Safety at Midblock Pedestrian Signals (2023)

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.

29   Safety Analysis—Findings The focus of the safety analysis was investigating changes in crash frequency by type (all, pedestrians, and rear end) due to the presence of the MPS. The evaluation only considered FI crashes because the California data did not include PDO crashes. Method The evaluation used a cross-sectional observational study approach. While an EB before-after method is commonly used for this type of evaluation, most of the MPSs have been installed for more than 10 years, which limits the availability of high-quality before data. Only a few MPSs have been installed in the past 3–5 years, which was too small of a sample. The team therefore used a cross-sectional study, which is a type of observational study where the outcome and the exposure are assessed at one point or for a short period of time in a sample population (47). The underlying assumption is that all sites should have similar characteristics (e.g., pedestrian volumes and driver behavior). Routinely collected data such as crash, traffic, and geometric data are often used for conducting cross-sectional data analyses. Cross-sectional studies can be conducted faster than before-after studies and are commonly used in traffic safety analysis. For instance, crash-frequency models or safety performance functions in the HSM (24) are developed based on cross-sectional data. Number of Sites for Treated and Control Groups Crash evaluations are beneficial when a control group of similar sites without treatment is identified. Three potential control groups were identified for the evaluation: • Control group 1 (called “All”) comprised all identified nontreated sites. This group included signalized intersections and intersections treated with other pedestrian crossing treatments, such as PHBs or RRFBs, or midblock segments that were not treated. • Control group 2 (called “2-leg”) comprised all nontreated sites with two legs. • Control group 3 (called “2-leg grey”) comprised all nontreated sites with two legs and what the research team termed “grey” pedestrian traffic control devices. The control group included pedestrian traffic control that is not pedestrian activated (e.g., crosswalk pavement markings and signs). Table 20 provides the number of sites by state and by type of traffic control present at the pedestrian crossing for the treated and control group 1 sites. Table 21 shows the number of sites in control group 2, and Table 22 shows the number of sites in control group 3. C H A P T E R 5

30 Safety at Midblock Pedestrian Signals Treated or Control Control Typea CA TX UT Total Treated MPS 150 11 32 193 Control group 1, All CBoverhead-24/7 0 1 0 1 Control group 1, All CBoverhead-PedAct 2 0 2 4 Control group 1, All CBroadside-24/7 4 0 0 4 Control group 1, All CBroadside-PedAct 1 1 4 6 Control group 1, All CW&Sign 61 18 19 98 Control group 1, All CW_only 1 0 0 1 Control group 1, All LED-Em 0 2 2 4 Control group 1, All LED-Em & Flags 0 0 4 4 Control group 1, All NoPedTCD 0 32 2 34 Control group 1, All PHB 0 4 10 14 Control group 1, All RRFB 23 2 2 27 Control group 1, All RRFB-Overhead 9 0 0 9 Control group 1, All Signal 198 220 61 479 Control group 1, All Stop-AllWay 3 0 0 3 Control group 1, All Stop-Cont 7 0 0 7 Control group 1, All Stop-ContwCB 1 0 0 1 Control group 1, All Stop-OneWayTraffic 3 0 0 3 Control subtotal 313 280 106 699 Grand total 463 291 138 892 a See Table 15 for descriptions of control type. Treated or Control Control Typea CA TX UT Total Treated MPS 150 11 32 193 Control group 2, 2-leg CBoverhead-24/7 0 1 0 1 Control group 2, 2-leg CBoverhead-PedAct 2 0 2 4 Control group 2, 2-leg CBroadside-24/7 4 0 0 4 Control group 2, 2-leg CBroadside-PedAct 1 1 4 6 Control group 2, 2-leg CW&Sign 60 18 19 97 Control group 2, 2-leg CW_only 1 0 0 1 Control group 2, 2-leg LED-Em 0 2 2 4 Control group 2, 2-leg LED-Em & Flags 0 0 4 4 Control group 2, 2-leg NoPedTCD 0 32 0 32 Control group 2, 2-leg PHB 0 4 9 13 Control group 2, 2-leg RRFB 10 2 0 12 Control group 2, 2-leg Stop-Cont 7 0 0 7 Control group 2, 2-leg Stop-ContwCB 1 0 0 1 Control group 2, 2-leg Stop-OneWayTraffic 3 0 0 3 Control subtotal 89 60 40 189 Grand total 239 71 72 382 a See Table 15 for descriptions of control type. Treated or Control Control Type a CA TX UT Total MPS RedDevice 150 11 32 193 Control group 3, 2-leg grey CW&Sign 60 18 19 97 Control group 3, 2-leg grey CW_only 1 0 0 1 Control group 3, 2-leg grey NoPedTCD 0 32 0 32 Control subtotal 61 50 19 130 Grand total 211 61 41 323 a See Table 15 for descriptions of control type. Table 20. Number of sites in treated and control group 1 (All). Table 21. Number of sites in treated and control group 2 (2-leg). Table 22. Number of sites in treated and control group 3 (2-leg grey control).

