variation_causal_quartet {quartets}

Gelman Variation Causal Quartet Data

Description

This dataset contains 88 observations, each generated under a different mechanism of variation of the treatment effect with respect to some pre-exposure characteristic, z:

• (1) Constant effect

• (2) Low variation

• (3) High variation

• (4) Occasional large effects

Usage

variation_causal_quartet

Format

A dataframe with 88 rows and 5 variables:

• dataset: The data generating mechanism

• exposure: exposure

• covariate: a pre-exposure factor

• outcome: outcome

• .causal_effect: Latent true causal effect

References

Gelman, A., Hullman, J., & Kennedy, L. (2023). Causal quartets: Different ways to attain the same average treatment effect. arXiv preprint arXiv:2302.12878.

Hullman J (2023). causalQuartet: Create Causal Quartets for Interrogating Average Treatment Effects. R package version 0.0.0.9000.