Interpret and Describe Causal Effect Estimates Using E-values
margot_interpret_table.Rd
This function interprets the output of causal effect analysis, providing textual descriptions of causal effect estimates. It categorises the strength of evidence for causality based on E-values and confidence intervals, and generates a detailed interpretation of the effect estimates according to specified causal scales (i.e., "causal_difference" or "risk_ratio") and estimands. This function now supports interpreting results on both the causal difference and risk ratio scales.
Arguments
- df
Data frame containing causal effect estimates, expected to include columns for outcome names, effect estimates (either differences or ratios), confidence intervals, E-values, and a summary estimate label. The structure of `df` should align with the specified `causal_scale`.
- causal_scale
Character string specifying the causal scale used in the analysis. Currently supports "causal_difference" for differences in means or medians, and "risk_ratio" for comparing ratios of probabilities or risks.
- estimand
Character string indicating the type of causal estimand interpreted: "PATE" (Population Average Treatment Effect), "ATE" (Average Treatment Effect), "ATT" (Average Treatment Effect in the Treated), or "CATE" (Conditional Average Treatment Effect).
Value
A character vector containing a detailed interpretation of each outcome in `df`, including the type of estimand, the causal contrast, E-values, and the strength of evidence for causality. The interpretation includes whether there is evidence for causality based on the E-value and confidence interval, tailored to the specified causal scale.