Safety Analysis—Findings 31 Models Preliminary analyses were conducted using the variables listed in Table 16. After preliminary analyses, some of the variables were eliminated because they were consistently not significant (e.g., type of median treatment) or were refined (e.g., creating posted speed limit groups). Table 23 lists the variables used in the safety analysis. Most of the MPSs identified were included in the evaluation. One site was removed from consideration during model development because it was frequently identified as an outlier. The site was on a surface street under several freeway overpasses. The 250-ft radius initially used to identify crashes was adjusted to a 150-ft radius; however, freeway crashes were suspected to still be in the crash count. Because of the additional challenges in determining which crashes were on the surface street and which were on the freeway, the research team decided instead to remove the site from the evaluation. For model development, the research team used negative binomial regression and first examined various combinations of variables. In some cases, a variable had to be regrouped during the analysis; for example, the posted speed limit along the major street (i.e., the street with the MPS) was regrouped into below 30 mph (yes or no). The model form presented reflects the findings from several preliminary regression analyses. The predicted crash frequency is calculated in Equations 1–3. E[C] = y × fst × eb0+badtADT+badpADP+blegslegs+bowIow+bpslIpsl+bmlanesMLanes+bpkIpk+bbkIbk × CMFtrt (1) with fst = ebcaCA+butUT (2) CMFtrt = ebtrtItrt (3) where E[C] = Predicted annual average crash frequency y = Number of years of crash data fst = State indicator variable CA = California indicator variable (= 1 if site is in CA; = 0 otherwise) UT = Utah indicator variable (= 1 if site is in UT; = 0 otherwise) ADT = Average daily traffic, vehicles per day ADP = Average daily pedestrian volume, pedestrians per day legs = Number of intersection legs Iow = One-way street indicator variable (= 1 if one-way street; = 0 otherwise) Ipsl = Indicator variable for posted speed limit (= 1 if < 30 mph; = 0 otherwise) MLanes = Number of lanes on the major street Ipk = Indicator variable for the parking on the road (= 1 if present; = 0 otherwise) Ibk = Indicator variable for the bike lane (= 1 if present; = 0 otherwise) CMFtrt = Crash modification factor for the treatment (i.e., MPS) Itrt = Indicator for the MPS (= 1 if present; = 0 otherwise) bj = Calibrated coefficients

32 Safety at Midblock Pedestrian Signals Variable Name Model Coefficient Description ADP badp Average daily pedestrians for site (total of all approaches) ADT badt Average daily traffic for the site (total of all approaches) BikeLane bbk Bike lane is present (yes or no) CA bca Site is in California (yes or no) CrossDist bdist Crossing distance Legs blegs Number of legs (2, 3, or 4) OneWay bow One-way vehicle operations present for 2-leg intersection (yes or no) OnStreetParking bpk On-street parking is present (yes or no) PostedSpd<30 bpsl Posted speed limit is less than 30 mph (yes or no) Treatment btrt MPS treatment present (yes or no) UT but Site is in Utah (yes or no) NumLanesMaj bm_lanes Number of lanes on the major street Table 23. Roadway variables used in safety analysis. Findings Pedestrian Fatal and Injury Crashes Table 24 provides the estimated regression coefficients for pedestrian FI crashes for the three control groups. As expected, both vehicle volume and pedestrian volume were signi ficant, with increasing activity being associated with more pedestrian crashes. A greater number of legs at the intersection was also associated with more pedestrian crashes when considering control group 1, which included intersections with two, three, and four legs along with all types of pedestrian traffic control. When focusing on sites with two legs, the presence of on-street parking was associated with fewer pedestrian crashes. After experimenting with different approaches for considering posted speed limit, the decision was to group that variable into those sites with 30 mph and more on the major road and those sites with less than 30 mph. For all control groups, the variable was significant, with a negative coefficient indicating that when the posted speed limit was 25 mph and less, fewer pedestrian