USF Libraries
USF Digital Collections

Evaluation of highway safety needs of special population groups, Phase 1


Material Information

Evaluation of highway safety needs of special population groups, Phase 1
Physical Description:
University of Tennessee, Knoxville. Transportation Center
Southeastern Transportation Center (U.S.)
University of South Florida. Dept. of Civil and Environmental Engineering
University of South Florida. Center for Urban Transportation Research
Center for Urban Transportation Research (CUTR)
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Aged automobile drivers--United States   ( lcsh )
Roads--United States--Safety measures   ( lcsh )
Traffic safety--United States   ( lcsh )
Teenage automobile drivers--United States   ( lcsh )
Automobile drivers with disabilities--United States   ( lcsh )
letter   ( marcgt )

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 42714385
usfldc doi - C01-00161
usfldc handle - c1.161
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader ntm 22 Ka 4500
controlfield tag 008 s flunnn| ||||ineng
datafield ind1 8 ind2 024
subfield code a C01-00161
(OCoLC) 42714385
0 245
Evaluation of highway safety needs of special population groups, Phase 1
Tampa, Fla
b Center for Urban Transportation Research (CUTR)
c 1998 October
Aged automobile drivers--United States
Roads--United States--Safety measures
Traffic safety--United States
Teenage automobile drivers--United States
Automobile drivers with disabilities--United States
2 710
University of Tennessee, Knoxville. Transportation Center.
Southeastern Transportation Center (U.S.)
University of South Florida. Dept. of Civil and Environmental Engineering.
University of South Florida. Center for Urban Transportation Research.
Dissanayake, Sunanda.
1 773
t Center for Urban Transportation Research Publications [USF].
4 856


Digitization Note : Original print copy was poor quality EVALUATION OF HIGHWAY SAFETY N EEDS OF SPECIAL POPULATION GROUPS Phase I


A REPORT PRESENTED TO THE SOUTHEAS1ERN TRANSPORTATION CENTER UNIVERSITY OF 1ENNESSBE ON EVALUATIONOFIDGHWAYSAFETYNEEDSOF SPECIAL POPULATION GROUPSPhase I By Sunanda Dissanayake, Research Assistant, J. John Lu, Ph.D., P.E., Assistant Professor, Hua Tan, Research Assistant Department of Civil and Environmental Engineering and Xuehao Chu, Ph.D., Research Associate, Patricia turner, Research Associate Center for Urban Transportation Research University of South Florida T8lllpa, Florida 33620 October 1998


ABSTRACI' Although the United S t ates has achieved dramatic progress in the past several decades in tenn.s of highway safety, there is however still room for improvement. Each year, tens .of thousands of people are killed and millions injured as a result of highway crashes, which in tum pose considerable economic loss to society. One area of highway safety that deserves further consideration is in relation to special population groups. This projec t concentrat es on the highway safety needs of six special population groups: older drivers, school aged children, young drivers, i.titemational tourists, ne w immigrants and people with disabilities. Ranking of highway safety importance was developed among the selected groups and critical highway safety issues and concerns were identified for each of the special population groups. These two basic tasks were accomplished by conducting two statewide surveys in which transportation professionals in the highway safety area vjere consulted and their views concerning !h e critical highway safety issues of the selected special population groups were identified. A multi criteria decision making approach was used in priorit izing the identified issues an

AC:KNOWLEDGEMENT This project was sponsored by tb.e Southeastern Transportation Center (STC) at the University ofTellllessee, which was established under the University Transportation Centers Program of U.S. Department of Transportation. The assistance and cooperation provided by the members of STC is greatly appreciated. The authors would also like to express their sincere appreciation to Matthew Wey for the support given in editing the report.


. TABLE OF CONTENTS ABSTRACT ACKNOWLEDGEI\IIENT TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES CHAPTER I. INTRODUCTION 1.1 Background 1.2 Research Statement 1.3 Research Pwposes and Objectives 1.4 Outlin e of the Report CHAPTER 2. SELECTED GROUPS AND THEIR SIGNIFICANCE 2.1 The Elderly 2.2 Intemational Tourists 2.3 School Aged Children .2.4 Young Drivers 2.5 People with Disabilities 2.6 New Immigrants CHAPTER 3. ME1HODOLOGY 3.1 Introduction 3.2 Projection Models 3.3 Suryeys 3.4 Ranking Model iv ii 111 iv vii X I I 4 s 6 7 7 9 11 12 13 14 15 1 5 15 .16 18


