Obsidian/Doktoritöö/260215 - Vilnius/Struktuur/11. Introduction.md

1 Introduction

Road traffic safety remains one of the major global sources of death and injury; each year more than one million people are killed in road crashes, and approximately 20 % of those victims die (World Health Organization, 2023). This situation seriously hinders progress toward the United Nations Sustainable Development Goals (SDGs), especially SDG 3.6, which calls for a halving of road‑traffic deaths and related injuries by 2030.

Traditionally, the road‑fatality rate—the number of deaths per 10 000 motor vehicles—has been the dominant indicator used by countries to assess traffic safety (Elvik, Vaa, & Christensen, 2019). Although this metric provides valuable information on mortality levels, it reflects only the severity dimension of crashes and neglects a critical component—the frequency of accidents and their total count. Consequently, a single‑indicator approach can lead policymakers to draw inaccurate conclusions when they require a comprehensive and nuanced picture of risk factors (Peden et al., 2020).

The present study proposes a multi‑indicator framework that combines the road‑fatality rate, the absolute number of fatalities, the number of fatal accidents, and the total number of traffic crashes into a single composite index. Using daily traffic‑volume and accident data supplied by the Estonian Road Administration for the period 2010–2018, and applying structural path analysis, we test two primary hypotheses:

• H1 – Accident severity influences road‑traffic safety more strongly than accident frequency.
  
• H2 – The multi‑indicator model explains a significantly larger proportion of RTS variance compared with the traditional single‑indicator approach.
  

Findings from this research will clarify which components should be prioritized in national and international traffic‑safety strategies and will support the achievement of SDG 3.6 through evidence‑based decision making. The following sections describe the data set, methodological procedures, path‑analysis results, and their policy implications.