This research integrates multiple analytical methods to study road fatalities in Attica, Greece, for the period 2016–2022, examining fatality patterns, the influence of COVID-19, and temporal trends. Descriptive analytics were used to summarise annual and categorical distributions, highlighting key risk groups and temporal fluctuations. Hierarchical clustering was applied to identify structural patterns among fatality categories based on their temporal evolution. A pre-/post-COVID comparison using two-sample t-tests assessed mean shifts across categories, and simple linear regression models characterised fatality trends over time. The statistically significant findings from the COVID comparison are a decrease in children killed (p=0.01) and an increase in single-vehicle crashes (p=0.03) in the post-COVID period. Non-national riders showed a large descriptive increase of +74% that did not reach statistical significance (p=0.99). Descriptive results confirm persistently high fatality counts among motorcyclists and pedestrians, with most incidents occurring within built-up areas and during nighttime conditions. These analyses provide a comprehensive insight into fatality dynamics in Attica and underscore the value of combining descriptive, clustering, and inferential approaches for regional road safety monitoring. The findings are consistent with Safe System principles and highlight the need for targeted interventions addressing vulnerable road users, post-COVID behavioural shifts, and long-term trend monitoring within Greece’s road safety strategy.
| ID | pc677 |
| Presentation | |
| Full Text | |
| Tags |