CHAPTER 4. PROJECTIONS OF n:J'i JlltO'BLEM MAGNITIJDE 20 4.1 Introduction 20 4.2 Methodology in the Model Building Process 22 4.2.1 Data Sources 22 4.2.2 Tlie Approach 22 4.3 Results 23 4.3.1 Models at National Level 23 4.3.2 Models for the State of Florida 26 4.4 Summary 27 CHAPTER 5. IDENTIFICATION OF CRITICAL GROUPS & ISSUES/CONCE:RNS 34 5.1 Introduction 34 5.2 Preliminary Identification of Special Population Groups and Issues ;35 5.3 Final Ranking ofissuesiConcems 38 CHAPTER 6. ANALYSIS OF SOME CRITICAL ISSUES AND CONCE:RNS 44 6.1 Older Drivers 44 6.2 School Aged Children 52 6.3 Young Drivers 72 6.4 International Tourists 73 6.5 People with Disabilities 77 6.6 New Immigrants 85 CHAPTER 7. COUNTERMEASURES AND MATRICES 88 7.1 Older Drivers 88 7.2 School Aged Children 91 7.3 Young Drivers 109 7.4 International Tourists 114 7.5 People with Disabilities 117 7.6Newlmmigrants 121 v


CHAPTER. 8. CONCLUSIONS/RECOMMENDATIONS 8.1Summary 8.2 Conclusions 8.3 Recommendations REFERENCES APPENDICES Appendix 1 Appendix2 vi 125 125 126 127 128 133 141 .


Table u Table 1.2 LJSTOFTABLES Comparison ofTll!IISpOrtation Fatalities to Total and Accidental Fatalities i Fatalities and Injuries by Transportation Mode 3 Table 1.3 Driver Crash Fatality Rates by Age-1996 4 Table2.1 Top ten States with Highest Percentages of Elderly and tlieir Resident Population 8 Table2.2 International Tourists to Florida and United States 10 Table2.3 Violations in Florida Crashes 13 Table3.1 Calculating the Index for Each Issue for Each Oroup 19 Table4.1 Models Developed for Older Drivers at National Level 24 T able4.2 Models Developed for Young Drivers at National Level 25 Table 4.3 SI1II'llll8lj' of the Forecasting Models for School Aged Children at National Level 28 Table4.4 SI1II'llll8lj' of the Forecasting Models for Florida 29 Table 5.1 Basic Statistics about Survey Responses 34 Table 5.2 Composition of Survey Respondents 35 Table 5.3 Ranking oflmportance among Selected Population Groups 36 Table 5.4 Order of Importance of Safety Issues/ Concerns 36 Table 5.5 Estimated Weights for Factors given in Survey 2 40 Table 5.6 Final Ranking oflssues!Concerns Through Different Sets of Weights 41 Table 6.1 Mean Font Legibility Distances in ftfm 45 Table 6.2 Daytime legibility Distances (Bump Signs) 46 Table6.3 Nighttime legibility Distances (Bump Signs) 47 Table 6.4 Proportion of Senior Drivers Who Indicated that Driving Activitie s Became More Difficult Now Compared to 10 years Ago 48 Table 6.5 Proportion of Senior Drivers Who Iridlcated that Highway Features Became More Difficult Now Compared to 10 years Ago 48 Table 6.6 Swnmary ofPBRTto an Unexpected Object 50 Table6.7 Summary ofPBRT to an Expected Object so vii


Table6.8 Decision-Reaction Tunes for Different Age Groups so Table6.9 Observed Critical Gap Values for Different Age Groups and Turning Movements 51 Table 6.10 Driving Knowledge Test Scores Related to Right Angle ll!ld Left Tum Aceidents 52 Table 6.11 Model Summary (Posted Speed-Injwy Severity Model) 55 Table6.12 (\NOVA Results 55 Table 6.13 Significance of the Model Coefficients 56 Table 6.14 Model Summary (VolumeInjury Severity Model) 59 Table 6.15 ANOV A Results 59 Table 6.16 Significance of the Model Coefficients 60 Table 6.17 ll!egal Passes by School Bus Type ahd Presence of Wheel Chair Lift 62 Table 6.18 Tune of Illegal Passing (>2 Table 6.19 Type of Roadway 63 Table 6.20 Vehicle Passed from the Same or Opposite Direction 63 Table 6.21 Vehicle Passed from Left or Right of the School Bus 64 Table6.22 Type of Vehicle that Illegally Passed 64 Table6.23 Number of Students at the School Bus Stop 65 Table 6.24 Strobe Light in Use at the Time of Illegal Passing 65 Table6.25 Reported Problem Areas in Relation with Signs 74 Table 6.26 Survey Results about Understanding of Traffic Control Devices 75 Table6.27 Persons with Disabilities by Degree 78 Table 6.28 Estimates of the Number ofVehicl es with Adaptive Equipment by Data Source 79 Table6.29 Wheelchair Users Injured and Killed from All Causes 81 Ta ble6.30 Wheelchair Users Injured and Killed by Type of Vehicle 81 Table6.31 Wheelchair Users Injured and Killed by Type of Activity 82 Table 6.32 Wheelchair Users Injured and Killed by Type of Vehicle and Selected Activity Type 83 Table6.33 Wheelchair Users Injured and Killed by Severity and Med ical D isposition 83 viil