crashes were present. A concern expressed by members of this project’s panel and by the National Committee on Uniform Traffic Control Devices (NCUTCD) Signal Technical Committee while review- ing these findings was the potential influence of a nearby driveway. The research team addressed this concern by developing additional models to explore the impacts of distance between the nearest driveway and the marked crosswalk. The treated sites were grouped as follows: • Treated group 1 (also called “All Treated Sites”) comprised all identified treated sites. These sites had driveways as close as 10 ft to the center of the MPS marked crosswalk. None of the MPS sites had a driveway that was closer than 10 ft to the crosswalk. • Treated group 2 (called “MPS with No Driveway within 50 ft” or “> 50 ft”) comprised all treated sites where the closest driveway was 50 ft away or more. • Treated group 3 (called “MPS with No Driveway within 100 ft” or “> 100 ft”) comprised all treated sites where the closest driveway was 100 ft away or more. Table 25 shows the results for the two additional models that used treated groups 2 and 3. These models considered pedestrian FI crashes along with the 2-leg grey control sites. The treated sites were restricted to those sites with either no driveway within 100 ft or no driveway within 50 ft of the marked crosswalk. The MPS treatment variable was significant (p < 0.05) for the model that included sites where driveways were at least 50 ft away from the marked cross- walk and marginally significant (0.05 < p < 0.1) for the model where the nearest driveway was

Model Coefficient, Variable Name • Ped FI Crashes • All Control Sites • All Treated Sites • Ped FI Crashes • 2-Leg Control Sites • All Treated Sites • Ped FI Crashes • 2-Leg Grey Control Sites • All Treated Sites badp, ADP 0.2225 0.4814 (< 0.0001) (< 0.0001) badt, ADT 0.4287 0.684 0.6486 (< 0.0001) (< 0.0001) (< 0.0001) bca, California −0.9175 −0.8201 −0.7996 (< 0.0001) (0.0027) (0.0076) blegs, Legs 0.3674 0 0 (< 0.0001) 0 0 bm_lanes, NumLanesMaj 0 0 0 0 0 0 bow, OneWay −0.6475 0 0 (0.0222) 0 0 bpk, OnStreetParking 0 −0.6433 −0.6478 0 (0.0012) (0.0028) bpsl, PostedSpd<30 −0.4138 −0.7583 −0.8132 (< 0.0001) (0.0013) (0.0015) btrt, Treatment −0.3131 −0.6423 −0.8083 (0.0511) (0.0054) (0.0015) but, Utah −1.0961 0 0 (< 0.0001) 0 0 b0 −7.3823 −10.1277 −9.8365 (< 0.0001) (< 0.0001) (< 0.0001) k0 0.3157 0.5199 0.4633 (< 0.0001) (0.0121) (0.0244) Number of treated and control sites in model (sites removed in outlier analysis) 889 sites (1 treated and 2 control sites removed) 381 sites (1 treated site removed) 322 sites (1 treated site removed) Notes: See Table 23 for descriptions of the variables. b0 = intercept and k0 = dispersion parameter. The coefficient 0 denotes that the corresponding variable was excluded from the model. Cells are highlighted in light grey with italic text when the p-value is between 0.05 and 0.1. Cells are highlighted in dark grey with bold white text when the p-value is less than 0.05. 0.4495 (< 0.0001) Model Coefficient, Variable Name • Ped FI Crashes • 2-Leg Grey Control Sites • MPS, No Driveway within 100 ft • Ped FI Crashes • 2-Leg Grey Control Sites • MPS, No Driveway within 50 ft badp, ADP 0.5142 0.4919 (< 0.0001) (< 0.0001) badt, ADT 0.6274 0.7355 (< 0.0001) (< 0.0001) bca, California −0.9178 −0.8692 (0.0048) (0.0057) bpk, OnStreetParking −0.5878 −0.644 (0.0011) (0.0048) bpsl, PostedSpd<30 −1.0343 −0.7728 (0.0011) (0.0077) btrt, Treatment −0.5192 −0.5909 (0.0701) (0.0274) b0 −9.7298 −10.725 (< 0.0001) (< 0.0001) k0 0.3846 0.3700 (0.056) (0.0522) Number of treated and control sites in model (sites removed in outlier analysis) 219 sites: 130 control and 89 treated (1 treated site removed) 250 sites: 130 control and 120 treated (2 treated sites removed) Notes: See Table 23 for descriptions of the variables. b0 = intercept and k0 = dispersion parameter. Cells are highlighted in light grey with italic text when the p-value is between 0.05 and 0.1. Cells are highlighted in dark grey with bold white text when the p-value is less than 0.05. Table 24. Estimated regression coefficient (and p-values) for pedestrian FI crashes by control group. Table 25. Estimated regression coefficient (and p-values) for pedestrian FI crashes by treated site groups revised with consideration of distance to driveways.

