/limited_luck_analysis

Primary LanguagePython

Limited Luck in Space

This repository contains the data and analysis code for the limited luck in space project.

Project structure

.
├── data/
│   ├── processed
│   └── raw
├── figures/
│   ├── main
│   └── si
├── outputs/
├── images/
├── README.md
├── preprocess.py
├── experiment1.py
├── experiment2.py
├── spatial_gamblers_fallacy.py
├── utils.py
├── jags_model_selection.R
└──  requirements.txt

data directory

Contains the raw data downloaded from qualtrics (within the raw directory). The data is in CSV format, where each row is the data from a single response to the experiment. Also contains the data after running preprocess.py (within the processed directory), which removes incomplete responses and rejects responses according to the pre-registration specifications.

figures directory

Contains the figures generated from the different scripts. Figures included in the main paper, are under the main directory. Figures for the supplementary information are under the si directory.

outputs directory

Contains the console outputs of each of the python scripts. The outputs contain regression tables and permutation test results.

images directory

Contains the images used for heatmap plots, downloaded from the Qualtrics experiments.

Project root

Contains the scripts used to perform the analysis and the python requirement file.

  1. preprocess.py - processes the raw data, removing irrelevant columns and rejecting responses.
  2. experiment1.py - runs the analysis and produces the plots for the first experiment.
  3. experiment2.py - runs the analysis and produces the plots for the second experiment, that should be a reproduction of the first.
  4. gamblers_fallacy.py - runs the analysis and produces the plots for the spatial gamblers fallacy experiment.
  5. utils.py - constants and shared utility functions.
  6. jags_model_selection.R - performs Bayesian model fitting and BIC based selection for unimodel vs bimodal hypothesis for the data.