Indirect methods to account for exposure in highway safety studies

Pendleton, Olga J. · 1996 · ROSA P / United States. Department of Transportation. Federal Highway Administration

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 report addresses the complex challenge of measuring exposure—the numerical opportunity for an accident to occur—in highway safety evaluations. While exposure is straightforward for roadway sections (e.g., vehicle-miles traveled) or intersections (e.g., annual average daily traffic), it is often undefined or unavailable for driver- or vehicle-based analyses, such as assessing older drivers in left-turn accidents. Due to these difficulties, the paper provides an instructional guide on indirect methods that treat exposure as a surrogate, allowing researchers to evaluate safety treatments without direct exposure data. The document details three primary indirect methodologies: before/after evaluation designs, case-control studies, and induced-exposure methodologies. For each method, the report provides step-by-step numerical formulas, critiques of advantages and disadvantages, and specific applications. The before/after designs are categorized into four types: simple before/after without a comparison group; before/after with a comparison group or condition using single time measurements; designs with multiple time measurements to test for comparability; and designs with replication. The text emphasizes that simple before/after tests are weak due to regression to the mean and lack of control for confounding factors. More robust designs utilize comparison groups (matched "yoked" sites or unmatched "reference" sites) or comparison conditions (e.g., daytime accidents serving as a control for nighttime accidents) to isolate treatment effects. The findings are illustrated through numerical examples, primarily using data from a Michigan study on raised pavement markers. In the simple before/after analysis, the study found a non-significant 0.53 percent reduction in accidents. When employing a comparison group (reference sites), the estimated reduction in nighttime accidents was 2.2 percent, which was also statistically insignificant. Using daytime accidents as a comparison condition yielded a 6.5 percent reduction in nighttime accidents, but this too failed to reach statistical significance. The report demonstrates how to calculate odds ratios and likelihood ratio chi-square statistics to determine significance, noting that valid conclusions depend on passing comparability tests when multiple years of data are available. The significance of this work lies in its role as a comprehensive guide for highway safety researchers. By providing clear computational procedures and critical assessments of each method’s appropriateness, the report enables researchers to design and evaluate safety studies effectively when direct exposure data are lacking. It highlights the importance of selecting valid comparison groups or conditions to avoid biases such as regression to the mean, thereby improving the reliability of safety treatment evaluations.

Key finding

The report provides instructional formulas and examples for before/after, case-control, and induced-exposure methodologies to handle exposure issues in highway safety studies when direct measures are unavailable.

Methodology

review

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).

StageOutcomeToolModelPromptAttemptsCompleted
discover success rosap 2 2026-05-23
archive success 1 2026-05-23
extract success cached 2 2026-06-10
clean success 1 2026-06-01
chunk success 1 2026-06-01
embed success 1 2026-06-02
enrich success 1 2026-05-23
promote success 1 2026-05-23
summarize success llm qwen3.6-27b-prismaquant summ-v5 3 2026-06-10
tag success vector_similarity 19 2026-06-11
verify partial 2 2026-06-10

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

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).