Konfound robustness for a coefficient (konfound extender) — morie_sensitivity

Thin interface to konfound::pkonfound: how many cases would need to be replaced with average-treatment-effect cases (or how large would an omitted-variable correlation have to be) to invalidate the inference? Pairs with morie_sensitivity_omitted_var_bias (which uses the Cinelli-Hazlett partial-R-squared framing instead of the Frank et al. percent-bias-to-invalidate framing).

Usage

morie_sensitivity_konfound(
  estimate,
  se,
  n,
  n_covariates = 0L,
  alpha = 0.05,
  ...
)

Arguments

estimate: Treatment-coefficient estimate.
se: Standard error of estimate.
n: Number of observations.
n_covariates: Number of covariates in the model (excluding the intercept and the treatment). Default 0.
alpha: Significance level. Default 0.05.
...: Additional arguments forwarded to konfound::pkonfound.

Value

A list of class morie_sensitivity_konfound with the percent-bias-to-invalidate, the impact-threshold-of-a- confounding-variable (ITCV), and the raw konfound object.

References

Frank, K. A., Maroulis, S. J., Duong, M. Q., & Kelcey, B. M. (2013). What would it take to change an inference? Educational Evaluation and Policy Analysis, 35(4), 437–460.