Install the app
Install this application on your home screen for quick and easy access when you’re on the go.
Just tap then “Add to Home Screen”
Install this application on your home screen for quick and easy access when you’re on the go.
Just tap then “Add to Home Screen”
Menu
Install this application on your home screen for quick and easy access when you’re on the go.
Just tap then “Add to Home Screen”
Member rate £492.50
Non-Member rate £985.00
Save £45 Loyalty discount applied automatically*
Save 5% on each additional course booked
*If you attended our Methods School in the July/August 2023 or February 2024.
Date: Monday 29 July – Friday 2 August 2024
Time: 09:00 – 12:00 CEST
This introductory course on causal inference techniques will teach you state-of-the-art tools for establishing causal relations in the social sciences.
Emphasising intuition, the objective of the course is to teach you how to evaluate the effects of policies, events, and other interventions in empirical research and apply these to your research questions. You will learn popular methods of causal inference through both a theoretical and an applied component. We will review the theory behind each of these designs in detail with the aim of comprehension and competency. There will be an emphasis on applicability of the learned methods by discussing how they have been used in previous research, and an applied session in R for each method. This course prepares you for implementing causal inference in your own research and provides a strong foundation for further study.
The purpose of this course is to teach you to establish the causes and effects of changes in the real world. You will gain the ability to differentiate correlation from causality by learning the five widely used causal research methodologies: randomised experiments, multiple regression and matching, instrumental variables and 2SLS design, difference-in-differences and regression discontinuity design. The aim is for you, at the end of the course, is to be able to apply these methods to your own research interests and conduct the analysis independently.
Upon completing this course, you will be able to:
3 ECTS credits awarded for engaging fully in class activities.
1 additional ECTS credit awarded for completing a post-course assignment.
Ria is an Assistant Professor at the University of Zagreb and Associate Researcher at the London School of Economics and Political Science.
Until recently, she was a lecturer at the University of Oxford where she taught a graduate course in Causal Inference.
She has a PhD in Economics from the Department of Political Economy at King's College London. Her research has been published or is forthcoming in journals such as the Journal of Public Economics, Journal of Politics, Political Behaviour, Labour Economics, Journal of Law and Economics, and Journal of Empirical Legal Studies.
To start, we will delve into the challenges of establishing causality in the social sciences. You will gain an understanding of the fundamental problem of causal inference and the potential outcomes framework, which provide a solid foundation for making causal claims in research. Through detailed discussions and examples, you will also learn about randomised experiments — how they work, why they are important, and how they can help establish causal relations in social science research.
As running randomised experiments is not always feasible and practical, we will discuss designs that assume that selection into the treatment groups is based on observables and how we can use multiple regression to overcome endogeneity. You will learn the potential and pitfalls of multiple regression and matching techniques used as causal methodology.
You will learn about instrumental variables (IVs) on the third day. Discussing existing research that has used this method, we will explore the assumptions underlying the use of IVs, how to estimate them using 2SLS, and ways to increase the interval validity of this method. We will also discuss ideas on how to find good IVs, and the appropriate model specifications.
You will learn how when public policies are adopted at a certain point in time for a selected group, this can be used to retrieve causal estimates utilising difference-in-differences (DiD). We will differentiate panel data from cross-sectional data. Further, we will discuss the DiD assumptions, how to verify them empirically, how to estimate DiD, and replicate previous work that uses this method. We will also briefly discuss the recent explosion of literature using this topic and some thoughts on how to navigate it.
In the last session, you will be introduced to regression discontinuity designs (RDDs) that allow us to approximate an experimental setting when a certain treatment or policy is based on a threshold. In the theoretical component, you will learn the required RDD assumptions, how to test their validity, and the model specifications used to estimate RDDs. We will replicate previous work that uses this method and discuss fuzzy RDDs and how they connect to instrumental variables. Finally, we will discuss how the methods we learnt over this week could be applied to the research questions of the participants.
The course is structured into five live Zoom sessions, each lasting 3 hours each day. Each session will cover a theoretical component introducing the topic of the session through examples of published research. The second part of each session will be an applied component using RStudio Cloud where we will implement the given methodology on relevant datasets, replicate previous work and interpret software output for each method we discuss.
The course provides a safe and collaborative environment for discussing students' work and published research. Our ultimate goal is to make the course useful, and the final session will focus on specific topics of interest to each student. You will additionally be provided with methodological and applied readings to deepen your understanding of the methods we learn.
There will also be designated ‘office hours’, during which you can sign up for a one-to-one consultation.
You should have a good understanding of basic statistics (linear regression, interaction effects, standard errors and hypothesis testing) and some introductory knowledge of R using RStudio.
This course description may be subject to subsequent adaptations (e.g. taking into account new developments in the field, participant demands, group size, etc.). Registered participants will be informed at the time of change.
By registering for this course, you confirm that you possess the knowledge required to follow it. The instructor will not teach these prerequisite items. If in doubt, please contact us before registering.