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 July/August 2023 or February 2024.
Date: Monday 22 – Friday 26 July 2024
Time: 10:00 – 13:30 CEST*
*including a 30 minute break
Field experiments, also known as randomised control trials (RCTs) conducted outside of a controlled laboratory setting, offer several advantages over traditional laboratory experiments, particularly for social scientists. This course introduces you to the design and analysis of field experiments. You will learn about the unique utility of RCTs via discussion of both statistical theory as well as practical replications and applied research examples.
By the end of this course, you will be confident in the theory and practice of randomised control trials (RCTs) in the social sciences. This hands-on and practical course will ensure participants are able to effectively design an experimental intervention in the field and evaluate its effects, execute a randomisation strategy, carry out power calculations, and complete a comprehensive statistical analysis of RCT data. You will be equipped with a toolkit to help resolve complex design issues like block randomisation, as well as potential threats to inference that may emerge in the field such as attrition and spillover effects, among other advanced topics.
3 ECTS credits awarded for engaging fully in class activities.
1 additional ECTS credit awarded for completing a post-course assignment.
Stuart J. Turnbull-Dugarte is an Associate Professor in Quantitative Political Science at the University of Southampton (UK) and Director of PhD programmes in Politics. He is one of the founding members of the Centre for Behavioural Experimental Action and Research (C-BEAR). Stuart received his PhD in Political Science from the Department of Quantitative Political Economy at King’s College London. He has published extensively using experimental methods including randomised control trials carried out in collaboration with political organisations and political parties.
The course will begin by providing a foundational overview of the problem with establishing causality in the social world. We will look at examples of how (and how not) randomised control trials can overcome the problem of identifying causal relationships in the real world. You will be introduced to the potential outcomes framework as well as the theory and practice of estimating the average treatment effect.
We will consider how design considerations can improve and/or bias estimations in randomised control trials. We will look at issues related to i) statistical power, ii) blocking & clustering in randomisation, iii) covariate-adjustment, and iv) randomisation inference.
We will turn our attention to estimating treatment effect heterogeneity in RCTs. This will cover treatment-by-covariate and treatment-by-treatment interactions, as well as the difference in the causal interpretation between the two heterogeneous effect types.
You will learn about some of the pitfalls (and unique opportunities) that emerge from executing randomised control trials. We will consider solutions to issues such as non-compliance, attrition, post-treatment bias, as well as the differences and interpretations of diverse estimands of interest (e.g. ITT, CACE). You will also learn how to estimate spillover effects. Finally, we will discuss the practical issues and problems that emerge from working with non-academic research partners and how to manage these relationships.
The final day will be given over to discuss fellow participants’ proposed randomised control trials and how the tools learned throughout the week can be applied to improve deigns in advancing of going to the field. To that end, you will receive detailed feedback from the instructor on design proposals and we will engage in a group-based discussion of the practical issues and potential field-based pitfalls that may emerge.
Suggested reading
Gerber, A. & Green, D. (2012) Field Experiments: Design, Analysis, and Interpretation. W.W. Norton & Company: New York.
The course is structured into five live Zoom session. Each Zoom session will consist of two 1.5 hours of instruction interrupted by a 30-minute break. The first part of each session will focus on statistical theory and the latter part of each session will focus on applied examples and replications in R. During the break period, you will be encouraged to discuss their own research projects and interests.
Throughout the course, you will be provided with both methodological readings and example RCT papers. You are expected to come ready to learn, participate in applied and hands-on practice in R, and keen to engage in discussions of your own RCT plans.
In addition to the 3 hours of direct class time, you will be invited to participate in one-to-one dedicated office hours with the instructor.
Good knowledge of linear regression (OLS) is required. Basic understanding of including multiplicative interactions preferred. Some familiarity with R is essential.
As a participant in this course, you will engage in a variety of learning activities designed to deepen your understanding and mastery of the subject matter. While the cornerstone of your learning experience will be the daily live teaching sessions, which total three hours each day across the five days of the course, your learning commitment extends beyond these sessions.
Upon payment and registration for the course, you will gain access to our Learning Management System (LMS) approximately two weeks before the course start date. Here, you will have access to course materials such as pre-course readings. The time commitment required to familiarise yourself with the content and complete any pre-course tasks is estimated to be approximately 20 hours per week leading up to the start date.
During the course week, you are expected to dedicate approximately two-three hours per day to prepare and work on assignments.
Each course offers the opportunity to be awarded three ECTS credits. Should you wish to earn a 4th credit, you will need to complete a post-course assignment, which will involve approximately 25 hours of work.
This comprehensive approach ensures that you not only attend the live sessions but also engage deeply with the course material, participate actively, and complete assessments to solidify your learning.
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.