34 Safety at Midblock Pedestrian Signals within 100 ft. The smaller sample size for treated group 3 (MPS with No Driveway within 100 ft) may be affecting the finding. In other words, an MPS provides a safety benefit for pedestrians regardless of the distance to the driveway. All Fatal and Injury Crashes Table 26 provides the estimated regression coefficients for all FI crashes for the three control groups along with a subset of MPS sites where the nearest driveway of any type is within 50 ft. Both volume variables—the number of vehicles entering and the number of pedestrians at the intersection—were significant for all control groups considered. Only for the model that considered all treated sites and 2-leg grey control sites was the treatment variable significant. As regards all crashes, the research showed a safety benefit from the MPS when considering all MPS sites (including those with driveways 10 ft or more from the crossing); however, when restricting the MPS sites to those with driveways at least 50 ft from the MPS, the crash reduction was not statistically significant, perhaps due to the limited sample size. In other words, the installation of an MPS has safety benefits for all users (vehicles, pedestrians, etc.) when Model Coefficient, Variable Name • FI Crashes • All Control Sites • All Treated Sites • FI Crashes • 2-Leg Control Sites • All Treated Sites • FI Crashes • 2-Leg Grey Control Sites • All Treated Sites • FI Crashes • 2-Leg Grey Control Sites • MPS with No Driveway within 50 ft badp, ADP 0.06096 0.3487 0.3582 0.3977 (0.0122) (< 0.0001) (< 0.0001) (< 0.0001) badt, ADT 0.6655 0.8905 0.8381 0.9588 (< 0.0001) (< 0.0001) (< 0.0001) (< 0.0001) bca, CA −0.8529 −1.4866 −1.2388 −1.3024 (< 0.0001) (< 0.0001) (< 0.0001) (< 0.0001) blegs, Legs 0.6391 0 0 0 (< 0.0001) 0 0 0 bm_lanes, NumLanesMaj 0.07914 0 0 0 (0.0043) 0 0 0 bow, OneWay −0.1922 −0.5341 −0.5035 −0.3611 (0.0586) (0.0157) (0.024) (0.179) bpk, OnStreetParking 0 −0.3828 −0.3900 −0.3793 0 (0.0161) (0.0186) (0.0398) bpsl, PostedSpd<30 −0.4172 −0.3339 −0.4117 −0.3481 (< 0.0001) (0.0567) (0.029) (0.1179) btrt, Treatment −0.1153 −0.1752 −0.4158 −0.3506 (0.3017) (0.3049) (0.0387) (0.1278) but Utah −0.7588 −0.6616 −0.5256 −0.6403 (< 0.0001) (0.0065) (0.058) (0.0433) b0 −8.3403 −10.0365 −9.5408 −10.872 (< 0.0001) (< 0.0001) (< 0.0001) (< 0.0001) k0 0.4872 0.7245 0.7045 0.6671 (< 0.0001) (< 0.0001) (< 0.0001) (< 0.0001) Number of treated and control sites in model (sites removed as part of outlier analysis) 891 sites (1 control site removed) 380 sites (2 control sites removed) 321 sites (2 control sites removed) 250 sites: 130 control and 120 treated (2 treated sites removed) Notes: See Table 23 for descriptions of the variables. b0 = intercept and k0 = dispersion parameter. The coefficient 0 denotes that the corresponding variable was excluded from the model. Cells are highlighted in light grey with italic text when the p-value is between 0.05 and 0.1. Cells are highlighted in dark grey with bold white text when the p-value is less than 0.05. Table 26. Estimated regression coefficient (and p-values) for all FI crashes by control group and by subset of MPS sites.

