KonFound-It!
konfound-it.org
- Free plan
- Web
7.3 Club score Visit konfound-it.org
What the maker publishes
- Audience
- The project describes its intended users as researchers, policy makers, and pragmatists evaluating statistical inferences. source
- Benchmark tool
- The project provides an interactive tool for sensitivity analyses calculated for the What Works Clearinghouse benchmark dataset. source
- Bias analysis
- The project develops sensitivity analyses that quantify how robust inferences are to omitted variables and other sources of bias. source
- Browser app
- The Shiny app runs sensitivity analyses in a web browser without installing software and is built on the konfound R package. source
- Browser app
- The Shiny app runs sensitivity analyses in a web browser without requiring software installation. source
- Deployment
- both source
- Free plan
- Yes source
- Guidance
- The Practical Guide is intended to help implement ITCV and RIR sensitivity analyses. source
- ITCV
- Impact Threshold for a Confounding Variable estimates the correlations an omitted variable would need with the predictor and outcome to invalidate an inference; the FAQ describes it as intended for linear models. source
- ITCV
- Impact Threshold for a Confounding Variable estimates how strongly an omitted variable would need to relate to the predictor and outcome to change an inference. source
- ITCV caveat
- The package documentation cautions that ITCV was originally derived for OLS standard errors and should be interpreted cautiously when reported standard errors are not OLS-based. source
- Known limitation
- The FAQ says there is currently no module specifically designated for multinomial outcomes. source
- Known model gaps
- The FAQ says there is no module specifically designated for multinomial outcomes, and that predictive models using machine learning and fixed effects are under development. source
- License
- The konfound R package is listed as MIT licensed. source
- Maker
- The site describes KonFound-It! as a project run by a group of researchers spanning multiple institutions. source
- Meta-analysis
- The R package includes mkonfound() for comparing or synthesizing sensitivity analyses across multiple analyses. source
- Purpose
- KonFound-It! provides interactive sensitivity analyses to help researchers, policy makers, and pragmatists evaluate the robustness of statistical inferences. source
- Purpose
- KonFound-It! provides sensitivity analyses to evaluate how robust statistical inferences are to potential bias. source
- Regression models
- Yes source
- RIR
- Robustness of Inference to Replacement estimates how much of an observed effect would need to be bias-driven, or how many cases would need to be replaced with cases showing no effect, to change an inference. source
- RIR
- Robustness of Inference to Replacement estimates how much of an observed effect would need to be bias-driven to change an inference. source
- R package
- The konfound R package supports sensitivity analysis for published studies and models fit in R, including RIR and ITCV calculations. source
- R package
- The konfound R package supports sensitivity analysis for models fit in R and published study results. source
- Scripting support
- Yes source
Checked against the maker's pages on 12 Oct 2026.