Proceeding From Observed Correlation to Causal Inference: The Use of Natural Experiments

  title={Proceeding From Observed Correlation to Causal Inference: The Use of Natural Experiments},
  author={Michael L. Rutter},
  journal={Perspectives on Psychological Science},
  pages={377 - 395}
  • M. Rutter
  • Published 1 December 2007
  • Psychology
  • Perspectives on Psychological Science
This article notes five reasons why a correlation between a risk (or protective) factor and some specified outcome might not reflect environmental causation. In keeping with numerous other writers, it is noted that a causal effect is usually composed of a constellation of components acting in concert. The study of causation, therefore, will necessarily be informative on only one or more subsets of such components. There is no such thing as a single basic necessary and sufficient cause… 
Causal Inference and Observational Research
It is argued that the study of twin pairs discordant on exposure, and in particular discordant monozygotic twins, provides a useful analog to the idealized counterfactual design and is concluded that twin researchers could make greater use of the discordant-twin design to strengthen causal inferences in observational research.
Causal Inference from Descriptions of Experimental and Non-Experimental Research: Public Understanding of Correlation-Versus-Causation
Participants drew causal inferences from non-experimental vignettes as often as they did from experimental vignette, and more frequently for causal statements and directions of association that fit with intuitive notions than for those that did not.
Using Natural Experiments to Test Environmental Mediation Hypotheses
Concepts of causation are discussed, as are alternative explanations of an association other than causation. Nine types of genetically sensitive natural experiments are described, together with their
Causal inference and developmental psychology.
  • E. Foster
  • Psychology
    Developmental psychology
  • 2010
The limitations of regression for causal inference are described and how new tools might offer better causal inference methods are described, in the context of a specific research question, the effect of family structure on child development.
IX. Risk, causation, mediation, and moderation.
This chapter seeks to pull together the findings from the English and Romanian Adoptee studies that were relevant for possible genetic moderation of the effects of institutional deprivation, and presents new data because these were not considered in earlier chapters.
Do natural experiments have an important future in the study of mental disorders?
In this review, a growing range of different types of natural experiment are described and it is highlighted that there is greater confidence about findings where there is a convergence of findings across different designs.
From correlates to causes: can quasi-experimental studies and statistical innovations bring us closer to identifying the causes of antisocial behavior?
The challenges to causal inference posed by observational studies are identified and quasi-experimental methods and statistical innovations that may move researchers beyond discussions of risk factors to allow for stronger causal inference are described.
Isolation in the construction of natural experiments
A natural experiment is a type of observational study in which treatment assignment, though not randomized by the investigator, is plausibly close to random. A process that assigns treatments in a
Living with observational data in biological anthropology.
Issues involved in using statistics to make valid causal inferences from observational data are discussed and methods recommended for this difficult objective have not been widely adopted in the biological anthropology literature.


Testing hypotheses on specific environmental causal effects on behavior.
It is concluded that environmental risk hypotheses can be (and have been) put to the test but that it is usually necessary to use a combination of research strategies.
Testing hypotheses on specific environmental causal effects on behavior
There have been strong critiques of the notion that environmental influences can have an important effect on psychological functioning. The substance of these criticisms is considered in order to
Identification of Causal Effects Using Instrumental Variables: Comment
It is shown that the instrumental variables (IV) estimand can be embedded within the Rubin Causal Model (RCM) and that under some simple and easily interpretable assumptions, the IV estimand is the average causal effect for a subgroup of units, the compliers.
Identification of Causal Effects Using Instrumental Variables
It is shown that the instrumental variables (IV) estimand can be embedded within the Rubin Causal Model (RCM) and that under some simple and easily interpretable assumptions, the IV estimand is the average causal effect for a subgroup of units, the compliers.
Estimating causal effects.
Although one goal of aetiologic epidemiology is to estimate ‘the true effect’ of an exposure on disease occurrence, epidemiologists usually do not precisely specify what ‘true effect’ they want to
Comorbidity: concepts, claims and choices
The presence of comorbidity may distort research findings but, if properly dealt with, it also provides important research opportunities to test hypotheses about causal mechanisms. Seven possible
Assessing Sensitivity to an Unobserved Binary Covariate in an Observational Study with Binary Outcome
This paper proposes a simple technique for assessing the range of plausible causal con- clusions from observational studies with a binary outcome and an observed categorical covariate. The technique
Toward a clearer definition of confounding.
  • C. Weinberg
  • Medicine
    American journal of epidemiology
  • 1993
This paper illustrates that bias can result when adjustment is made for any factor which is caused in part by the exposure under study and is also correlated with the outcome under study.
Although marriage is associated with a plethora of adult outcomes, its causal status remains controversial in the absence of experimental evidence. We address this problem by introducing a
What can mendelian randomisation tell us about modifiable behavioural and environmental exposures?
Using genetic variants as a proxy for modifiable environmental factors that are associated with disease can circumvent some of the problems of observational studies, and the inferences that can be drawn from mendelian randomisation studies depend on the different ways in which genetic variants can proxy for environmentally modifiable exposures.