Safety Analysis—Findings 35 there is a driveway 10 ft or more away, and no safety disbenefit was found when there is no driveway within 50 ft. Rear-End Fatal and Injury Crashes Table 27 provides the estimated regression coefficients for RE crashes for the three control groups. The ADT was always significant, while pedestrian counts (average daily pedestrians, ADP) were significant for all crashes except RE crashes. Members of the NCUTCD Signal Technical Committee asked whether the findings for RE crashes would be different if PDO crashes were considered. Members noted that many RE crashes tend to be PDO crashes. Because PDO crashes are not available for the California sites, the research team reviewed the data for all severity level crashes using only Texas and Utah data. Table 28 provides the model results. As shown in Table 27, treatment was found to be significant when considering all control sites, indicating that the MPS is associated with a decrease in FI RE crashes. When using all control sites in only Texas and Utah along with all severity level crashes (see Table 28), treatment is not significant. Treatment is not significant for any of the control groups when considering PDO crashes. Model Coefficient, Variable Name • RE FI Crashes • All Control Sites • All Treated Sites • RE FI Crashes • 2-Leg Control Sites • All Treated Sites • RE FI Crashes • 2-Leg Grey Control Sites • All Treated Sites badp, ADP 0 0.1759 0.1757 0 (0.0578) (0.0965) badt, ADT 0.7918 0.8109 0.7092 (< 0.0001) (< 0.0001) (< 0.0001) bca, CA 0.5275 0 0 (< 0.0001) 0 0 blegs, Legs 0.3873 0 0 (< 0.0001) 0 0 bm_lanes 0.1243 0 0 (0.0009) 0 0 bow, OneWay 0 −1.0243 −0.9767 0 (0.0275) (0.0346) bpk 0 0 −0.443 0 0 (0.0544) bpsl −0.6003 −0.7051 −0.6829 (< 0.0001) (0.0105) (0.0265) btrt −0.3768 −0.2588 −0.1303 (0.0364) (0.2842) (0.6624) but 0.9135 0.9524 0.6693 (< 0.0001) (0.0002) (0.0335) b0 −11.3643 −11.0553 −9.8592 (< 0.0001) (< 0.0001) (< 0.0001) k0 0.5938 0.859 0.7223 (< 0.0001) (0.002) (0.0086) Number of treated and control sites in model (sites removed in outlier analysis) 893 sites (no site removed) 381 sites (1 control site removed) 322 sites (1 control site removed) Notes: See Table 23 for descriptions of the variables. b0 = intercept and k0 = dispersion parameter. The coefficient 0 denotes that the corresponding variable was excluded from the model. Cells are highlighted in light grey with italic text when the p-value is between 0.05 and 0.1. Cells are highlighted in dark grey with bold white text when the p-value is less than 0.05. Table 27. Estimated regression coefficient (and p-values) for RE FI crashes by three control groups.

