Natural Experiments Based on Geography

2015 ◽  
Vol 4 (1) ◽  
pp. 65-95 ◽  
Author(s):  
Luke Keele ◽  
Rocío Titiunik
Keyword(s):  
Natural Experiment ◽  
Original Study ◽  
Causal Effects ◽  
Natural Experiments ◽  
Ballot Initiatives ◽  
Causal Inferences ◽  
And Control ◽  
Administrative Boundary

Political scientists often attempt to exploit natural experiments to estimate causal effects. We explore how variation in geography can be exploited as a natural experiment and review several assumptions under which geographic natural experiments yield valid causal estimates. In particular, we focus on cases where a geographic or administrative boundary splits units into treated and control areas. The different identification assumptions we consider suggest testable implications, which we use to establish their plausibility. Our methods are illustrated with an original study of whether ballot initiatives increase turnout in Wisconsin and Ohio, which illustrates the strengths and weaknesses of causal inferences based on geographic natural experiments.

scholarly journals Geographic Boundaries as Regression Discontinuities

2015 ◽  
Vol 23 (1) ◽  
pp. 127-155 ◽  
Author(s):  
Luke J. Keele ◽  
Rocío Titiunik
Keyword(s):  
Voter Turnout ◽  
Regression Discontinuity ◽  
Regression Discontinuity Design ◽  
Natural Experiments ◽  
Causal Inferences ◽  
Political Advertisements ◽  
Geographic Boundaries ◽  
Random Fashion ◽  
And Control ◽  
Administrative Boundary

Political scientists often turn to natural experiments to draw causal inferences with observational data. Recently, the regression discontinuity design (RD) has become a popular type of natural experiment due to its relatively weak assumptions. We study a special type of regression discontinuity design where the discontinuity in treatment assignment is geographic. In this design, which we call the Geographic Regression Discontinuity (GRD) design, a geographic or administrative boundary splits units into treated and control areas, and analysts make the case that the division into treated and control areas occurs in an as-if random fashion. We show how this design is equivalent to a standard RD with two running variables, but we also clarify several methodological differences that arise in geographical contexts. We also offer a method for estimation of geographically located treatment effects that can also be used to validate the identification assumptions using observable pretreatment characteristics. We illustrate our methodological framework with a re-examination of the effects of political advertisements on voter turnout during a presidential campaign, exploiting the exogenous variation in the volume of presidential ads that is created by media market boundaries.

scholarly journals A Tutorial on the Use of Differences-in-Differences in Management, Finance, and Accounting

2021 ◽  
Vol 25 (1) ◽  
Author(s):  
Rafael Felipe Schiozer ◽  
Frederico Abou Mourad ◽  
Theo Cotrim Martins
Keyword(s):  
Financial Crisis ◽  
Natural Experiment ◽  
Management Research ◽  
Natural Experiments ◽  
Causal Inferences ◽  
2008 Financial Crisis ◽  
Treatment Assignment ◽  
Practical Guide ◽  
Degree Of Risk ◽  
The 2008 Financial Crisis

ABSTRACT Context: natural experiments or quasi-experiments have become quite popular in management research. The differences-in-differences (DiD) estimator is possibly the workhorse of these techniques. Objective: the goal of this paper is to provide a tutorial that serves as practical guide for researchers considering using natural experiments to make causal inferences. Methods: we discuss the DiD advantages, concerns, and tests of validity. We also provide an application of the technique, in which we discuss the effect of government guarantees on banks’ degree of risk, using the 2008 financial crisis as a natural experiment. The database used, as well as the Stata and the R scripts containing the analyses, are available as online appendices. Conclusion: DiD may be used to tackle endogeneity concerns when treatment assignment is random.

scholarly journals Do natural experiments have an important future in the study of mental disorders?

2019 ◽  
Vol 49 (07) ◽  
pp. 1079-1088 ◽  
Author(s):  
Anita Thapar ◽  
Michael Rutter
Keyword(s):  
Cigarette Smoking ◽  
Conduct Problems ◽  
Natural Experiment ◽  
Causal Effect ◽  
Natural Experiments ◽  
Causal Inferences ◽  
Hyperactivity Disorder ◽  
Naturally Occurring ◽  
Being Bullied ◽  
Hostile Parenting

AbstractThere is an enormous interest in identifying the causes of psychiatric disorders but there are considerable challenges in identifying which risks are genuinely causal. Traditionally risk factors have been inferred from observational designs. However, association with psychiatric outcome does not equate to causation. There are a number of threats that clinicians and researchers face in making causal inferences from traditional observational designs because adversities or exposures are not randomly allocated to individuals. Natural experiments provide an alternative strategy to randomized controlled trials as they take advantage of situations whereby links between exposure and other variables are separated by naturally occurring events or situations. In this review, we describe a growing range of different types of natural experiment and highlight that there is a greater confidence about findings where there is a convergence of findings across different designs. For example, exposure to hostile parenting is consistently found to be associated with conduct problems using different natural experiment designs providing support for this being a causal risk factor. Different genetically informative designs have repeatedly found that exposure to negative life events and being bullied are linked to later depression. However, for exposure to prenatal cigarette smoking, while findings from natural experiment designs are consistent with a causal effect on offspring lower birth weight, they do not support the hypothesis that intra-uterine cigarette smoking has a causal effect on attention-deficit/hyperactivity disorder and conduct problems and emerging findings highlight caution about inferring causal effects on bipolar disorder and schizophrenia.

