Skip to content

datanautical/datanautical.github.io

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 

Repository files navigation

I’m David, a B.S. in Data Analytics graduate with over 15 years experience in Telco, Utilities and Security. I build practical solutions at the intersection of data, automation, and systems—turning complex, real-world problems into clear insights, reproducible processes, and outcomes we can actually use.


🚀 Featured Projects

Below are the featured projects I’m most proud of right now. Each includes a short overview and a link to the repo.


1) Turning Churn into Opportunity

Do contract terms have an impact on customer churn?

Summary

This project applies statistical analysis to a telecommunications dataset to determine whether customer contract type is significantly associated with churn behavior. Using the Chi-Square Test of Independence and Cramér’s V effect size, my analysis demonstrates that month-to-month customers churn at substantially higher rates than long-term contract customers.

The focus of this project is on statistical rigor, reproducibility, and clear business interpretation.


Highlights

  • Cleaned and prepared ~7,000 customer records for analysis
  • Constructed contingency tables for categorical comparison
  • Performed Chi-Square statistical testing with assumption validation
  • Calculated Cramér’s V to measure practical effect size
  • Visualized churn rates across contract types
  • Translated statistical results into actionable business retention insight
  • Designed for full reproducibility inside a Jupyter Notebook

Tech Stack

  • Python
  • Pandas
  • SciPy
  • Matplotlib
  • Jupyter Notebook

Repo: https://github.com/datanautical/Turning_Churn


2) Data Visualization: Income, Poverty, and Unemployment Patterns

How do income and poverty levels relate to unemployment across U.S. states?

Summary

This project uses Tableau to explore the relationship between median household income, poverty rate, and unemployment across U.S. states.

Through a series of scatter plots and a geographic dashboard, the analysis reveals a consistent inverse relationship between income and unemployment, and a direct relationship between poverty and unemployment. The visuals highlight important outliers, regional patterns, and income-stratified comparisons that demonstrate unemployment outcomes are not evenly distributed.

The goal of this project is to demonstrate how thoughtful visual encoding, filtering, and dashboard design can turn raw data into clear, exploratory insight for decision makers.


Key Insights

  • Higher median household income is generally associated with lower unemployment
  • Higher poverty rates are associated with higher unemployment
  • Puerto Rico is a major outlier with both the lowest income and highest unemployment
  • Even among high-income states (>$70k median income), unemployment varies significantly
  • Coastal and regional patterns emerge when income filtering is applied
  • Income alone does not fully explain unemployment differences — regional structure matters

Visualization Highlights

  • Scatter plots comparing:
    • Income vs Unemployment
    • Poverty vs Unemployment
  • Linear regression overlays to summarize trends
  • Bubble size encoding state population
  • Consistent red-to-green color scale to represent unemployment severity
  • Discrete income-band filters for stratified analysis
  • Interactive dashboard combining map and bar chart for geographic and comparative views
  • Carefully designed tooltips, axis formatting, and layout for clarity

Tech Stack

  • Tableau Public
  • Data encoding & dashboard interaction design
  • Visual analytics principles

Links

About

My Portfolio

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors