Statistical Studies
This project combines two statistical analysis initiatives aimed at uncovering insights into critical societal issues. The first analysis focuses on dropout rates in schools, exploring how factors such as gender, region, and school type influence these rates. The second investigates the determinants of women’s principal resource provider status in households using logistic regression. Both projects employ advanced statistical techniques, robust data visualization, and thoughtful analysis to provide actionable insights into educational and social dynamics.
STATISTICSSTATSMODELSDATA VIZUALISATION
1 min read
Main Features
School Dropout Analysis: Identifies key factors influencing school dropout rates, analyzing variables like region, gender, and school sector with statistical methods and visualizations.
Women’s Resource Provider Status: Uses binary logistic regression to determine the predictors of women being primary earners in households, including household size, income, and number of active workers.
Advanced Statistical Techniques: Includes Kruskal-Wallis, Dunn’s test, Mann-Whitney U, and correlation analysis for non-parametric data, alongside logistic regression for predictive modeling.
Rich Visualizations: Generates detailed plots, such as boxplots and scatter plots, for insightful data presentation.
Technology Stack
Data Handling: pandas, numpy
Statistical Analysis: scipy, statsmodels, sklearn
Data Visualization: matplotlib, seaborn
Data Formats: CSV and .sas7bdat for input datasets
Machine Learning: Logistic regression for predictive modeling