Instead of laboratory: Analyzing data from natural experiments (Nobel Memorial Prize in Economic Sciences 2021)

2022 ◽  
pp. 5-22
Author(s):  
K. I. Sonin
Keyword(s):  
Empirical Analysis ◽  
Statistical Methods ◽  
Nobel Prize ◽  
Labor Economics ◽  
Natural Experiment ◽  
Heterogeneous Reactions ◽  
Causal Effects ◽  
Natural Experiments ◽  
Policy Changes ◽  
New Approach

The 2021 Nobel Prize in Economic Sciences was awarded to David Card, Joshua Angrist, and Guido Imbens for advancing methodology to establish casual relationships in economics. Their approach brought the notion of the natural experiment, situations in which heterogeneous reactions of different groups of people to chance shocks or policy changes allows to elicit causal effects, to the forefront of empirical analysis, and spearheaded a revolution in development of statistical methods needed to analyze the data. After the initial contributions in labor economics and economics of education, the new approach has become a new standard in economic sciences.

Natural “Natural Experiments” in Economics

2000 ◽  
Vol 38 (4) ◽  
pp. 827-874 ◽  
Author(s):  
Mark R Rosenzweig ◽  
Kenneth I Wolpin
Keyword(s):  
Natural Experiment ◽  
Recent Literature ◽  
Economic Models ◽  
Empirical Knowledge ◽  
Natural Experiments ◽  
Simple Economic

The recent literature exploiting natural events as “natural experiment” instruments is reviewed to assess to what extent it has advanced empirical knowledge. A weakness of the studies that adopt this approach is that the necessary set of behavioral, market, and technological assumptions made by the authors in justifying their interpretations of the estimates is often absent. The methodology and findings from twenty studies are summarized and simple economic models are used to elucidate the implicit assumptions made by the authors and to demonstrate the sensitivity of the interpretations of the findings to the relaxation of some of these assumptions.

scholarly journals Psychological Pressure in Competitive Environments: Evidence from a Randomized Natural Experiment

2010 ◽  
Vol 100 (5) ◽  
pp. 2548-2564 ◽  
Author(s):  
Jose Apesteguia ◽  
Ignacio Palacios-Huerta
Keyword(s):  
Natural Experiment ◽  
Psychological Effects ◽  
Control Groups ◽  
Sequential Order ◽  
Socioeconomic Outcomes ◽  
Competitive Environments ◽  
First Mover Advantage ◽  
Psychological Pressure ◽  
And Control ◽  
First Mover

Emotions can have important effects on performance and socioeconomic outcomes. We study a natural experiment where two teams of professionals compete in a tournament taking turns in a sequence. As the sequential order is determined by the random outcome of a coin flip, the treatment and control groups are determined via explicit randomization. Hence, absent any psychological effects, both teams should have the same probability of winning. Yet, we find a systematic first-mover advantage. Further, professionals are self-aware of their own psychological effects and, when given the chance, they rationally react by systematically taking advantage of these effects. (JEL C93, D03, D82, L83)

scholarly journals Passive Smoking Increases Exposure to Polycyclic Aromatic Hydrocarbons and Leads to Prolonged Lipid Oxidation: Lessons From a Natural Experiment in Travelers From Los Angeles to Beijing

2020 ◽  
Author(s):  
Yan Lin ◽  
Yifang Zhu ◽  
Xinghua Qiu ◽  
Fen Yin ◽  
Chi-Hong Tseng ◽  
...  
Keyword(s):  
Lipid Peroxidation ◽  
Los Angeles ◽  
Passive Smoking ◽  
Natural Experiment ◽  
Smoking Prevalence ◽  
Geometric Mean ◽  
Urinary Cotinine ◽  
Polycyclic Aromatic ◽  
And Control ◽  
The Impact

Abstract Background: There are drastic differences in smoking prevalence and control policies among different countries. However, the impact of international travels on passive smoking exposures and subsequent health effects remained unknown. Methods: We recruited 27 non-smokers who travelled from Los Angeles to Beijing for 10 weeks in 2014 and 2015. Urine samples (n=197) were collected before (LA-before), during (Beijing), and after (LA-after) the trip, for the assessment of biomarkers of passive smoking (cotinine), PAHs exposure (hydroxylated-PAHs), and lipid peroxidation (malondialdehyde and 8-isoprostane). Results: The geometric mean concentrations of urinary cotinine were 0.13, 1.50, and 0.22 µg/g creatinine in LA-before, Beijing, and LA-after, respectively. Likewise, hydroxylated-PAH levels were significantly higher in Beijing as compared to LA-before or LA-after (p<0.001), in association with the urinary cotinine concentrations (p<0.05). Traveling from Los Angeles to Beijing increased urinary concentrations of malondialdehyde by 51.8% (95%CI, 29.1% to 78.5%), which did not return to baseline levels, up to 4 - 10 weeks after traveling back to Los Angeles. Remarkably, while urinary 8-isoprostane concentrations were similar between Beijing and LA-before, those levels increased by 75.5% (95%CI, 48.5% to 108%) after returning to Los Angeles (LA-after), and were significantly higher than baseline (LA-before, 63.6%; 95%CI, 31.1% to 104%). Urinary concentrations of malondialdehyde and 8-isoprostane were positively associated with urinary cotinine concentrations, independently of hydroxylated-PAHs. Conclusion: Traveling from Los Angeles to Beijing increased exposures to passive smoking, which contributed to increased PAHs exposures and increased lipid peroxidation in the urine.Trial Registration: Not applicable.