Table6 .34 Type of Vehicle Adapted for Use by Persons with Disabjlities 84 Table6 .35 Top Five Modifications/Adaptations in Use by Persons with Disabilities 85 Table7.1 Countermeasure Matrix for the Safety Issues of Older Drivers 92 Table7.2 Countermeasures for Speeding in and around School Zones 98 Table 7.3 Traffic Operations and Design Recommendations for Casual or New Adult and Teenage Bicyclists, Urban Section, No Parking 100 Table 7.4 Traffic Operations and Design Recommendations for Casual or New Adult and Teenage Bicyclists, Urban Section, With Parking 101 Table 7.5 Traffic Operations and Design Recommendations for Casual or New Adult and TeeJge. Bicyclists, Rural Section 102 Table7.6 Countermeasures oflllegal Passing of School Bus 104 Table7.7 Countermeasure Matrix for the Safety Issues oflntemational Tourists Table 7.8 Summary of the Policy Issues and Countermeasures .119 ix


LIST O F FIGURES Figure 2 1 Driver Fatalities in Florida Crashes in 1993 9 Figure4.l Observed and Estimated F atality Rates for O l der DriversU.S. 3 0 F igure4. 2 Observed and Estimated Fatality Rates for Young Drivers U.S. 30 Figure4.3 O bserved and Estimated Fatality Rates for Scho o l ChildrenU.S. 31 Figure4.4 Observed and Estimated Fatality Rates fo r Older Drivers-FL 32 Figure 4.5 . Obs e rved and Estimated F a tality Rates for International Tourists -FL 32 Figure4. 6 Observed and Estimated Fatality Rates for Young Drivers-FL 33 Figure 4.7 Observed and Estimated Fatality Rat e s for School Children-FL 33 Figure 6.1 Statist i cal Relationship between P osted Speed and Injwy Severity 55 Figure6. 2 Defi nition o f School Zone 5 7 Fig ure 6 3 Effect of D istance on Model Split 58 Figure 6.4 S t atistical Rel ati on ship between Averag e Dail y Traffic and Injwy Severity 59 Figure6.5 Driving Popul atio n and Subgroups 79 X


CHAPTER 1 : INTRODUCilON 1.1 Backgroun d Safety perfonnance of a transportation system is usUally assessed by examiniqg changes over time in the number and rate of deaths or injuries, and exposure to risk in each mode. By these standards, statistics clearly show that most modes have become safer over the past couple of decades, despi t e more intensive use. Much ofthe" improvement results ftom use of . innovations developed through research, education efforts by the transportation community, and improved emergency response and care. However, there are some risk factors hidden in ihe aggregates that continue to prompt concerns about transportat i on safety. Table 1.1 indicates a comparison of transportation fatalities to total an

Table 1.1: Comparison of Transportation Fatalities to Total and Accidental . Fatalities: Selected Years 1970 1980 1992 1993 Number % Number. % Number % Number % Year ('000) ('000) ('000) ('000) Resident 203,904 -227,225 255,039 257,800 pOpulation Deaths from 1,921 100 1,990 100 2,176 100 2.269 100 all causes Accidental 115 6 0 106 5 3 87 4.0 91 4.0 deaths . Transportation 56 2.9 55 2 7 42 1.9 43. 1.9 deaths Share of -49. 3 -51.7 -48 2 -47. 1 transportation deaths (Data Source: Transportation Statistics Annual.Report 1997, Bureau of Transportation Statistics) . One area that demands further attention is the safety determination of special population groups. From examining current statistics, it is observed that certain population subsets are o ver represen t ed in highway crashes. Table 1.3 shows driver crash and fatality rates per 100 Million Vehicle Miles of Travel (VMT) and per 1000 licensed drivers by age for 1996. Certain age groups experienced crash/fatality rates above the average values. One reason promoting this occurrence could be that current highway designs and policie s are not effectively reflecting the highway s .afety needs of sucjl populations. I.o rnosi cases, the design of highway facilities and traffic control devices, and corresponding policies and regulations are bssed o.n the average characteriStics of all population groups Howe.,er,. certaii; population subsets may have a greater probability in being involved in highway c rashes. Unfortunately, studies conducted on the safety problems of special population 2


.8toUJlS are limited in number. If the safety concerns are identified through researeh, countermeasures could be detennined to address each of the issues. The most suitable . countermeasure could either be a policy change or a design change, depend ing on the issue tha! il$ being taken into consideration. Table 1.2: Fatalities and Injuries by Transportation Mode Year Aviation Motor Rail Transit Wate r Pipelin e (All types) Vehicles (All types) 1985 1584 43,825 454 NfA 1247 33 1990 86 2 44,599 599 339 956 9 1991 983 41,508 586 300 984 14 1992 979 39,250 591 273 912 .15 1993 803 40,150 653 281 910 17 1994 1050 40,716 611 320 853 22 . 1995 961 41,798 567 274 882 21 lojuries 1985 606 N/A 31,617 N/A 2929 126 1990 477 3,231,000 22,736 54,556 3997 76 1991 503 3,097,000 21,374 52,125 4077 98 1992 455 3,070,000 19,408 55,089 3850 118 1993 428 3,189,000 17,284 52;668 3719 112 1994 525 3,307,000 14,850 58,193 4263 1968 1995 462 3,507,000 12,546 56,991 5110 64 N/A-Not avail. able (Data Source: Transportation Statistics Annual Report 1997, Bureau of Transportation Statistics) Among several subsets, six special population groups, experienc ing highway siifety concerns . were selected for the pwpose of this project. Selected special population groups include the elderly, international tourists, school aged children, young drivers, new imm.igrants, and persons with disabilities. 3


