posts, data science, or survey analytics.

The module will enable students to:

• understand and apply the logic of hypothesis testing in a variety of political science contexts.


• understand and interpret statistical analyses in published political science research.


• master the mathematics behind ordinary least squares and related regression models.


• translate theories into empirical models.


• conduct their own basic regression analyses using empirical datasets, both manually and with software, commensurate with analyses published in political science journals.


• assess the goodness of fit of empirical models.


• estimate causal effects of treatment variables on outcomes.


• effectively present quantitative results using R and modern document formats embedded in statistical software.

By the end of this module, students will be expected to be able to:

1. formulate theories in ways that are amenable to single and multiple hypothesis testing and be able to diagnose violations of basic assumptions.


2. understand, and be able to improve upon, statistical analyses and their interpretations in political science journals.


3. have practical experience with conducting high-quality quantitative political science research as well as with the implementation of basic regression models, both using ready-made functions/packages in R and manually/from scratch.


4. master the mathematics and statistical theory underlying hypothesis testing, ordinary least squares, time series analysis, panel and multilevel models, and causal inference techniques.


5. know how to handle complex data structures and implement appropriate models, including temporal and hierarchical dependence.


6. confidently apply causal inference techniques to estimate treatment effects.


7. present statistical results effectively.

Indicative Syllabus:

• Introduction to Quantitative Methods and R


• Hypothesis testing


• Linear regression I: Linear regression with one/two regressors


• Linear regression II: Ordinary least squares with multiple regressors


• Linear regression III: Assumptions of linear regression


• Methods for panel and multilevel data


• Interactions in OLS


• Causal inference I: Experimental design


• Causal inference II: Difference in difference and regression discontinuity


• Causal inference III: Matching, instrumental variables, and synthetic controls

The module will be delivered over 2 hours per week.