ESRA logo
Tuesday 14th July      Wednesday 15th July      Thursday 16th July      Friday 17th July     

Wednesday 15th July, 16:00 - 17:30 Room: L-102

Natural Experiments in Survey Research

Convenor Professor Henning Best (University of Wuerzburg )
Coordinator 1Dr Gerrit Bauer (University of Munich )

Session Details

In this session we are particularly interested in papers on identification of treatment effects in natural experiments, research combining surveys with natural-experimental designs, papers that employ multiple methods of treatment estimation, and innovative ways to design or analyze natural experiments in cross-sectional and especially panel surveys.

Though experiments are generally regarded as the royal road to causal inference, natural experiments often face serious problems: endogeneity, insufficiencies in standardizing treatment- and control conditions, and self-selection into study- and control group. Advances in data analysis have tackled these problems, and methods such as IV-regression, conditional fixed-effects models and propensity score matching help in identifying unbiased treatment effects.

We are not only interested in applications of natural experiments in the social sciences but especially encourage submissions on methods and designs. Examples of such methods include applications of IV-regression, conditional fixed-effects models, propensity score matching and regression discontinuity designs.

Paper Details

1. Natural Experiments in the Social Sciences: Overview, Typology & Evaluation Criteria
Dr Gerrit Bauer (LMU Munich)
Professor Henning Best (University of Wuerzburg)

An increasing number of papers published in sociology and political science make use of the research design “natural experiment”, indicating that not a researcher but nature sorts individuals into treatment- and control conditions. Some of these studies rely on real randomization (e.g. lottery studies or weather shocks), while other studies rely on “as-if-randomization” (e.g. policy interventions restricted to one particular group). In our presentation, we provide examples for different types of natural experiments and summarize how they make use of exogenous (“natural”) variation when searching for causal effects. Following a typology discussed by Dunning (2012), we

2. The short term effect of the escalation of 2008-crisis on the relation between social vulnerability and welfare and labour market threats. A natural experiment approach.
Dr Marie Valentova (LISER Luxembourg)
Miss Marie-sophie Callens (LISER)

This paper uses the European Value Study from 2008/2009 to investigate the short term effect of the escalation of 2008-crisis on the relation between social vulnerability and the perception of immigration related threats (labour market and welfare threat). We use a natural experiment approach comparing two groups of Luxembourg residents to evaluate the impact of the crisis: respondents interviewed before (control group) and after (treated group) the escalation of the crisis employing propensity score matching technique. Due to the used research design (natural experiment with survey data), this paper estimates the causal effect of the crisis.