Zimbabwe Poverty & Demographics Analysis - Statistical Inference

📈
Data AnalysisDemographic AnalysisStatistical InferencePythonPandasGeospatial AnalysisDevelopment EconomicsPublic PolicyHypothesis TestingData VisualizationSocioeconomic Analysis

MY ROLES

Data AnalystDevelopment Researcher

TIMELINE

2020

TOOLS USED

Demographic Analysis, Statistical Inference, Python, Pandas, Geospatial Analysis, Development Economics, Public Policy, Hypothesis Testing, Data Visualization, Socioeconomic Analysis

DESCRIPTION

This project analyzed Demographic and Health Survey (DHS) data for Zimbabwe to understand the polarized distribution of electricity and cooking fuel access across different regions. The analysis involved summarizing household and household member data by DHS clusters to identify patterns and relationships between utility access, poverty levels, and population characteristics. The study sought to explain why certain areas have clear access to utilities while others completely lack them, with minimal middle ground.

OUTCOMES

  • Identified key factors affecting utility access: poverty levels and population density were statistically significant
  • Determined that population size had minimal impact on electricity access but slight significance for cooking fuel access
  • Developed evidence-based policy recommendations for utility infrastructure development in developing regions

FEATURES

DHS cluster analysis, geospatial data integration, variable relationship mapping, statistical hypothesis testing, socioeconomic trend visualization

CHALLENGES

Working with complex survey data spanning multiple domains. Isolating causal factors when many variables are interrelated. Creating meaningful visualizations to demonstrate polarized distributions. Developing actionable recommendations for development planning.

APPROACH

Preprocessed and summarized household and individual data from five DHS datasets (household data, household member data, births data, cluster information, geographic shapefile). Conducted exploratory analysis to identify overall trends across DHS clusters. Applied inferential statistics to test relationships between utilities access and demographic factors. Demonstrated that the polarized distribution of utilities was primarily explained by the economic inefficiency of implementing infrastructure in areas with low population density and high poverty.