Logistic Interaction Scenario Explorer
Translate four specified logit coefficients into probabilities and contrasts for two binary predictors.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Translate four specified logit coefficients into probabilities and contrasts for two binary predictors. All default values are fictional.
Calculation results
E=0, M=0: 0.119203; E=1, M=0: 0.214165
E=0, M=1: 0.182426; E=1, M=1: 0.377541
Risk difference when M=0: 0.0949621
Risk difference when M=1: 0.195115
Difference in risk differences: 0.100153
Assumptions and limits
Uses logit(p)=intercept+exposure×E+modifier×M+interaction×E×M, with E and M coded 0/1. Each coefficient must be finite within ±100. Other covariates are absent or absorbed into this intercept at a fixed profile. These are hypothetical conditional predictions, not fitted estimates, standardized population risks, causal effects, or clinical predictions. No coefficient covariance is supplied, so no confidence interval or interaction test is calculated. Risk-scale and log-odds interaction differ; extreme logits may round to 0 or 1.
SciPy: inverse logit · Read the practical guide · Download blank worksheet (TXT)