https://github.com/pmichaillat/p-hacking

Code and data for the paper "Critical Values Robust to P-hacking"
https://github.com/pmichaillat/p-hacking

critical-values hypothesis-testing metascience p-hacking paper-with-code statistical-significance

Last synced: 3 days ago
JSON representation

Code and data for the paper "Critical Values Robust to P-hacking"

• Host: GitHub
• URL: https://github.com/pmichaillat/p-hacking
• Owner: pmichaillat
• License: mit
• Created: 2023-04-05T23:45:58.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-05-08T16:03:14.000Z (6 months ago)
• Last Synced: 2024-05-09T17:10:01.311Z (6 months ago)
• Topics: critical-values, hypothesis-testing, metascience, p-hacking, paper-with-code, statistical-significance
• Language: MATLAB
• Homepage: https://pascalmichaillat.org/12/
• Size: 148 KB
• Stars: 4
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# Critical Values Robust to P-hacking: Code and Data

This repository contains the code and data accompanying the paper "Critical Values Robust to P-hacking", written by [Adam McCloskey](https://adammccloskey.com) and [Pascal Michaillat](https://pascalmichaillat.org), and published in the [Review of Economics and Statistics](https://doi.org/10.1162/rest_a_01456) in April 2024.

## Paper webpage

The paper is available at https://pascalmichaillat.org/12/.

## Code

The MATLAB code within the `code` folder generates the 9 figures presented in the paper.

### Main script

The main script, `run.m`, orchestrates the production of the 9 figures. The script generates each figure in a MATLAB window and saves the figures as PDF files. Additionally, the data used to generate each figure are stored in CSV files.

### Production scripts

The main script produces the individual figures using the following scripts:

+ `figure1A.m` – Produce figure 1A

+ `figure1B.m` - Produce figure 1B

+ `figure2A.m` - Produce figure 2A

+ `figure2B.m` - Produce figure 2B

+ `figure3A.m` - Produce figure 3A

+ `figure3B.m` - Produce figure 3B

+ `figure4.m` - Produce figure 4A, figure 4B, and figure 4C

### Display script

The production scripts use the `visualDisplay.m` script to format figures, axes, and plots. The formatting is designed to provide a clear visual display of the results.

## Figures

The 9 figures generated by the code are saved as 9 PDF files, and the underlying data are saved as 9 CSV files. The files are stored in the `figures` folder, each corresponding to a specific figure in the paper:

+ `figure1A.pdf`, `figure1A.csv` – Figure 1A

+ `figure1B.pdf`, `figure1B.csv` – Figure 1B

+ `figure2A.pdf`, `figure2A.csv` – Figure 2A

+ `figure2B.pdf`, `figure2B.csv` – Figure 2B

+ `figure3A.pdf`, `figure3A.csv` – Figure 3A

+ `figure3B.pdf`, `figure3B.csv` – Figure 3B

+ `figure4A.pdf`, `figure4A.csv` – Figure 4A

+ `figure4B.pdf`, `figure4B.csv` – Figure 4B

+ `figure4C.pdf`, `figure4C.csv` – Figure 4C

## Software

The results were obtained using MATLAB R2023b on an Apple Silicon Mac running macOS Sonoma 14.3.1.

## License

The content of this repository is licensed under the terms of the MIT License.