Guidance for the Design and Application of Shoulder and Centerline Rumble Strips (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

170 A P P E N D I X H This appendix presents the SPF results developed for two- lane roads based on all nontreatment sites (i.e., sites without centerline rumble strips: BA-No RS). The same SPFs apply to roads classified as urban and rural and to tangent and curved roadways. Each line in each table provides the regression coef- ficients and their precision estimates for a given SPF where one was estimated. For example, using Table H-1 for two-lane roads in Pennsylvania: where: SPD = 0 if the posted speed limit is ≥ 55; otherwise 1 WIDTH = roadway width (ft) Expected total crashes/mi/yr exp= − +5 59 0 58. . ln . . . ADT RHR SPD WIDTH e AD–5.59 ( + + + ) = × 0 10 0 30 0 02 T e e e0.55 RHR SPD WIDTH× × ×0 10 0 30 0 02 1. . . ( ) SPF Results for TOT, FI, and SSOD Crashes on Selected Roadways Without Centerline Rumble Strips An empty cell in a row where an SPF was estimated indi- cates the corresponding regression coefficient was not statis- tically significant at the 0.15 level or the coefficient’s sign was not of the expected direction. For some analyses, the sample size was not sufficient to estimate an SPF directly from data. An approximation was employed in which a factor was applied to the SPF for total crashes in Table H-1. This factor is the ratio of the sum of all relevant crashes at the NT sites to the sum of all (total) crashes. The dispersion parameter is not adjusted, i.e., the one for total crashes applies for all crash type subsets for which SPFs are so approximated.

Intercept lnADT RHR SPD WIDTH Dispersion State Number of sites Estimate SEa Estimate SE p–valueb Estimate SE p–value Estimate SE p–value Estimate SE p– value Estimate SE R2LR MN 244 -7.31 0.88 0.81 0.09 <.0001 0.22 0.09 0.022 0.74 0.16 30.25% WA 228 -7.46 1.20 0.95 0.13 <.0001 0.57 0.12 46.37% PA 603 -5.59 0.42 0.58 0.04 <.0001 0.10 0.04 0.005 0.30 0.06 <.0001 0.02 0.005 0.0007 0.29 0.03 91.31% a SE: standard error of estimate. b p–value: significance level. Intercept lnADT RHR SPD WIDTH Dispersion State Number of sites Estimate SEa Estimate SE p–valueb Estimate SE p–value Estimate SE p–value Estimate SE p– value Estimate SE R2LR MN 244 A factor of 0.41 is applied to the SPF for total crashes in Table H -1 WA 228 A factor of 0.44 is applied to the SPF for total crashes in Table H -1 PA 603 -5.55 0.47 0.55 0.06 <.0001 0.31 0.07 <.0001 0.02 0.006 0.0005 0.36 0.04 86.91% a SE: standard error of estimate. b p–value: significance level. Intercept lnADT RHR SPD WIDTH Dispersion State Number of sites Estimate SEa Estimate SE p–valueb Estimate SE p–value Estimate SE p–value Estimate SE p– value Estimate SE R2LR MN 244 A factor of 0.12 is applied to the SPF for total crashes in Table H -1 WA 228 A factor of 0.05 is applied to the SPF for total crashes in Table H-1 PA 603 -7.78 0.95 0.73 0.11 <.0001 0.26 0.13 0.044 -0.02 0.01 0.10 0.60 0.03 39.36% a SE: standard error of estimate. b p–value: significance level. Table H-1. SPF results for TOT crashes on two-lane rural roads based on all nontreatment sites. Table H-2. SPF results for FI crashes on two-lane rural roads based on all nontreatment sites. Table H-3. SPF results for SSOD crashes based on all nontreatment sites.