Safety Evaluation of Wet-Reflective Pavement Markings

Chapter 6. Before–After Evaluation Results

Aggregate Analysis

Table 9 through table 14 provide the estimates of expected crashes in the after period without treatment, the observed crashes in the after period, and the estimated CMF and its standard error for all crash types considered. Sideswipe-opposite-direction crashes were not analyzed because of the very low number of crashes. The effects for dry-road crashes, which were not specifically evaluated as a target crash type, were inferred from the effects for total and wet-road crashes; they are shown in these tables for information purposes. Results are provided separately for each State as well as all States combined. All results obtained are reported in this section. Recommended CMFs are presented in chapter 8.

The results for North Carolina freeways in table 9 indicate reductions for injury, wet-road, and nighttime wet-road crashes although only the reductions for injury and wet-road crashes were statistically significant at the 95-percent confidence level. All other crash types indicate a slight increase in crashes, but only the increase for dry-road was statistically significant.

Table 9. Results for North Carolina freeways.

The results for Wisconsin freeways in table 10 indicate reductions for total, injury, run-off-road, wet-road, dry-road, and nighttime crashes that are all statistically significant at the 95-percent confidence level. Sideswipe-same-direction and nighttime wet-road crashes had non-statistically significant increases.

Table 10. Results for Wisconsin freeways.

The results for Wisconsin multilane roads in table 11 indicate reductions for total, injury, run-off-road, wet-road, dry-road, and nighttime crashes that were all statistically significant at the 95‑percent confidence level. Sideswipe-same-direction and nighttime wet-road crashes had non-statistically significant increases crashes had non-statistically significant decreases, while nighttime wet-road crashes had negligible and non-statistically significant increases.*

Table 11. Results for Wisconsin multilane roads.

The results for Minnesota two-lane roads in table 12 indicate reductions for total, wet-road, dry-road*, nighttime, and nighttime wet-road crashes, none of which were statistically significant at the 95‑percent confidence level. However, the results for wet-road crashes were statistically significant at the 90-percent confidence level. SInjury, run-off-road, and sideswipe-same-direction crashes* and nighttime wet-road crashes had non-statistically significant increases. For Minnesota two-lane roads, the total numbers of crashes were low, so lack of statistical significance in the analysis results was not unexpected. The indications of reductions in wet, nighttime, and nighttime wet-road crashes do still support the hypothesis that wet-reflective markings reduce these types of crashes.

* Revised 3/12/2019

Table 12. Results for Minnesota two-lane roads.

The results for Minnesota freeways in table 13 indicate reductions for total, injury, sideswipe-same-direction, wet-road, and nighttime crashes but none were statistically significant at the 95‑percent confidence level. The results for wet-road crashes were statistically significant at the 90‑percent confidence level. Run-off-road and nighttime wet-road crashes had non-statistically significant increases. For Minnesota freeways, as with the Minnesota two-lane roads, the total numbers of crashes were low, so the statistical insignificance was not unexpected. However, the indications of reductions in wet and nighttime crashes do still support the hypothesis that wet-reflective markings reduce these types of crashes.

Table 13. Results for Minnesota freeways.

The combined results for all freeway sites (North Carolina, Wisconsin, and Minnesota) in table 14 indicate reductions for total, injury, run-off-road, wet-road, nighttime, and wet-road nighttime crashes, but only those for injury and wet-road crashes were statistically significant at the 95-percent confidence level. The results for sideswipe-same-direction and dry-road crashes showed negligible and non-statistically significant increases for these crash types.

Table 14. Results for combined States freeways.

Disaggregate Analysis

An attempt was made to further analyze the combined freeway dataset for wet-road crashes to identify site characteristics for which the safety benefits were greater. Only wet-road crashes were considered because this was the principal target crash and the only one with a consistent and statistically significant effect in each of the three States. Only freeways were considered because the datasets for multilane and two-way roadways had too few crashes for such an analysis.

A number of variables were investigated, including the following:

• Surface width.
• Shoulder width.
• Area type (urban versus rural).
• Number of lanes.
• Presence of shoulder rumble strip.
• AADT.
• Expected wet-road crash frequency per mi prior to treatment.

No differences or clear trends were seen for any of these variables and the estimated CMFs. Therefore, for this dataset, the expected effect of this strategy on wet-road crashes on freeways was the same regardless of differences in these aspects of the roadway environment.