This end-to-end data analytics project investigates global compensation trends in the data science field, drawing on a dataset of 607 cleaned salary records spanning 2020 to 2022. The raw dataset was cleaned and enriched using Python, with experience levels, employment types, and remote-work status mapped into descriptive categories and outliers flagged using the IQR method. SQL was then used to answer 30 business questions covering experience-level pay gaps, geographic salary variation, remote-work premiums, and company-size effects, while an interactive Streamlit dashboard brings the findings to life through live filtering and visualization. The project aimed to surface actionable compensation insights for job seekers, recruiters, and HR teams navigating the global data science job market.
- Clean and validate raw salary data, flagging duplicates and outliers
- Engineer descriptive features for experience level, employment type, and remote status
- Answer 30 business questions using SQL to uncover compensation patterns
- Build an interactive Streamlit dashboard with live filters and KPI cards
- Deliver actionable salary benchmarking insights to stakeholders
A combination of data science libraries, SQL, and a web app framework was used to build an efficient and reproducible analysis pipeline:
- Data cleaning, deduplication, and outlier detection (IQR method)
- Feature engineering for categorical and ordinal variables
- Exploratory and statistical data analysis with SQL and Python
- Interactive dashboard design and data storytelling
- Compensation analysis and business reporting
📁 Data-Analytics-Project/
├── 📂 Dataset/
│ ├── Raw_Data.csv
│ └── Cleaned_Data.csv
├── 📂 Python/
│ ├── Data_Cleaning_Script.py
│ ├── Data_Cleaning.ipynb
│ ├── EDA.ipynb
│ ├── Data_Visualization.ipynb
│ ├── Dashboard.py
│ └── Requirements.txt
└── 📂 SQL/
├── Data_Cleaning.sql
├── EDA.sql
└── Business_Queries.sql
A visual walkthrough of the analysis outputs — from the interactive Streamlit dashboard to salary distribution charts built in Python.
After cleaning, the dataset retained 607 salary records across 50 countries and three years (2020–2022). Executive-level professionals earned the highest average salary at roughly $199K, more than triple the entry-level average of around $62K, while Senior-level and Mid-level roles averaged about $139K and $88K respectively. Data Scientist, Data Engineer, and Data Analyst were the three most common job titles in the dataset. Around 63% of recorded roles were fully remote, and remote positions were consistently competitive with on-site pay. The United States and a handful of countries with smaller sample sizes, such as Russia, topped the list for average company-location salary.
This project exemplifies how SQL, Python, and interactive dashboards can be combined to turn raw compensation data into clear, decision-ready insights. By cleaning and structuring the data, answering 30 targeted business questions, and surfacing the results through a live Streamlit dashboard, the project produced a comprehensive view of how experience, location, remote status, and company size shape data science pay. The insights gained can support salary benchmarking, recruitment strategy, and negotiation, and the workflow itself serves as a reusable template for future compensation-analysis projects.