DATA-557-Project

Overview

This repository contains an applied statistics project that investigates potential sex-based disparities in faculty salaries and promotions at a U.S. university. It includes:

• app.py – Main Streamlit application
• promotions_analysis.py – Analysis functions/classes for faculty promotions
• salary_bias_analysis.py – Functions/classes related to salary data and potential bias
• salary.txt – Faculty salary dataset (1976–1995, excluding medical school)
• requirements.txt – Python dependencies
• .streamlit/config.toml – Streamlit configuration (theme, layout)

---

Key Research Questions

1. 1995 Salaries
   Investigate if there is evidence of sex bias in the most recent salary records.
2. Starting Salaries
   Compare entry-level pay for men and women over time.
3. Salary Increases (1990–1995)
   Determine whether salary growth rates differ by sex.
4. Promotions (Associate to Full)
   Examine whether one sex has a higher or quicker promotion rate.

---

Project Highlights

• Statistical Methods: T-tests, two-proportion z-tests, Kaplan-Meier analysis, OLS and logistic regression.
• Data Handling: Pandas for data cleaning and merging.
• Visualizations: Plotly for bar charts, box plots, line plots, and survival curves.
• Interactive App: Streamlit to create a dynamic, filterable dashboard for exploring results.

---

How to Run

1. Clone the Repository

   git clone https://github.com/rapugustino/DATA-557-Project.git
   cd DATA-557-Project

2. Install Dependencies

   pip install -r requirements.txt

3. Run the App

   streamlit run app.py

   • The app will open in your default web browser at http://localhost:8501.

4. Access the Dashboard Navigate to the dashboard in your web browser and explore the results.

Acknowledgments

• Data & Research Questions: Provided by Professor Scott Emerson.
• Project Context: Developed within the framework of an applied statistics course, guided by Professor Katie Wilson’s instruction.
• Team Collaboration: Collaboration between team members each contributing to data preparation, analysis, and interpretation. The team includes Kyle Cullen Bretherton, Aravindh Manavalan, Richard Pallangyo, Akshay Ravi, and Vijay Balaji S.