Exploratory Data Analysis (EDA)

Exploratory data analysis of the Stack Overflow Developer Survey 2025 (results.csv) — examining what drives developer compensation worldwide.

Overview

This project uses main.ipynb to analyze developer salary data. It cleans the raw survey responses, then explores:

• Salary distribution — median/mean annual compensation (USD)
• Top paying programming languages — median salary per language (min 500 respondents)
• Remote vs on-site work — how work setup affects compensation
• Geographic disparities — median salary across the most represented countries
• Experience vs salary — relationship between years coding and compensation

Requirements

• Python 3.8+
• Jupyter Notebook
• Pandas
• NumPy
• Matplotlib
• Seaborn

Installation

# Create a virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install pandas numpy matplotlib seaborn jupyter

Usage

1. Place the survey data (results.csv) in the project root.
2. Run the notebook:

jupyter notebook main.ipynb

Data & Cleaning

The notebook loads only the columns needed to reduce memory usage:

ResponseId, MainBranch, Employment, RemoteWork, EdLevel, YearsCode, DevType, Country, LanguageHaveWorkedWith, DatabaseHaveWorkedWith, PlatformHaveWorkedWith, ConvertedCompYearly

Cleaning steps:

• Drop rows with missing salary
• Filter to realistic annual salaries ($5,000 – $500,000)
• Convert YearsCode text values (Less than 1 year → 0.5, More than 50 years → 51) to numeric
• Explode the semicolon-separated LanguageHaveWorkedWith column for per-language analysis

Key Findings (as of this run)

• Median global salary: ~$XX,XXX USD — computed in the notebook
• Remote workers earn more on median than on-site workers
• US, Germany, and India dominate respondent counts
• Experience has a weak-to-moderate positive correlation with salary

Update this section with the actual figures printed by the notebook.

Project Structure

Exploratory Data Analysis/
├── main.ipynb    # EDA notebook
├── results.csv   # Survey dataset (2025 schema)
└── README.md