External Validity: From Do-Calculus to Transportability Across Populations

@article{Pearl2015ExternalVF,
  title={External Validity: From Do-Calculus to Transportability Across Populations},
  author={Judea Pearl and Elias Bareinboim},
  journal={Probabilistic and Causal Inference},
  year={2015}
}
The generalizability of empirical findings to new environments, settings or populations, often called "external validity," is essential in most scientific explorations. This paper treats a particular problem of generalizability, called "transportability," defined as a license to transfer causal effects learned in experimental studies to a new population, in which only observational studies can be conducted. We introduce a formal representation called "selection diagrams" for expressing… 

Figures from this paper

Generalizability of causal and statistical relations

TLDR
The work develops a theoretical framework for understanding, representing, and algorithmizing the generalization problem as encountered in many practical settings in data-intensive fields, and puts many apparently disparate generalization problems under the same theoretical umbrella.

Generalizing experimental findings: identification beyond adjustments

We aim to generalize the results of a randomized controlled trial (RCT) to a target population with the help of some observational data. This is a problem of causal effect identification with multiple

A Review of Generalizability and Transportability

When assessing causal effects, determining the target population to which the results are intended to generalize is a critical decision. Randomized and observational studies each have strengths and

Towards causally interpretable meta-analysis: transporting inferences from multiple studies to a target population

We take steps towards causally interpretable meta-analysis by describing methods for transporting causal inferences from a collection of randomized trials to a new target population,

Toward Causally Interpretable Meta-analysis: Transporting Inferences from Multiple Randomized Trials to a New Target Population.

We take steps toward causally interpretable meta-analysis by describing methods for transporting causal inferences from a collection of randomized trials to a new target population, one trial at a

Generalizing Study Results: A Potential Outcomes Perspective.

TLDR
This work discusses how a version of direct standardization (the g-formula, adjustment formula, or transport formula) or inverse probability weighting can be used to generalize a causal effect from a study sample to a well-defined target population, and demonstrates their application in an illustrative example.

Generalizing experimental results by leveraging knowledge of mechanisms

TLDR
This work uses structural causal models and a refined version of selection diagrams to represent knowledge of local mechanisms that produce the outcome of interest and decides whether it entails the invariance of probabilities of causation across populations, which then enables generalization.

A Unified Approach to Generalized Causal Inference

TLDR
Generalized causal diagrams, or g-dags, make explicit what is being assumed in discussing the generalizability of experiments, which allows for constructive discourse and informed research agendas.

A Unified Approach to Generalized Causal Inference

TLDR
Generalized causal diagrams, or g-dags, make explicit what is being assumed in discussing the generalizability of experiments, which allows for constructive discourse and informed research agendas.

Target Validity and the Hierarchy of Study Designs.

TLDR
It is shown that even perfect internal validity does not ensure that a causal effect will be unbiased in a specific target population, and a joint measure of the validity of an effect estimate with respect to a specific population of interest may be more useful.
...

References

SHOWING 1-10 OF 112 REFERENCES

Transportability across studies: A formal approach

TLDR
It is demonstrated how transportability analysis can guide the transfer of knowledge among non-experimental studies to minimize re-measurement cost and improve prediction power.

A General Algorithm for Deciding Transportability of Experimental Results

TLDR
This article establishes a necessary and sufficient condition for deciding when causal effects in the target population are estimable from both the statistical information available and the causal information transferred from the experiments, and provides a complete algorithm for computing the transport formula.

Transportability of Causal Effects: Completeness Results

TLDR
This paper establishes a necessary and sufficient condition for deciding when causal effects in the target domain are estimable from both the statistical information available and the causal information transferred from the experiments.

Causal Transportability with Limited Experiments

TLDR
This paper establishes a necessary and sufficient condition for causal effects in the target domain to be estimable from both the non-experimental information available and the limited experimental information transferred from the source.

Meta-Transportability of Causal Effects: A Formal Approach

TLDR
This paper provides a complete algorithm for computing the transport formula, that is, a way of fusing observational and experimental information to synthesize an unbiased estimate of the desired eects in the target domain.

Causation, Prediction, and Search, 2nd Edition

TLDR
The authors axiomatize the connection between causal structure and probabilistic independence, explore several varieties of causal indistinguishability, formulate a theory of manipulation, and develop asymptotically reliable procedures for searching over equivalence classes of causal models.

Compound Treatments and Transportability of Causal Inference

TLDR
The causal effect of compound treatments and their identifiability conditions are described and how such issues of “transportability” are related to the consistency condition in causal inference are discussed.

Causal inference in statistics: An overview

This review presents empiricalresearcherswith recent advances in causal inference, and stresses the paradigmatic shifts that must be un- dertaken in moving from traditionalstatistical analysis to

Pearl's D-Separation: One More Step Into Causal Thinking

It is frequently claimed that experiments can attest to causal assertions, whereas cross-sectional survey research cannot underwrite similar causal claims. Some logical deficiencies with deriving

Estimating causal effects of treatments in randomized and nonrandomized studies.

A discussion of matching, randomization, random sampling, and other methods of controlling extraneous variation is presented. The objective is to specify the benefits of randomization in estimating
...