This repository contains data and supplementary material for Chapter 5 of the thesis Role of Robust Statistical Methods in the Credibility Movement of Psychological Science.
The associated OSF repository can be found here: https://osf.io/znfc2/
The repository is organised as follows:
data
folder:helper_data/raw_data_database_osf_public.csv
contains the list of included and excluded studies, reasons for exclusions, links to raw data and dois/links to papers. Each paper has been assigned an ID. This ID is used to for the matching dataset files and analysis files specified below.processed_data/mod_df.RDS
contains the processed data. Use this file along withr_docs/data_processing
to get additional information about the metrics of interestraw_data
contains all raw data from included studies.
r_docs
folder:- Data processing (
data_processing.qmd
) and analyses (data_analysis.qmd
) ran on processed data to obtain typical estimates for metrics (skewness, kurtosis, heteroscedasticity, etc) - Individual analysis files for each paper labelled with appropriate ID.
- Data processing (
objects
folder:lm
folder has the exports of all linear models from individual papersmodels
contains the results and MCMC draws from all Bayesian models fitted to obtain typical estimates. Note that this file was too large to host on github. It's hosted on OSF instead - you can access it through the repository link above, or use this link that links directly to the file: https://osf.io/ts9fb
scripts
folder contains helper functions used for re-analyses of linear models and for extracting information about residual metrics.