36 Safety at Midblock Pedestrian Signals Crash Modification Factors The CMF for the MPS is determined in Equation 4: CMFtrt = ebtrt (4) The various models, along with control groups and treated groups, resulted in several cases where the treatment variable is significant. These results can be considered for developing a CMF for the MPS treatment. Table 29 summarizes the results from this research for CMF consideration by crash type, control group, and treated group. The research team selected a recommended MPS CMF for each crash type. For the crash type of all FI crashes and RE FI crashes, only one of the control groups resulted in a statistically significant CMF. For all FI crashes, the control group was the sites with two legs and nonactive pedestrian treatments (i.e., grey). For the RE FI crashes, the 2-leg grey control Model Coefficient, Variable Name • Texas and Utah only • RE All Severity Crashes • All Control Sites • All Treated Sites • Texas and Utah only • RE All Severity Crashes (TX and UT) • 2-Leg Control Sites • All Treated Sites • Texas and Utah only • RE All Severity Crashes (TX and UT) • 2-Leg Grey Control Sites • All Treated Sites badp, ADP 0.0017 0.0465 −0.0036 (0.9665) (0.7216) (0.9800) badt, ADT 0.8581 0.8190 0.8596 (< 0.0001) (0.0007) (0.002) bbk, BikeLane 0.0456 0.1741 −0.0171 (0.7052) (0.5366) (0.9609) bdist, CrossDist 0.0056 0.0005 0.0023 (0.1292) (0.9504) (0.8051) blegs, Legs 0.3428 0 0 (< 0.0001) 0 0 bm_lanes 0.0972 0.0703 0.0030 (0.1241) (0.5999) (0.9849) bow, OneWay −0.1159 −0.2593 −0.3993 (0.6580) (0.7106) (0.6188) bpk −0.3841 0.0543 −0.3523 (0.0131) (0.8659) (0.3236) bpsl −0.7745 −0.7309 −0.4002 (0.0001) (0.0468) (0.3909) btrt −0.1887 −0.1617 −0.1725 (0.4102) (0.5911) (0.6282) but 0.8068 0.7429 0.9104 (< 0.0001) (0.0579) (0.0528) b0 −11.0212 −9.9169 −9.8773 (< 0.0001) (< 0.0001) (< 0.0001) k0 0.5715 0.6681 0.6220 (< 0.0001) (0.0004) (0.0018) Number of treated and control sites in model (sites removed in outlier analysis) 429 sites (no site removed) 143 sites (1 control site removed) 112 sites (1 control site removed) Notes: See Table 23 for descriptions of the variables. b0 = intercept and k0 = dispersion parameter. The coefficient 0 denotes that the corresponding variable was excluded from the model. Cells are highlighted in light grey with italic text when the p-value is between 0.05 and 0.1. Cells are highlighted in dark grey with bold white text when the p-value is less than 0.05. Table 28. Estimated regression coefficient (and p-values) for RE FI crashes by control group for all severity level crashes in TX and UT.

Safety Analysis—Findings 37 group did not produce a statistically significant coefficient, but the control group that included all comparison sites did. More than one of the models for pedestrian FI crashes resulted in a statistically significant coefficient, and the research team recommends the model that considered those MPSs where a driveway was not within 50 ft of the marked crosswalk. Comparison with Other Studies The MPS has been compared to the PHB and the half signal as an appropriate treatment for a marked pedestrian crossing. Chapter 2 summarizes the PHB and half-signal treatment characteristics. Two previous research studies on the PHB have generated recommended CMFs. Table 30 provides a comparison of the CMFs developed for the PHB from an NCHRP study published in 2017 (11) and from an ADOT study published in 2019 (12), along with the CMFs suggested from the research in this study. Table 31 summarizes the characteristics of the treated and control sites included in the two PHB studies, along with this MPS study. Crash Type Control Group Treated Group # Sites in Modela Comments MPS Coefficient CMF FI 699 sites All 193 MPS sites 891 sites Model included all sites identified in the study −0.1153 Not significant FI 189 sites 2-leg 193 MPS sites 380 sites Model included only 2-leg sites −0.1752 Not significant FI 130 sites 2-leg grey 193 MPS sites 322 sites Model included 2-leg sites where the control sites had nonactive pedestrian treatments −0.4158 0.660 FI 130 sites 2-leg grey 122 MPS sites 250 sites Model included MPS sites where the nearest driveway was more than 50 ft from the marked crosswalk −0.3506 Not significant Ped FI 699 sites All 193 MPS sites 889 sites Model included all sites identified in the study −0.3131 Not significant Ped FI 189 sites 2-leg 193 MPS sites 381 sites Model included only 2-leg sites −0.6423 0.526 Ped FI 130 sites 2-leg grey 193 MPS sites 322 sites Model included 2-leg sites where the control sites had nonactive pedestrian treatments −0.8083 0.446 Ped FI 130 sites 2-leg grey 122 MPS sites 250 sites Model included MPS sites where the nearest driveway was more than 50 ft from the marked crosswalk −0.5909 0.554 Ped FI 130 sites 2-leg grey 90 MPS sites 219 sites Model included MPS sites where the nearest driveway was more than 100 ft from the marked crosswalk −0.5192 Marginal significant RE FI 699 sites All 193 MPS 893 sites Model included all sites identified in the study −0.3768 0.686 RE FI 189 sites 2-leg 193 MPS sites 382 sites Model included only 2-leg sites −0.2588 Not significant RE FI 130 sites 2-leg grey 193 MPS sites 324 sites Model included 2-leg sites where the control sites had nonactive pedestrian treatments −0.1303 Not significant a Number of sites in model after removing outliers. Shading indicates the recommended CMF for the given crash type. Table 29. Potential CMFs for the MPS treatment.