Table 1.3: Driver Crash Involvement and Fatality Rates by Age in 1996 Crash Fatality Crash Fatality Driver Involveme n t R a te Involvement Rate Age Rate (per 100 (per 100 Rate (per (per 1000 million VM1) million I 000 Licensed Licensed VMD Drivers) Drivers) 15 -19 1 938 3.2 175 0.29 20-24 886 2. 0 103 0.23 25-29 635 1.2 82 0 1 5 30-34 540 1.0 72 0.13 35-39 541 0.9 70 0.11 40-44 427 0.7 59 0.10 45-49 451 0.7 61 0.09 50-54 383 0.7 47 0.09 55-59 371 0.8 43 0.09 60-64 372 1.0 39 0.10 65-69 410 1.2 37 0.10 70-74 521 1.8 37 0.1 3 75-79 654 2 .9 37 0. 1 7 80-84 782 5.2 36 0.24 85+ 9 1 2 7.9 36 0.3 1 Total 580 1 2 68 0.14 Note: Bold values represent above average crash rates ( Source: Research Note on "Crash D a ta and Rates for Age-Sex Groups of Drivers, 1996" NHTSA, January 1998) 1. 2 Research Statement The unique characteristics of certain special population groups, s u ch as those studied in this research, have become important issues to the highway safety community. There is a great potential for reducing the number of traffic crashes (and e conomic l oss) by addressing the safety issues r elevant to these subsets. Once again, the number of studies addressing the individual highway safe ty needs of special population groups are limited In approaching the safety problem from this vieWpoint, it is first necessary to fully understand the unique characteristics associated with each special population group. It will be advantageous to rank the selected special population groups by their importance in terms of safety, allowing their corresponding critical issues and concerns to be identified Projecting th e magnitude of the 4


problem until the end of the study period is useful in selecting the critical groups. Once the critical safety issues are identified for each population subset, there is a need to investigate probable solutions or cotintermeasures to each of the issues. Florida's warm tropical climate and recreational amenities attract many special populations, including senior citizens, immigrants, and tourists As such, Florida offers an excellent environment to examine the roadway safety issues in association with these sub groups. With the need for addressing the safety issues and concerns in terms of different population subsets, Southeas t ern Transportation Cente r at the University of Tennessee (established under University Transportation Centers Program of US Department of Transportation) funded the project "Evaluation of Highway Safety Needs of Special Population Groups" in May 1997. TIIis study subsequently iden tifie s the most critical special population groups. and their corresponding critica l safety issues along with the countermeasures in improving; the identified issues. 1.3 Research Purposes and ObjectiveS The primary goal of this research is to identify critical safety issues of selected special population groups and then suggest countermeasures in the form of design or policy changes for the identified critical safety issues of the selected special population groups. The specific objectives ofthis study are: I. To select specific population segments that have special highway safety issues and/or concerns, 2. To identify critical safety issues for each of the selected speeial population groups, 3 To perform quantitative analysis to evaluate the significance of safety of each special population group in terms of magnitude, severity of incidents, and future magnitudes due to population, socio-economic and behavior changes, and 4. To i dentify potential changes in public: policy and engineering practices that are implemented to a(jdress the identified critical safety issues. 5


More specifically, this research concluded the following: (1) The ranking of importance among the six selected special population groups: The six special population groups considered in this stud)', older drivers, school-aged . children, international tourists, young drivers, new immigrants, and people with disabilities were ranked according to their highway safety importance. (2) The projections of the magnitude of the problem until Year 2010 are quantified in terms of number and/or rate of crashes: Time series data on crash statistics obtained for each special population group were used to develop models that forecast the magnitude and severity of the problem. These models will enable the number and/or rate of crashes to be projected until the end of Year 2010. (3) The determination of critical safety issues for each population group:. For each special population group, critical safety issues/concerns were ranked in accordance with the importance in terms of safety. ( 4) The suggesting of possible countermeasures: Countermeasures for each of the critical safety issues identified above were then suggested. 1.4 Outline of the Report This report consists of eight chapters. Chapter I p rovides a comprehensive introduction t o this report. Chapter 2 focuses on general details of the selected six special population groups ai:td their significance described by using statistics Chapter 3 explains the methodology employed in achieving the previously mentioned objectives, which is addresSed through three subsections: projection models, surveys, and ranking. model. The models developed for the projections of the problem are identified in Chapter 4. Two sets of models have been developed for Florida and the United States. Chapter 5 presents the results of the two stirVeys conducted and exl'lains how the fmal ranking of issues and concerns were obtained. Chapter 6 addresses in detail some of the critical issues/concerns. Chapter 7 examines the countermeasures identified for each of the'issue/concem for each special population group: Finally, Chapter 8 provides the conclusions and recommendations of this study. 6


