Simulated dataset and R code for MCA of occupational risks and Industry 4.0 technologies

Zenodo – Full Description Title: Simulated dataset and R code for MCA of occupational risks and Industry 4.0 technologies Abstract This repository provides simulated datasets and reproducible R scripts for performing Multiple Correspondence Analysis (MCA), Adjusted Standardized Residuals (ASR), bootstrap analyses, and cluster-stability assessments. The scenarios simulate responses related to Industry 4.0 technologies and occupational risk categories across sample sizes N = 500, 1000, 2000. All materials support step-by-step reproduction of the analysis pipeline, including visualization of MCA results, ASR diagnostics, bootstrap summaries, and publication-ready figures. Contents simulate_mca_data.R — Script to generate the simulated dataset (seed = 20251031). simulated_dataset.csv — Generated dataset (can be reproduced using the script above). mca_analysis.R — Full pipeline: MCA, ASR computation, bootstrap procedures, cluster-stability indices, automatic CSV detection, SPSS export, and per-N outputs. outputs/ — Numerical results and saved objects (MCA eigenvalues, coordinates, contributions), ASR matrices, bootstrap summaries, stability indices, and figures (PNG). Figures/ — Publication-ready MCA plots and screeplots. README.txt — Project description and instructions for replication. citation.txt, citacoes.txt, citation.bib, citation.ris, citation.xml — Citation files in multiple formats. license.txt — Licensing information (CC BY 4.0). Repository Overview This project demonstrates a complete, end-to-end workflow for multivariate categorical data analysis using MCA. It includes: Automatic detection of dataset files in the outputs directory. Simulation of categorical data with a fixed random seed for exact reproducibility. MCA execution using FactoMineR and visualization with factoextra and ggplot2. Computation of Adjusted Standardized Residuals (ASR) to assess local associations. Bootstrap analyses for evaluating coordinate stability and ranking robustness (e.g., Jaccard indices, adjusted Rand index). Cluster-stability checks across simulated scenarios. Per-N exports of eigenvalues, coordinates, screeplots, contributions, and ASR tables. Utilities for SPSS export and 600-dpi figure generation suitable for publication. Requirements and Reproducibility R ≥ 4.5.2 (ucrt build recommended) Required packages: tidyverse, FactoMineR, factoextra, boot, janitor Optional packages for high-quality exports: ragg, magick, rsvg Reproduction steps setwd("PATH/TO/PROJECT") source("simulate_mca_data.R") — Generates simulated_dataset.csv (seed = 20251031). source("mca_analysis.R") — Runs MCA, ASR, bootstrap, cluster stability, and writes all outputs. The main figures are fully reproducible using the fixed seed (20251031). Bootstrap results may vary slightly due to resampling. Notes and Recommendations The dataset is simulated for methodological demonstration and does not represent any specific population. Chi-square warnings may appear for contingency tables; ASR values and bootstrap diagnostics should be used for interpreting associations. For publication-quality graphics, re-render figures from ggplot objects or vector files (PDF/SVG) at 600 dpi with increased font size; avoid upscaling low-resolution bitmaps. Exact reproducibility requires maintaining the seed (20251031) and consistent R package versions. License and Funding License: Creative Commons Attribution 4.0 (CC BY 4.0). Funding: Universidade Federal do ABC; CAPES. How to Cite Destiné, A.; Fávero, P.B.; Rodrigues, L.H.; Ribeiro Rodrigues, L. (2025). Simulated dataset and R code for MCA of occupational risks and Industry 4.0 technologies. Zenodo.DOI: https://doi.org/10.5281/zenodo.17575199