Examining injury severity in left turning crashes at intersections
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates the factors influencing injury severity in left-turn crashes at intersections, addressing a gap in traffic safety research where specific crash patterns are often overlooked. Motivated by the high frequency and danger of intersection crashes, particularly involving motorcycles and left-turning vehicles, the authors aim to identify contributory factors—such as human attributes, traffic flow, roadway geometry, and environmental conditions—that affect crash outcomes. The research utilizes police-reported crash data from 163 signalized intersections in Xi’an, China, covering the period from 2005 to 2010. From an initial pool of 1,368 collisions, 391 left-turn crashes were identified, with 317 observations retained for final analysis after accounting for missing data. The methodological approach compares four statistical models: Ordered Logit Model (OLM), Heterogeneous Choice Model (HCM), Generalized Ordered Logit Model (GOLM), and Partial Proportional Odds Model (PPOM). The study argues that traditional ordered models often violate the parallel-lines assumption, which assumes constant coefficients across severity levels. To address this, the authors employ PPOM, which allows certain variables to have varying effects across different injury severity levels. The dependent variable is injury severity, categorized on a five-point ordinal scale ranging from no injury to fatal injury. Independent variables include driver age, alcohol/drug impairment, safety belt usage, vehicle type, impact points, traffic volume, truck percentage, and environmental conditions like weather and lighting. The results indicate that the Partial Proportional Odds Model provides the best fit, outperforming OLM, HCM, and GOLM in terms of pseudo R-squared and Akaike Information Criterion. Key findings reveal that very young drivers (≤18 years) and older drivers (>65 years) are more prone to severe crashes, with very young drivers showing a significant association with higher severity during rush hours and in foggy or rainy conditions. The involvement of trucks significantly increases injury severity, likely due to reduced visibility and capacity. Impact points are critical; front-end collisions for both vehicles result in the most severe injuries due to higher collision forces. Additionally, safety belt usage and alcohol/drug impairment were found to violate the parallel-lines assumption, indicating their effects vary across severity levels. Environmental factors such as ice, snow, and fog also significantly contribute to higher injury severities. The significance of this study lies in its demonstration that PPOM is a superior tool for analyzing crash severity when the parallel-lines assumption is violated by specific variables. The findings provide actionable insights for traffic engineers and policymakers, highlighting the need for targeted safety interventions for vulnerable age groups, restrictions on truck movements at intersections, and improved visibility measures. By accurately identifying factors that were previously underestimated, such as the specific impact of safety belts and alcohol across different severity levels, the study contributes to more precise safety performance evaluations and countermeasure development for left-turn crashes.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | OpenAlex-citations | — | — | 1 | 2026-06-18 |
| archive | success | unpaywall | — | — | 2 | 2026-06-25 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-18 |
| chunk | success | chunk | — | — | 1 | 2026-06-18 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-18 |
| promote | success | — | — | — | 1 | 2026-06-18 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-18 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: crash risk outcomes