CHAPTER2: SELECTED GROUPS AND THEIR SIGNIFICANCE Among several special popwation groups in existence, six groups that are known to experience critical highway safety issues were selected for this project. These special . popwation groups include the elderly, international tourists, school-aged children, young drivers, new immi grants, and persons with disabilities. A brief description of the significance of each of the su bsets is provided in the following paragraphs. 2.1 The Elderly In 1900, 4 percent of the population in the United States lived to be 65 years old or more. In 1984, approximately 12 percent of the population were 65 or older. By the Year 2000, ' it is estimated that as much as 13 percent of the population in the United States will be over 65 years of age. Between the years 2020 and 2030, 20 percent of the populatio n wili . -be 65 years or older even if birth rates stay low (TRB Special Report 218, 1988).' According to the current trends, the majority of the elderly population relies on the automobile for most of their transportation needs. Although preserving the mobility of elderly is a primary concern, traffic safety statistics indicate that the elderly as a group experience a far greater risk of injury and fatality when using the roadways. Based on the statistics, the. number of crashes involving elderly drivers are less than the average ; since they Jess. On the other hand, the elderly arc more likely to be killed or severely injured when involved in a Decreased physical and mental capabilit ies of the elderly reduce their desire to maintain mobility through use. of the automobile. This problem is even more acute when you consider that today's highway envhonm ent, in most cases, is based on the performance characteristics of the younger population. The impact of the elderly is mo re significant in Florida than other states, since it has the highest percentage of elderly as shown in Table 2.1. Due to rich and diverse ethnic and cultural it is likely that the percentage of will continue to grow. Diversity is the factor that sets Florida's older population apart from that of most other states. Serving as a retirement magnet for the last couple of decades, Florida has 7


attracted and continues to attract widely diverse ethnic groups from acroSs the United States mainly for the shared reasons of climate recreation and taxes. Table 2.1 : Top Ten States with Highest Percentages of Elderly as ofJuly I, 1995 and Their Resident Populati on State Resident Population 65 Years or Older Population Total (Thousands) Rank Percentage Rank Florida 14,166 4 18.6 I Pennsylvania 12,072 5 15.9 2 Rhode Island 990 43 I5.7 3 West Virginia 1,7 93 34 15.3 4 Iowa 2,842 30 I5.2 5 Arkansas 2,484 33 14.5 6 North Dakota 641 47 14.5 6 South Dakota 729 45 14.4 8 Connecticut 3,275 28 143 9 Massachusetts 6,074 13 I4.2 10 .. .Y.) r;: ' : :, c.:.:_:.:.:.: ., _.! (Adapted from the Department of Census Data, From http:\\www .ee The majority of the elderly in Florida are both physically and self sufficient 1bis self-sufficiency is directly linked to economic resources, representing one of the state's largest revenue sources; Social Security retirement benefits totaling approximately $30 billion annually (Florida Department of Health and Rehabilitative Services, I989). In recognizing the economic impact of Florida's elderly, the state should endeavor to better Jiving standlirds of the elderly in all aspects of their lives, especially transportation The number of motor vehicle crashes in Florida during 1996 by age group of drivers is . illustrated in Figure 2.1. According to the figure, which depicts fatality rates per number of drivers and per miles driven, older drivers exlu"bit a disproportionate level offatality in crashes. These data indicate the combined risk of being involved in a crash, an injury, and dying in a crash. When ail older driver is involved in a crash he or she has th!' greatest risk of being fatallY injured. s


10 9 i 8 7 6 5 'I; 4 i 3 2 z 1 0 "' cs ; ;:1; .. ;! I!! .. -.. :3 ' ' ' ' ' "' 0 ... ... "' :g :3 ... -"' "' .., ... ... Age Group Figure 2.1: Driver Fatalities in Florida Crashes in 1993 (Source: Special Report on Driver Age Differences m Traffic Crashes, Flo rida Department ofHighway Safety and Motor Vehicles, 1995) As made C'\ftdent by the high fatality rates per vehicle miles driven, the safe transportation of the elderly is a primary concem. In addition to studies conducted on accid ent rates, further research on behavioral and demand analysis issues needs to be; conducted to improve roadway facility perfonnance and design standards for the elderly. The aforementioned reasons necessitate the selection of older drivers as one of the special population groups requiring in depth analysis in terms of safety. 2 .2 International Tourists International tourists play an important role in the touriSm industry of the United States. Florida attracts a considerable percentage of international tourists coming to U.S. Of the 22 million international tourists that came to U.S. in 1995, 19.2% arrived in Florida. Statewide total taxable sales as a result of tourismJrCc'reation accounted for $ 3 8 billion in 1996, which marks an increase of7.3 o/o from the 1995 figure (Florid;& Tourism Industry Marketing Corporation, 1996). 9


