Analyzing Government Budget Speeches in Sub-Saharan Africa: Public Accountability and Energy Access

This dataset accompanies a research project that analyzes government budget speeches in Sub-Saharan Africa (SSA) in relation to public accountability and energy access. It includes a full pipeline for text extraction, processing, and statistical analysis using R and Stata.

The data folder contains:

Budget speech PDFs (source documents)

Textual analysis outputs including SDG keyword counts and summary statistics

Governance and development indicators from V-Dem and World Bank datasets

Merged panel datasets (with and without imputation)

Scripts for processing, regression analysis, and LASSO-based variable selection

The R script performs keyword extraction and exploratory factor analysis on the speech texts, while the Stata scripts conduct regression analyses to investigate relationships with governance and development indicators.

All files are organized to ensure full reproducibility, and README files are included to guide users through the workflow.