███╗ ███╗ ██████╗ ██████╗
████╗ ████║ ██╔════╝ ██╔══██╗ ██████╗ ██╗ ██╗███████╗██████╗
██╔████╔██║ ██║ ██║ ██║██╔═══██╗██║ ██║██╔════╝██╔══██╗
██║╚██╔╝██║ ██║ ██████╔╝██║ ██║██║ █╗ ██║█████╗ ██████╔╝
██║ ╚═╝ ██║ ██║ ██╔═══╝ ██║ ██║██║███╗██║██╔══╝ ██╔══██╗
██║ ██║ ╚██████╗ ██║ ╚██████╔╝╚███╔███╔╝███████╗██║ ██║
╚═╝ ╚═╝ ╚═════╝ ╚═╝ ╚═════╝ ╚══╝╚══╝ ╚══════╝╚═╝ ╚═╝
MCPower GUI is a desktop application for Monte Carlo power analysis. It provides a graphical interface to the MCPower library, letting you plan sample sizes and estimate statistical power for complex study designs — without writing any code.
| Platform | Link |
|---|---|
| Windows | MCPower.exe |
| Linux | MCPower-linux |
| macOS | MCPower-macos.zip |
| if above do not work | SourceForge |
No Python installation required — these are standalone executables.
- Download
MCPower.exe. - Double-click to run.
Note: Windows SmartScreen may show a warning ("Windows protected your PC") because the application is not code-signed. Code signing certificates cost ~$100/year, which is not feasible for a free open-source project. The app is safe — you can verify the source code in this repository. To proceed: click More info → Run anyway.
Your antivirus software may also flag the file. This is a known false positive caused by the packaging tool (PyInstaller) used to create standalone Python executables. Many legitimate open-source applications trigger the same warning.
- Download
MCPower-linux. - Open a terminal in the download folder:
- File manager: Right-click in the folder → Open Terminal Here
- Or manually: open a terminal and run
cd ~/Downloads
- Make the file executable and run it:
chmod +x MCPower-linux ./MCPower-linux
- Download
MCPower-macos.zip. - Double-click the
.zipto unzip it —MCPower.appwill appear in the same folder. - Drag
MCPower.appto your Applications folder (optional but recommended). - Double-click
MCPower.appto launch it.
First launch — Gatekeeper warning: macOS will block the app with a message like "MCPower cannot be opened because it is from an unidentified developer." This happens because the app is not signed with a paid Apple Developer certificate. To allow it:
- Open System Settings → Privacy & Security
- Scroll down to the Security section
- You will see a message: "MCPower was blocked from use because it is not from an identified developer" — click Open Anyway
- On the next prompt, click Open
You only need to do this once. After that, the app opens normally.
Full documentation is available in-app via the Documentation menu item.
Statistical power is the probability that a study will detect a real effect when one exists. A power analysis helps you determine:
- Find Power: Given a sample size, what is the probability of detecting your expected effects?
- Find Sample Size: What sample size do you need to achieve a target power level (e.g., 80%)?
Traditional power formulas work for simple designs but break down with interactions, correlated predictors, categorical variables, or non-normal data. MCPower uses Monte Carlo simulation — it generates thousands of synthetic datasets under your assumptions, fits the statistical model to each, and counts how often the effects are detected. This approach handles arbitrary complexity.
Just type your formula. Enter your model the way you'd write it in R or a stats textbook — outcome = treatment + covariate + treatment*covariate. MCPower handles everything else: parsing the formula, setting up the simulation, and managing interactions and factor coding. No programming required. Logistic regression support is coming soon.
Scenarios show you the full picture. Real studies rarely match textbook conditions — effect sizes may be smaller than expected, distributions may be skewed, or variance may not be constant. One checkbox enables automatic robustness testing. MCPower runs your analysis under optimistic, realistic, and worst-case conditions, so instead of a single number you get a range that shows how sensitive your design is to violated assumptions.
Upload your data, skip the guesswork. Drop in a CSV and MCPower auto-detects variable types (continuous, binary, or categorical), preserves real distributions, and handles correlations between predictors. String columns (e.g. "control", "drug_a", "drug_b") are supported and auto-detected as factors. Factor levels use the original values from your data — so cyl with values [4, 6, 8] shows up as cyl[4], cyl[6], cyl[8] instead of abstract indices. No need to manually specify whether each variable is normal, skewed, or categorical — MCPower samples from the empirical distribution. This is especially useful when you have pilot data or a related dataset and want your power analysis to reflect actual conditions.
| Model setup | ANOVA mode (dark) |
|---|---|
 |
 |
| Define formula, variable types, and effect sizes | Set up factors with custom levels and proportions |
| Running analysis | Results — power chart |
|---|---|
 |
 |
| Progress tracking with simulation count | Scenario comparison and replication script |
| Results — power curves (dark) |
|---|
 |
| Find Sample Size — power curves across sample sizes |
- Model tab — Define your study design: enter a formula, set variable types, specify effect sizes, and optionally upload empirical data with correlations.
- Analysis tab — Choose analysis mode (Find Power or Find Sample Size), set target power, enable scenarios, and select multiple testing corrections.
- Results tab — View power tables, bar charts, power curves, scenario comparisons, and a replication script you can run independently.
MCPower uses R-style formula notation:
| Formula | Meaning |
|---|---|
y = x1 + x2 |
Two main effects |
y = x1 + x2 + x1:x2 |
Two main effects + interaction |
y = x1 * x2 |
Shorthand for x1 + x2 + x1:x2 |
y ~ x1 + (1|group) |
Random intercept (mixed model) |
y ~ x1 + (1 + x1|group) |
Random intercept + slope (mixed model) |
Both = and ~ are accepted as separators between the dependent variable and predictors.
MCPower GUI is built on MCPower. MCPower's mixed-effects solver is validated against R's lme4 across 95 scenarios with 100% pass rate.
GPL v3. If you use MCPower in research, please cite:
Lenartowicz, P. (2025). MCPower: Monte Carlo Power Analysis for Complex Statistical Models (Version ) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.16502734
Replace <your version> with the version you used — check with import mcpower; print(mcpower.__version__).
@software{mcpower2025,
author = {Lenartowicz, Pawe{\l}},
title = {{MCPower}: Monte Carlo Power Analysis for Complex Statistical Models},
year = {2025},
publisher = {Zenodo},
doi = {10.5281/zenodo.16502734},
url = {https://doi.org/10.5281/zenodo.16502734}
}This project is free and open-source. If you'd like to support its development, donations are appreciated!