How naren1704/ai-dev-survey-analysis works
This R project analyzes the Stack Overflow 2025 developer survey, from downloading and cleaning responses through statistical analysis, predictive modeling, clustering, and publication of results. Its main workflow is documented in the README; analysis scripts 02–06 read the cleaned dataset, while the runner executes the pipeline and renders the report. The graph reflects documented relationships only: source excerpts were unavailable, so it does not infer specific function calls or unshown data transfers.
Main entry points: run_all.R Pipeline runner · processed Cleaned dataset · 01_data_cleaning.R Data cleaning · 06_paper_benchmark.R Model benchmark
Components 20
Data preparation
- Download setup
scripts/00_setup.R - Raw survey data
data/raw - Data cleaning
scripts/01_data_cleaning.R - Cleaning functions
R/cleaning.R - Cleaned dataset
data/processed - Analysis setup
R/setup.R
Analysis
- Exploratory analysis
scripts/02_eda.R - Statistical tests
scripts/03_statistical_tests.R - Regression models
scripts/04_modeling.R - Developer clustering
scripts/05_clustering.R - Model benchmark
scripts/06_paper_benchmark.R - Statistical helpers
R/stats.R - Pipeline runner
scripts/run_all.R
Research outputs
- Analysis tables
outputs/tables - Analysis figures
outputs/figures - Benchmark outputs
paper/tables - Research paper
paper/main.tex - Report source
reports/final_report.Rmd - Rendered report
reports/final_report.html
Other components
- Stack Overflow survey
Connections 24
- Download setup → Stack Overflow survey: downloads data
README.md - Download setup → Raw survey data: stores data
README.md - Raw survey data → Data cleaning: provides CSV
README.md - Data cleaning → Cleaned dataset: writes dataset
README.md - Data cleaning → Cleaning functions: uses functions
README.md - Cleaned dataset → Exploratory analysis: read by
README.md - Cleaned dataset → Statistical tests: read by
README.md - Cleaned dataset → Regression models: read by
README.md - Cleaned dataset → Developer clustering: read by
README.md - Cleaned dataset → Model benchmark: read by
README.md - Pipeline runner → Data cleaning: runs step
README.md - Pipeline runner → Exploratory analysis: runs step
README.md - Pipeline runner → Statistical tests: runs step
README.md - Pipeline runner → Regression models: runs step
README.md - Pipeline runner → Developer clustering: runs step
README.md - Pipeline runner → Model benchmark: runs step
README.md - Pipeline runner → Rendered report: renders report
README.md - Exploratory analysis → Analysis tables: writes results
README.md - Statistical tests → Analysis tables: writes results
README.md - Regression models → Analysis tables: writes results
README.md - Developer clustering → Analysis tables: writes results
README.md - Model benchmark → Benchmark outputs: writes results
README.md - Model benchmark → Research paper: supports paper
README.md - Report source → Rendered report: rendered as
README.md