Evaluation of Post-Mounted Delineators and Curve Warning Message Markings Phase 1: Designing a Sampling Plan

About the report:

In 2022, the Virginia Transportation Research Council published Guidance for and Effectiveness of Low-Cost Delineation Treatments, a report that evaluated the effectiveness of several low-cost delineation countermeasures (Flintsch et al., 2022a). That report recommended that the Virginia Department of Transportation (VDOT) consider a pilot study of using post-mounted delineators (PMDs) and curve warning message markings (CWMMs). In response, this study’s purpose was to design a sampling plan to evaluate the safety effectiveness of PMDs and CWMMs. The research team conducted a literature review, chose a countermeasure design for PMDs and CWMMs, and designed a statistically robust sampling plan. This study is the first phase of a three-phase effort to evaluate PMDs and CWMMs by deploying them on a sample of VDOT roads. A second phase, a future technical assistance study, will deploy the sampling plan and monitor PMDs and CWMMs in the field following the plan developed in the phase 1 study. A third phase of the study will conduct the before-and-after crash analysis, crash modification factor development, and benefit-cost analysis.

The research team limited the scope to two-lane, rural roads. For PMDs, the scope was limited to roads without guardrails present on the outside of curves because VDOT guardrails typically have delineators already attached to the posts or web at 40- or 80-foot increments. A proportionate stratified sampling approach was evaluated and deemed appropriate for the study. For both countermeasures, a sample size of approximately 100 sites was determined to be sufficient to detect a meaningful treatment effect with adequate statistical power. The reference group of sites identified in this study is appropriate for use with both countermeasures. In both cases, the test of comparability indicated that the odds ratio estimates of crashes in the treatment and reference groups were not statistically significantly different from 1.0. The research team recommended that VTRC plan the phase 2 effort to implement the sampling plan for PMDs and CWMMs.

Authors