38 Safety at Midblock Pedestrian Signals Crash Type PHB CMF 2017 NCHRP Study (11) PHB CMF 2019 ADOT Study (12) MPS CMF NCHRP 3-141 (this study) All crash types, all severity levels 0.820 0.818 Not generated All crash types, FI only Not generated 0.748 0.660 Pedestrian crashes, all severity levels 0.432 0.543 Not generated Pedestrian crashes, FI Not generated 0.550 0.554 Rear-end crashes, all severity levels Not generated 0.795 Not generated Rear-end crashes, FI Not generated 0.714 0.686 Rear-end and sideswipe crashes, all severity levels 0.876 Not generated Not generated Sites PHB CMF 2017 NCHRP Study (11) PHB CMF 2019 ADOT Study (12) MPS CMF NCHRP 3-141 (this study) Number of treated sites 27 PHB with advance stop markings and signs sites 52 PHB sites 193 MPS sites (used for all or RE CMFs) 122 MPS sites (used for Ped CMF) Treated sites— number of legs distribution Not provided 2 legs = 11 sites (21%) 3 legs = 17 sites (33%) 4 legs = 24 sites (46%) 2 legs = 193 (100%) 2 legs = 122 (100%) Treated sites— posted speed limit distribution Not provided 35 mph or less = 22 sites (42%) 40–45 mph = 30 sites (57%) 193 sites 35 mph or less = 179 sites (93%) 40–45 mph = 14 sites (7%) 122 sites 35 mph or less = 115 sites (94%) 40–45 mph = 7 sites (6%) Control group 3,129 crashes at 287 sites in Charlotte, Portland, Phoenix, Scottsdale, Tucson, and St. Petersburg that did not have the following treatments: PHB, RRFB, or refuge island 101 unsignalized 56 signalized 130 pedestrian crossings with sign and markings, only markings, or neither signs nor markings Control sites— number of legs distribution Not provided Unsignalized: 3 legs = 42 (42%) 4 legs = 59 (51%) Signalized: 3 legs = 5 (9%) 4 legs = 51 (91%) Unsignalized: 2 legs = 130 (100%) Control sites— posted speed limit distribution Not provided Unsignalized: 35 mph or less = 41 sites (42%) 40–45 mph = 60 sites (59%) Signalized: 35 mph or less = 24 sites (43%) 40–45 mph = 32 sites (57%) Unsignalized: 35 mph or less = 111 sites (85%) 40–45 mph = 19 sites (15%) Table 30. Comparison of CMFs for PHB and MPS pedestrian treatments by crash type. Table 31. Comparison of treated and control site characteristics used to generate CMFs for pedestrian treatments in Table 30.

Safety Analysis—Findings 39 In general, the CMFs for the MPS developed in this study are similar to the CMFs identified for the PHB. A needed caution when comparing the CMFs for the MPS and the PHB is the difference in characteristics of the sites in each study. The MPS sites all have two legs, while only 21% of the PHB sites in the ADOT study have two legs. The MPS is appropriate for locations with only two legs (midblock), while the PHB is appropriate for locations with three or four legs. Almost all the MPS sites (93%) had 35 mph or lower posted speed limits. For the PHB sites in the ADOT study, only 42% were on 35 mph or lower posted speed limit roads, with the majority (57%) being on roads with 40 or 45 mph posted speed limits.