The country of origin of international tourists traveling to Florida is listed in Table 2.2. There are significant differences between the transportation systems and signing used in .the United States and that of other countries. Although traffic signs are uniform in U.S., international tourists may not be able to correctly interpret or properly r espond to some traffic signs. Typical problems may inClude understanding of words and symbols, abbreviations and connotations of words, order of information, and cardinal directions. Table 2.2: International Tourists to Florida and United States CountrY Total to Florida (Thousands) 10 Total to U.S. (Thousands) ' 0


International tourists are usually unfamiliar with the area in which they are driving. The driver's familiarity with the road indicates the level ofknowledge that he/she has with the road in question, including location of signs and exits, roadway surface and geometric conditions and so forth. A driver in an unfamiliar roadway may exhibit atypical driving behavior when faced With a situation that in.;olves the perception and reaction of a sudden stimulus, such as in the cases of tail-gating, lane change, and inadequate gap acceptance. In addition, most international tourists are unfamiliar with U. S. driving laws and customs. Thus, international tourists may have a greater pote n tial to be involved in traffic related incidents. Certain design and policy changes could potentially assist international tourists in guidance and navigational tasks that are necessary to safely traverse unfamiliar roadways. In consideration of the above factors, intemational tourists were selected as one of the population subsets requiring further 2 3 School Aged Children . Ensuring the safety of children en-route to and from school is a growing public concern in United States. Injuries and fatalities involving motor vehicles and children traveling to and from school, whether it be on foot, bicycle, or in school buses, generate intense media exposure (more than i t is for any other special population group) and attracts the greatest public outcry to remedy the problem. A growing volume of literature points to four major contributing risk factors for children traveling to and from school: (I) longer walking and bicycling commut es; (2) increased traffic congestion and vehicular speeds along arterial routes during hours; (3) lack of crossing guards at major zone intersections; and, ( 4) lack of motorist knowledge regarding school bus stopping laws. According to a study conducted by the Center for Urban Transportation Research {CUTR) at University of South Florida, approximately I 0,600 motorists illegally pass stopped school buses in Florida during a typical school day {Baits, 1998). During a 180day school year, this amounts to about 1.9 million motorists illegally passing stopped school buses during a typical school year. The alarming number of illegal passes along with other factors raises several broad but important issues about motorists in Florida II


. The most important concerns identified are motoiist understanding of traffic control devices, including signalization and traffic signage, and their driving respons i bilities as defined in certain state laws. More research is needed to be conducted on the effectiveness of applying engineering teclmiques that are geared towards improving safety around schools Fw:ihermore, the existing guidance .to assi st the sch<>

Combining these characteristics with alcohol use creates a particularly deadly mix. . Investigations nee_d to be carried out to find suitable design and policy changes that would assist in changing this alanning trend . Table 2.3: Violations in Florida Crashes (per 1;000 drivers) Age Group Careless Failure to Alcohol Traffic Lane Driving Yield Related Control Change 1524 7.33 4.87 1.01 0 .85 0.42 25-69 2 35 2 .03 0.88 0.31 0.20 70+ 1.27 2 96 0.11 0 33 0.16 (Data Source: Florida Department ofHighway Safety and Motor vehicles) 2.5 People with Disabilities Full participation of the disabled in society is essential for their independent living,: overal l well being, and economic--sufficiency. In addition, the independence of the elderly relieves society of the cost that would be used to take care of the elderly. For participation, they require access to the transportation system in a safe barrier-free way. The Americans with Disabilities Act (ADA) of 1990 was passed to increase access to public facilities for persons with disabilities The Census Bureau estimates that as of the end of 1994, about 54 million non institutional persons were Jiving in the U.S. with some sort of disability, 9 9 % with severe disability (NHTSA, 1997). Along with the increase of access to public facilities, persons with disabilities are being employed in increasing numbers. According to the same source, in.l995, 26.1 % (3.7 million) of those persons betweeii the ages of.21 and 64 with severe disabilities were employed. In addition, 25.9 % of the 2 3 million long term users of mobility equipment (wheelchair, cane, walker clutches) in this age group were employed, amounting to 600,000 per5oris. Many people with disabilities need specific types of modifications made to and/or adaptive equipment added to their motor vehicles to meet their transportation needs. The safety of this population subset is of a 13


different nature and requires careful consideration. Further research is needed to address the safety of their access to the transportation system. 2.6 New Immigrants hnm.igrants are coming to United States in record numbers, many of whom come from vastly different environments than that of U.S. One component that differs from one country to the next is the transportation system. For many new immigrants walking or bicycling is the only familiar mode of transportation. Many of whom are not even familiar with these activities under very high vehicular volumes or speeds. Immigrants, :who do not have long time exposure to safe driving habits, are forced to leain to drive for . the first time. This learning process may be more difficult due to language barriers and cultural differences. Furthermore, many new immigrants learn to drive from friends or family members who may also lack the necessary experience. As a result of the barriers, differences, and difficulties, there is a strong need to investigate through research the safety concerns of new immigrants. 14


