Network Analysis and Community Detection R-Code for Mental Health Assessment Data from Veteran and First Responder Families

Here's the revised Zenodo description: Description: This repository contains the complete R analysis code and supporting data files for the manuscript "Mental health symptom networks in veteran and first responder family members" published in Communications Psychology. The code performs network analysis and community detection on mental health assessment data from 317 treatment-seeking family members of trauma-exposed veterans and first responders. Key Analyses Included: Network estimation using Gaussian graphical models (EBICglasso) Bootstrap stability analysis for network reliability Exploratory Graph Analysis (EGA) for community detection Dimension stability assessment across bootstrap samples Node centrality and expected influence calculations Power analysis for sample size adequacy High-resolution figure generation (300+ DPI) for all manuscript figures Data Files Included: This repository includes the underlying data files supporting Figures 1-3 from the manuscript, which can be fully reproduced using the provided R code. These CSV files contain the processed analysis results including bootstrap stability data, network edge weights, and community detection outcomes, allowing for complete replication of all visualizations without access to the original raw clinical data. Data Structure: The code processes responses from three standardized clinical measures: PTSD Checklist for DSM-5 (PCL-5), Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Disorder-7 (GAD-7). Overlapping items were removed to ensure each symptom was represented only once in the network. Output Files: All manuscript figures in multiple high-resolution formats (PNG, TIFF, SVG/PDF) CSV data files for Figures 1-3 enabling complete reproduction Network edge weights and confidence intervals Node centrality measures and community assignments Bootstrap stability results and power analysis outputs Software Requirements: R version 4.0+ Required packages: bootnet, qgraph, EGAnet, networktools, tidyverse, ggplot2, and others (full list in code) Reproducibility: All analyses use set random seeds (seed = 2711) to ensure complete reproducibility. The code includes comprehensive documentation, parameter specifications, and session information for transparency. The included data files allow researchers to reproduce all manuscript figures without requiring access to the original clinical dataset. Citation: If you use this code or data files, please cite both the original manuscript and this code repository. The analysis methodology follows established best practices for psychological network analysis and community detection as implemented in the bootnet and EGAnet R packages. Keywords: network analysis, mental health, veterans, first responders, family members, PTSD, depression, anxiety, community detection, bootstrap stability, R code, reproducible research, data transparency