Global Terrorism Database (GTD) – Data Analysis & Tableau Dashboard

Project Overview

This project presents an end-to-end data analysis and visualization of the Global Terrorism Database (GTD).
It combines data cleaning, exploratory data analysis (EDA) in Python with an interactive Tableau dashboard to uncover global patterns, trends, and characteristics of terrorist incidents over time.

The project is designed as a portfolio piece, demonstrating skills in data wrangling, analytical reasoning, and data storytelling through visualization.

---

Objectives

• Clean and prepare a large, real-world dataset with complex categorical structures
• Explore temporal, geographic, and tactical patterns of terrorist attacks
• Translate analytical findings into an interactive Tableau dashboard for intuitive exploration
• Communicate insights clearly to both technical and non-technical audiences

---

Dataset

Source: Global Terrorism Database (GTD), maintained by the National Consortium for the Study of Terrorism and Responses to Terrorism (START), University of Maryland.

• Event-level data on terrorist attacks worldwide since 1970
• 200,000+ incidents with detailed attributes (location, attack type, weapon type, casualties, targets, etc.)
• Publicly available for research and educational use

Detailed definitions and methodological notes are documented in the official GTD Codebook (GTD_Codebook.pdf).

---

Project Structure

• GTD_Data_Cleaning_and_EDA.ipynb – Data cleaning and exploratory data analysis
• GTD_Dashboard.twbx – Tableau packaged workbook (interactive dashboard)
• GTD_Codebook.pdf – Official GTD methodology and variable definitions
• README.md – Project documentation
• Screenshot.png – Screenshot of the Tableau Dashboard

---

Data Cleaning & Exploratory Analysis

The Jupyter Notebook covers:

• Initial data inspection and variable selection
• Handling missing values and GTD-specific encodings
• Feature simplification for analysis and visualization
• Exploratory analysis of:
  • Distributions of selected features
  • Correlations
  • Linear relationships

The focus is on analytical clarity, not heavy modeling, ensuring that insights remain interpretable and transferable to visualization.

---

Tableau Dashboard

The Tableau dashboard provides an interactive interface to explore the GTD data, including:

• Global and regional trends in terrorist incidents
• Temporal patterns across years
• Distribution of attack types and targets
• High-level comparisons across regions and countries

The dashboard is designed for:

• Fast insight generation
• User-driven exploration via filters and tooltips
• Clear visual hierarchy and storytelling

Open GTD_Dashboard.twbx in Tableau Desktop to explore the dashboard interactively.

---

Tools & Technologies

• Python (pandas, numpy, matplotlib, seaborn, statsmodels)
• Jupyter Notebook
• Tableau Desktop

---

Notes & Limitations

• GTD data collection methods and coverage vary over time and geography; trends should be interpreted with caution
• Some variables are subject to reporting bias or missing information, as documented in the GTD Codebook
• This project focuses on descriptive and exploratory analysis, not prediction or causal inference

---

License & Data Usage

• The GTD dataset is provided for research and educational purposes
• Users must comply with the GTD End User License Agreement

---

Author

Portfolio project by Devin Mavric
Data Analysis | Visualization | Tableau | Python

Screenshot