CHAPTER3 :METRODOLOGY .. The methodology used in achieving the objectives mentio ned in chapter I is discuSsed in the followi.llg sub-sections; projection models, surveys, and ranking models . 3.1 Introduction The six spe<:ial population groui'S identified for preliminary investigation were ranked in the . order of importance in terms of higl;tway safety. In the meanwhile, relevant issues and conce rns were identified. As such, two surveys w ere conducted in which variou s transportation professio.nals in highway safety were asked to identify critical special population groups and their conesponding issues and concerns. Projection mo

which looked into the safety needs of transit, the study horizon was selected to conclude at the Year 2010. However, the pUipOse of the models was not to explain the occurrence of traffic crashes, but rather to simply exp ress the crash Illte of the considered special populatio n group during the study period. Therefore, the time point within the study period or simply . the year under consideration was selected as the dependent variable. It was difficult to select a suitable period of data to be used in the model building process. Depend ing on the starting point in time chosen, different results were yielded. Thus, it was . decided to develop the proj ection models using the longest period as possible which spanned the histocy ofFARS (Fatal Accident Reporting System) database. In some Florida models, the use of FARS was not justified because data were available for non-fatal era$ statistics. Due to limited resources, this project did not an!llyze Florida's crash database to obtain the necessary statist i cs. Instead information and data available in various publications were used in the projection model building process. 3.3 Surveys Two surveys were conducted to identify critical special population groups and develop a fmal ranking among identified critical issues/concems for all the population groups. The survey fonns used in the first and second surveys are given in Appendix 1 and Appendix 2. The ftrst survey was intended to provide preliminary identification of critical special population groups and their corresponding issues and concerns The second survey addressed a selected set of issues and concerns in detail, considering the effect of several important factors. The fust page of Survey I was used to obtain the IliDking of importance among special population groups in terms of safety. Six special population groups were listed and the respond ents were required to select one of six choices. The choices includ ed a 'No Opinion . response and others that varied from 1 to 5, with 1 indicating 'Less Important' to 5 indicating 'More Important'. The remaining portion of the survey listed the identified issues and concerns of each special population group. Each of the identified issues/concerns required 16


that a check mark be placed next too the opinion of the respondent. In adrutioo, the respondents were given the opportunity to indicate their views by ad ding any sub group whom they consider as having safety concerns. Respondents could also add more safety . issues/concerns for each population group, in addition to the ones listed in the questionnaire. This preiiminary survey form was faxed to the respective individuals and pbone calls were made to ensure a high return rate. The population targeted in the survey included traffic/transportation/safety engineers, Jaw enforcement officers, adminimative officers dealing with safety-related matters, university academics, and Community Traffi<1 Safety Team (CTS1) members in Florida. There is oile CTST for every county in Florida, which works towards increasin g the traffic safety in their county. Based on the guidance achieved through the first survey, a second survey was conducted in a more detailed manner so that more information could be gathered about the selected saJ:ety .. issues/concerns. The possibility of making an evaluation of an issue was also taken into consideration when eliminating or adding some of the issues. The second survey fol!D was also faxed to the same target population with only minor !=hanges to the list (mostly due to employee transfers) . Since this survey form required much more time and energy by the respondents, it was expected to have a lower return rate than that of the :first survey. As a result, it was decided to mail a survey form along with a self addressed stamped envelope with the faxed survey form. This procedure was expected to provide a greater degree of flexibility for the respondents, thereby increasing the rate of return. Follow-up phone calls were also made at three s t ages: before faxing, several days, and after the deadline mentioned in the cover l etter Further reminders were made for the non-responses if . complete negative CQmrnenis were not observed during the earlier phone calls. Detailed information gathered by the second survey included the following: impact do these issues have on crasblacciden\ rates? -How effective would it be to use roadway design changes to address each issue? -How effectiv e would it be to use policy changes to address each of these issues? -How costly would it be to implement design or policy change to address each -How easr would it be to implement the change? 17


