Salary Analysis Based on Job Mode & Education Level

Author: Hon Wa Ng
Date: October 2024

Overview

This project analyzes salary variations across different job modes (remote vs. in-person) and education levels (Bachelor’s, Master’s, and Professional degrees) using statistical methods such as Welch’s t-test, ANOVA, and bootstrapping. The dataset is sourced from Stack Overflow’s Developer Survey 2024.

Objectives

• Compare salaries between remote and in-person workers.
• Evaluate salary trends across education levels.
• Perform statistical testing to determine significance in salary differences.
• Use bootstrapping to ensure robust comparisons.

Methodology

1. Exploratory Data Analysis (EDA)

• Visualizing salary trends by job mode and education level.
• Identifying distributions and potential outliers.

2. Hypothesis Testing

• Welch’s t-test: Compare remote vs. in-person salaries.
• ANOVA: Assess salary differences across education groups.
• Kolmogorov-Smirnov Test: Check for normality.
• Levene’s Test: Evaluate variance equality.

3. Bootstrapping for Robustness

• Generate 10,000 resampled salary means.
• Compute pairwise salary differences.
• Normalize mean differences using Welch’s t-score.

4. Result Interpretation

• Statistical significance of job mode and education level on salary.
• Confidence in findings using bootstrapped validation.

Key Findings

• Remote vs. In-Person Salaries:

  • Significant salary differences identified using t-tests and bootstrapping.
  • Bootstrapped results align with the original t-test conclusions.

• Education Level and Salary:

  • ANOVA shows statistical significance in salary differences.
  • Higher education levels correlate with increased salary, but variability exists.

Data Source

• Stack Overflow Developer Survey 2024 (https://survey.stackoverflow.co/)
• Contains salary, education level, work mode, and demographic information.

Repository Structure

Salary_Analysis_StackOverflow/
│── data/
│   │── sample_data.csv  # Reduced dataset
│   │── .DS_Store        # System file (recommended to remove)
│
│── LICENSE              # License for the project
│── reduce_size.ipynb    # Notebook for dataset size reduction
│── salary_analysis.ipynb # Jupyter Notebook for analysis
│── .DS_Store            # Duplicate system file (recommended to remove)

Notes:

• .DS_Store files are automatically created by macOS. It is recommended to remove them using:

  find . -name ".DS_Store" -delete
  
  

---

Installation & Usage

1. Clone the Repository

- git clone https://github.com/your-username/Salary_Analysis_StackOverflow.git cd Salary_Analysis_StackOverflow

2. Install Dependencies

pip install -r requirements.txt

3. Run the Analysis

python salary_analysis.py

---

Results Visualization

This project includes multiple visualizations such as:

• Salary trends by experience (Line plot).
• Average salary by education level (Bar chart).
• Job mode salary distributions (Histograms & KDE plots).
• Outlier detection & removal using IQR.
• Bootstrapped confidence intervals & statistical tests.

Sample Output:

• Bootstrapped t-test result: t-statistic: -2.35, p-value: 0.018
• Conclusion: Remote workers earn significantly more than in-person workers.

---

Data Source

• Stack Overflow Developer Survey 2024
• Contains salary, education level, work mode, and demographic information.

---

Next Steps & Improvements

• Extend analysis to other factors (e.g., programming language, country).
• Implement Machine Learning models for salary prediction.
• Automate data processing with Pandas & NumPy optimizations.

---

Notes

Remove .DS_Store files (macOS system files) before pushing to GitHub

find . -name ".DS_Store" -delete

---