Data-driven insights into road accident severity in sub-Saharan Africa : a multiple correspondence analysis for SDG-aligned policy

Abstract

In recent years, several impactful studies have provided stakeholders with actionable insights aimed at reducing accident severity, aligning with Sustainable Development Goals 3 and 11, which target a reduction in global deaths and injuries by 2030. Building upon this foundation, the present study applies the Multiple Correspondence Analysis (MCA) technique to uncover complex and latent relationships among categorical variables influencing road accident severity across Sub-Saharan Africa. The dataset comprises 12,316 accident records spanning 2017 to 2020, with 22 carefully selected categorical variables relevant to driver demographics, environmental conditions, vehicle characteristics, and road infrastructure. Through MCA, the dimensionality of the original 182 dimensions was reduced to 29 dimensions based on eigenvalue retention, with the first two dimensions accounting for 60.2% of the total variance. The resulting MCA biplot reveals distinct quadrant-based groupings of variables. The top-right quadrant demonstrates a strong positive correlation among factors such as younger drivers (aged 18-30), vehicle ownership, type of vehicle, service year, presence of medians or lanes, specific accident-prone areas, and weekdays. This cluster suggests that accident severity is significantly influenced by driver age and vehicle characteristics in particular contexts. This study revealed the interrelationships among key features, offering a data-driven foundation upon which policymakers and transport authorities can design and implement targeted interventions. These may include stricter licensing regulations for younger drivers, the enforcement of improved vehicle safety standards, and strategic infrastructural enhancements in identified high-risk zones. The findings provide a strong foundation for the expansion of sustainable road safety strategies and contribute to the growing discourse on mitigating accident severity in Sub-Saharan Africa.