Advanced Topics in Survey Experiments

Do you know the basics of survey and experimental research? This workshop is designed to take the next step in truly supercharging both your skills and your awareness of the state of the art. The survey experiments bootcamp introduces the most prevalent advanced topics in social and political science applications. After a quick review of randomized experimental and causal inference basics, we will get into design considerations like how to increase internal and external validity, and how to design valid experimental outcome measures. We will discuss the replication crisis and how to address the issues it exposes, such as taking statistical power seriously, and best practices for pre-registration or treatment administration. We will get into more complicated designs, like multiple treatments, interactions, factorial designs (like conjoints), and conditional or heterogeneous treatment effects. We will conclude with an examination of the increasing popularity of within-person designs and stimulus sampling approaches. If time allows, we will get into fun examples of experimental survey designs and causal inference with observational survey data. While the course is theoretical, I will show you how to effectively design implementations for the topics discussed with the help of artificial intelligence in any statistical programming language.

Pre-requisites:

Research mindset. Survey design and implementation basics. Randomized experimental basics. Multiple regression basics (including understanding of interactions). 90-second elevator pitch of your background and research ready at the beginning of the first session. No preparatory readings are required before the session. Readiness to work with full attentiveness during the day-long session. (This session will be fast-paced and dense.)

Date & Time:

The training will take place online on 26-November 2025, from 09:00 to 17:00. The course is primarily aimed at GIGA researchers (external participants can also register).

About the Trainers:

Levente Littvay (Levi) is Research Professor at ELTE Centre for Social Sciences, Hungarian Academy of Sciences Centre of Excellence and a Senior Visiting Researcher at the Democracy Institute of Central European University, where he also used to be Professor of Political Science (2007-2023). He holds a PhD in Political Science and an MS in Survey Research and Methodology from the University of Nebraska-Lincoln and taught numerous research methods workshops globally and online, including introductory, advanced, and multilevel regression and structural equation modelling, experiments, causal inference, impact evaluation, latent variable models, measurement theory, missing data, introductory stats, survey design, R, research design and AI in research courses. Founder of MethodsNET, and head of Team Survey in Team Populism where he helped spawn the Leader Profile Series and the New Populism series with The Guardian.