See also Experimental design.

Pearl, J. (1995). Causal diagrams for empirical research.

*Biometrika*,*82*(4), 669–688. doi:10.1093/biomet/82.4.669Pearl, J. (2009). Causal inference in statistics: An overview.

*Statistics Surveys*,*3*, 96–146. doi:10.1214/09-SS057General introduction to causal diagrams, confounding criteria, counterfactuals, and so on.

Scott Cunningham’s

*Causal Inference: The Mixtape*, a free causal inference book (though written for Stata). Includes discussion of matching, regression discontinuity, instrumental variables, differences-and-differences, and other causal designs commonly used in economics.Daniel, R. M., Kenward, M. G., Cousens, S. N., & Stavola, B. L. D. (2011). Using causal diagrams to guide analysis in missing data problems.

*Statistical Methods in Medical Research*,*21*(3), 243–256. doi:10.1177/0962280210394469A treatment of missing data using causal DAGs, rewriting the usual MCAR/MAR/MNAR distinction in causal terms and giving criteria for determining when complete case analyses are suitable.

Elwert, F., & Winship, C. (2014). Endogenous selection bias: The problem of conditioning on a collider variable.

*Annual Review of Sociology*,*40*(1), 31–53. doi:10.1146/annurev-soc-071913-043455A DAG-based discussion of endogenous selection, i.e. conditioning on colliders. Missing data and sampling biases, such as non-response bias, can be causes of endogenous selection, so the principles here can apply to analysis of survey data and missing data.

Mohan, K., & Pearl, J. (2021). Graphical models for processing missing data.

*Journal of the American Statistical Association*,*116*(534), 1023–1037. doi:10.1080/01621459.2021.1874961A unified treatment of missing data in causal terms, extending Daniel et al (2011).

Whittemore (GitHub), a Clojure-based DSL for performing causal queries.