scholarly journals Are post-error adjustments influenced by beliefs in free will? A failure to replicate Rigoni, Wilquin, Brass and Burle, 2013

2020 ◽  
Vol 7 (11) ◽  
pp. 200664
Author(s):  
Charlotte Eben ◽  
Zhang Chen ◽  
Emiel Cracco ◽  
Marcel Brass ◽  
Joël Billieux ◽  
...  
Keyword(s):  
Free Will ◽  
Statistical Power ◽  
Original Study ◽  
Control Group ◽  
Control Groups ◽  
Correct Trial ◽  
Manipulation Check ◽  
Subject Design ◽  
And Control ◽  
Larger Sample

In this pre-registered study, we tried to replicate the study by Rigoni et al. 2013 Cognition 127 , 264–269. In the original study, the authors manipulated the participants’ belief in free will in a between-subject design and subsequently measured post-error slowing (i.e. slower responses after an incorrect trial compared with a correct trial) as a marker of cognitive control. They found less post-error slowing in the group with reduced belief in free will (anti-free will group) compared with a control group in which belief in free will was not manipulated. In the present study, we used the same task procedure and the same free will manipulation (Crick text) in an attempt to replicate these findings. However, we used an online procedure and a larger sample size in order to address concerns about statistical power. Similar to the original study, we also used a questionnaire to measure beliefs in free will as an independent manipulation check. We found a difference in the scores on the questionnaire, thus a reduced belief in free will, after reading the Crick text. However, we did not find any difference in post-error slowing between the anti-free will and control groups. Our findings are in line with several other recent findings suggesting that the Crick text manipulation affects the participants’ self-reported belief in free will but not their behaviour. The present study can be considered a high-powered failed replication attempt.

scholarly journals The Separation and Reunification of Germany: Rethinking a Natural Experiment Interpretation of the Enduring Effects of Communism

2020 ◽  
Vol 34 (2) ◽  
pp. 143-171 ◽  
Author(s):  
Sascha O. Becker ◽  
Lukas Mergele ◽  
Ludger Woessmann
Keyword(s):  
World War Ii ◽  
Natural Experiment ◽  
West Germany ◽  
Extensive Literature ◽  
World War ◽  
East And West Germany ◽  
Cautious Interpretation ◽  
East And West ◽  
And Gender ◽  
And Control

German separation in 1949 into a communist East and a capitalist West and their reunification in 1990 are commonly described as a natural experiment to study the enduring effects of communism. We show in three steps that the populations in East and West Germany were far from being randomly selected treatment and control groups. First, the later border is already visible in many socio-economic characteristics in pre-World War II data. Second, World War II and the subsequent occupying forces affected East and West differently. Third, a selective fifth of the population fled from East to West Germany before the building of the Wall in 1961. In light of our findings, we propose a more cautious interpretation of the extensive literature on the enduring effects of communist systems on economic outcomes, political preferences, cultural traits, and gender roles.

Mixing Methods: A Bayesian Approach

2015 ◽  
Vol 109 (4) ◽  
pp. 653-673 ◽  
Author(s):  
MACARTAN HUMPHREYS ◽  
ALAN M. JACOBS
Keyword(s):  
Qualitative Data ◽  
Causal Effects ◽  
Causal Process ◽  
Causal Inferences ◽  
Prior Beliefs ◽  
Joint Learning ◽  
Multimethod Research ◽  
Qualitative And Quantitative ◽  
Qualitative Evidence ◽  
Process Level

We develop an approach to multimethod research that generates joint learning from quantitative and qualitative evidence. The framework—Bayesian integration of quantitative and qualitative data (BIQQ)—allows researchers to draw causal inferences from combinations of correlational (cross-case) and process-level (within-case) observations, given prior beliefs about causal effects, assignment propensities, and the informativeness of different kinds of causal-process evidence. In addition to posterior estimates of causal effects, the framework yields updating on the analytical assumptions underlying correlational analysis and process tracing. We illustrate the BIQQ approach with two applications to substantive issues that have received significant quantitative and qualitative treatment in political science: the origins of electoral systems and the causes of civil war. Finally, we demonstrate how the framework can yield guidance on multimethod research design, presenting results on the optimal combinations of qualitative and quantitative data collection under different research conditions.

Sign in / Sign up

Export Citation Format

Share Document