Identifying Highly Correlated Variables Relating to the Potential Causes of Reportable Wisconsin Traffic Crashes

Qin, Xiao; Shaon, Mohammad Razaur Rahman; Schneider, Robert J; Noyce, David A.; Bill, Andrea; Farid, Yashar Zeinali; Lord, Dominique · 2019 · ROSA P / Wisconsin. Department of Transportation

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This research addresses the rising trend of traffic fatalities in Wisconsin, which increased for three consecutive years leading up to 2017, reversing a long-term decline. The study aims to identify the most pertinent behavioral and engineering variables associated with reportable crashes to inform safety countermeasures. Recognizing that crashes result from a complex confluence of factors, the researchers sought to move beyond single-approach analyses by evaluating crash causes across diverse contexts, including macroscopic area-level trends, microscopic site-specific roadway characteristics, and granular event-oriented driver behaviors. The study employed a three-pronged methodological approach using data from 2015–2017. First, area-level modeling analyzed crash counts at the census tract level, integrating socioeconomic data from the U.S. Census with roadway information from the Wisconsin Information System for Local Roads (WISLR). Second, site-specific modeling utilized random parameters regression on roadway segment data from the Wisconsin State Trunk Network (STN) to account for local variations in highway design and traffic characteristics. Third, event-oriented modeling focused on individual crash events, utilizing Wisconsin traffic citation data to quantify the impact of risky driving behaviors and driver errors. The researchers evaluated over 100 variables across more than a dozen statistical models, including multinomial probit models for driver error and propensity score matching for causal inference regarding specific factors like speed, seatbelt use, and roadway conditions. Key findings include the identification of significant correlations between specific traffic violations and crash occurrences. The analysis of traffic citation data revealed trends in speeding and impairment-related offenses, linking them to jurisdiction-level crash rates. The area-level models highlighted the influence of socio-demographic factors and infrastructure changes on crash frequencies. Site-specific models identified critical roadway design elements that affect safety performance under varying human factors. Furthermore, the driver error models demonstrated that driver error is the primary cause of over 90% of crashes, with specific demographic and behavioral factors significantly influencing error types in both rural and urban settings. The causal inference analysis using propensity score matching provided evidence on the treatment effects of specific conditions, such as the presence of work zones, street lighting, and horizontal curves, on crash outcomes. The significance of this work lies in its comprehensive reference for traffic safety professionals, offering state-of-the-art methodologies for crash prediction and causal analysis. By integrating behavioral data with engineering variables, the study provides a robust framework for identifying high-priority safety issues. The findings support the development of targeted countermeasures, such as improved roadway designs and enforcement strategies, to address the specific causes of crashes. The research underscores the importance of data quality and the integration of diverse data sources to effectively reduce traffic fatalities and severe injuries in Wisconsin.

Key finding

The study successfully identified highly correlated variables relating to crash causes by developing and calibrating crash prediction models for area-level, site-specific, and pedestrian/bicycle contexts while quantifying the impact of risky driving behaviors using traffic citation data.

Methodology

dataset

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. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• incidence prevalence
• sex gender
• demographic disparities
• pre crash contributing factors
• induced exposure
• urban rural setting

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, observational prevalence
• Methodological Resource: dataset resource