Skip to contents

Gelman Variation Causal Quartet Data

Source: R/variation_causal_quartet.R
variation_causal_quartet.Rd

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

variation_conceptual_causal_quartet

Format

variation_causal_quartet: A dataframe with 88 rows and 4 variables:

  • dataset: The data generating mechanism

  • x: exposure

  • z: a pre-exposure factor

  • y: outcome

variation_conceptual_causal_quartet: A dataframe with 88 rows and 4 variables:

  • dataset: The data generating mechanism

  • z: A pre-exposure factor

  • causal_effect: The true causal effect

Details

The dataframe variation_conceptual_causal_quartet contains the 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.