A Unified Approach to Generalized Causal Inference

35 Pages Posted: 2 Aug 2013 Last revised: 29 Aug 2013

See all articles by Fernando Martel García

Fernando Martel García

Cambridge Social Science Decision Lab Inc.

Date Written: August 1, 2013


Randomized controlled trials and natural experiments have been criticized for their lack of generalizability (external validity), questioning their usefulness to social science and policy. Here I show how three common approaches to generalizability -- the heuristic, statistical, and structural approaches -- are each incomplete on their own, and how generalized causal diagrams, or g-dags, can achieve a complete representation of the problem. G-dags combine theory and evidence to (1) make inferences from a study to a population or subgroup; (2) combine two or more studies that are not generalizable on their own into a generalized inference; (3) encode and test generalizable knowledge; and (4) provide a link to boosting algorithms as generalized additive models. More generally, g-dags make explicit what is being assumed, or questioned, in discussing the generalizability of experiments, which allows for constructive discourse and informed research agendas.

Keywords: external validity, generalizability, causal diagrams, dags, experiments, field experiments, program evaluation, social experiments

JEL Classification: C9

Suggested Citation

Martel García, Fernando, A Unified Approach to Generalized Causal Inference (August 1, 2013). Available at SSRN: https://ssrn.com/abstract=2304970 or http://dx.doi.org/10.2139/ssrn.2304970

Fernando Martel García (Contact Author)

Cambridge Social Science Decision Lab Inc. ( email )

Washington, DC
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics