The Transparent-Door Monty Hall Problem: Replication Package

Published: 25 October 2026| Version 2 | DOI: 10.17632/4yr34tx4vj.2
Contributor:

Description

Replication materials for "The Transparent-Door Monty Hall Problem: An Experimental Study of Learning and Bias" (Chen & Wang, Journal of Economic Psychology, DOI forthcoming). Contains trial-level experimental data (82 subjects, 30 rounds), subject-level questionnaire data, Python scripts that reproduce the figures, regression tables, questionnaire tests, and the reinforcement-learning and epiphany-learning model fits, and the corresponding output files. Analysis code is released under the MIT License (see LICENSE). See README.md for run order and variable definitions.

Files

Steps to reproduce

Steps to reproduce 1. Environment: Python 3.11.13. Install the pinned packages with: pip install -r requirements.txt (numpy 2.0.1, pandas 2.3.2, scipy 1.16.1, statsmodels 0.14.5, matplotlib 3.10.6). No GPU is needed. 2. Unzip the archive and open a terminal in analysis/code/. Paths are resolved relative to each script, so the folder can be placed anywhere. 3. Run the scripts in this order: a. python process_data.py Merges the inputs into analysis/temp/processed_data.csv. b. python draw_group_learning_curve.py Figures 2-4 (analysis/output/switch_rate_*.eps/.pdf). c. python run_reg.py Logit regressions for Tables 1 and 2 (analysis/output/regression_results.txt). d. python t-test_questionaire.py Questionnaire tests in Section 4.3 (analysis/output/questionaire_ttest_results.txt). e. python fit_learning_models.py RL and EL fits for each subject and period range (analysis/output/model_fitting_results.csv and model_fitting_summary.txt). Takes under a minute on a typical laptop. f. python make_appendix_tables.py Per-subject tables in Appendix D (analysis/output/appendix_model_fit_tables.tex). g. python descriptive_stats.py Descriptive switch rates and trial counts quoted in the text (analysis/output/descriptive_stats.txt). 4. Compare the outputs in analysis/output/ with the files already included. Figure 1 (the 3-door schematic) and the Online Appendix screenshots are not produced by these scripts. Variable definitions and the model specification are in README.md.

Institutions

Categories

Psychology, Economics

Licence