Greek Road Accident Analysis, 2007–2021

Overview

This exploratory portfolio project analyzes annual road-accident statistics published by the Hellenic Statistical Authority (ELSTAT) for 2007–2021. It combines descriptive analysis with a fictional road-trip scenario as a data-storytelling device.

The statistical findings are real within the supplied dataset; the narrative framing is fictional. The project does not calculate individual travel risk or recommend a genuinely "safest" route, time, region, or vehicle.

Questions explored

• How did recorded injury-causing road accidents change over the period studied?
• How are recorded accidents distributed across the available geographic and categorical variables?
• Which descriptive patterns can be communicated effectively through charts and an interactive exploration?
• How can a fictional scenario make a public dataset more accessible without changing the underlying results?

Data

Source: Hellenic Statistical Authority, road-traffic-accident statistics for 2007–2021.

The repository should include:

• The exact ELSTAT dataset URL
• The download date
• A data dictionary or explanation of the fields used
• A clear distinction between raw and processed data

ELSTAT describes the 2021 release as statistics on injury-causing road accidents: ELSTAT 2021 road-accident release.

Workflow

1. Import and inspect the ELSTAT time-series file.
2. Clean field names, types, and missing values.
3. Calculate descriptive summaries for the available years and categories.
4. Create visualizations and, where present, an interactive widget.
5. Translate selected patterns into a clearly labelled fictional narrative.

Tools

List only the tools that are present in the repository, for example:

• Python
• pandas
• Matplotlib and/or Seaborn
• Jupyter Notebook
• Interactive-widget library, if actually included

How to use the project

1. Clone or download the repository.
2. Install the packages listed in requirements.txt.
3. Run the notebooks in the documented order.
4. Review the descriptive outputs and the limitations before interpreting the narrative.

Interpretation and limitations

• Accident counts alone do not measure travel risk because they do not account for exposure such as traffic volume, kilometres travelled, or number of journeys.
• Historical aggregate patterns cannot guarantee the safety of an individual trip.
• Apparent differences may reflect reporting practices, road use, population, infrastructure, or other unmodelled factors.
• The analysis is descriptive and does not establish causality.
• The fictional scenario is a communication technique, not a real person or a decision model.