How niuch of a priority to address each issue/concern? For each of these factors, respondents were required to rate the factor on a scale of 1 to 5 for each of the issue/concern corisidered. The rating was expected to be based on the question of . "how important the factor is in addressing each safety issue/concern." These ratings given by respondents were used to develop the final ranking model. . 3.4 Ranking Model The final ranking model was developed by evaluating the responses given in th e second survey. Weights were assigned to each of these factors based on the significance andmore accurate understanding about the critical safety issues were obtained by calculating an ihdex for each issue. The higher .the index, more important the issue would be. The magnitude of the index was used to rank the issues/concerns in a more accurate and quantitative way. For a given population group, the index for each issue was: Where; 6 lj= 'L,WiXij lj = index for issue j W, = weight of the factori Xu = mean ranking of issue j when cOnsidering factor i i = factors being considered (1, 2, 3, 4, 5 and 6) j =issues The method of calculating the index for each of the issues under the special population . groups is siunmarized in Table 3.1. One of the most important and also difficult steps in this procedure was determining the . weights . As such, weights were obtained by using two methods. The results of the two methods were compared to assess the reliability of ihe outcome. The two methods used were . (1) assigning weights through a focus group, and (2) using the responses to the survey its1i!f. 18


Table 3.1: Calculating the Index for Each isSIIe for Each Special Population Group Issue Attributes Index i=l I=2 i-3 . i=6 19 ...... w, w, wl W; ...... w, 1 Xu X,, x), Xu Xit I, 2 x,, X.. Xn X;, Xi, I, . . . .. . X,; x,; xl.l X;; Xit I; J . . . The focus group in the flfst method comprised of the research team who assigned weights to each of the factors depending on their personal opinion. on importance The mean of the weights given by the focus group was taken as the final weight for that factor used in estimating the indices. Even though it was possible to conduct a third to obtaiJl the weights, it was decided that the research team bad a better understanding and greater knowledge of each factor, and hence the importance of each factor in comparison. Weights were assigned for factors irrespective of the special population group and issues/concerns. The second method used the mean of the rating given by the respondents to estimate the weights for each factor. 1bis method enabled the weights to be assigned in terms of each special population group. Rankings were obtained by using these individual weights and also by considering the overall mean weight. The rankings obtained by the fust method was compared with the results of the. second method. to assess the overall reliability of assigning weights to each of the factors. 19


CHAPTETR 4: PROJECTIONS OP 'i'm! PROBLEM MAGNITUDE 4.1 Introductio n Forecasting the safety performance of the highway system in tenns of crash frequ ency or crash .rate has several uses in legislation, courts and police enforce ment, police deterrence safety improvement measures, publicity and education programs tax levies, and academics (Hakim et a!, 1991) : However, forecasting is not an easy task and the only thing certain about forecasting is its uncertainty For instance, a set of independent varia bles include d !n a forecasting model may fit availab le statistics very well, but nonetheless. may not explain safety pe!iormance in the future. One reason for this difficulty is that new indepen!lent variables may become relevant in the period following the model building process {Sivak, 1987). The effect of multi-collinearity among included indepen dent variables may also contri bute to a model's inability to explain variation in crash counts. The effect of several other variables not included in the models may also affect the unce rtainty associated with forecasting models (Partyka, 1984) Models developed in the past for forecasting highway safety performance were intent on trying to discover the causes for these changes. To counter the changes, researchers have employed several other types of dep endent variables in the modeling process. The most commonl y observed dependent variables are based on the number of fat alities, fatality r ates per population, per registered vehicle, and vehicle miles traveled. Almost all of these models were used at macro level for total populatio n of an entiie counliy. The independent variables are diverse, prompting researchers to try differen t sets of variables. In one of the earliest models, Smeed used population and n umber vehicl es to obtain fatali ty rates {Smeed, 1949). Sivak tried to estimate fatalities per. registered motor vehicle in each state by using homicide .rate, suicide .rate, fatality rate from nontraflic accidents, unemployment .rate, personal income, density of physicians, alcohol consumption, motor vehicles per capita, road . mileage per vehicle, .sex and age distribution of drivers, and attain ed education as independent variables (Sivak, 1983). However, only three variables, homicides per capita, proportio n of drivers under 25 years of old, and fatality rate from non-traffic accidents were .. 20


proven to be significant predictors. Partyka in 1984 used a set of economic parameters (number of employed workers, unemployed workers, and non-labor force) as independent variables to forecast the total number offatalities. Some other variables employed in the past include fuel consumption, Gross National Product (GNP), vehicle length, average speed, . . traffic volume, traffic density, urban/rural travel, environmental factors, and combinations of the above. Irrespective of the type and number of independent variables used in the model, some resean:hers have conducted studies on the unsuccessfulness of the previously developed models (Sivak, 1987 and Partyka, 1991) . Another limitation to the models explained earlier, Is that they need each independent variable to forecast for the year considered. The prediction for each of these variables is also equally uncertain, ,reducing the accuracy of the final prediction of safety performance. As such, models that do not require additional forecasting of independent variables are expected t o perform better. Safety problems associated with different population subsets are nof the same, because some groups are experiencing higher crash rates than the average population. For example, driver crash involvement and fatality rates by age group for a four-year period are given in Table 1.3 of Chapter 1. Teenage drivers aged 15 to 19 years have the highest crash involvement rates per I 00 million vehicle miles traveled (more than tlu:ee times the national average). A situation is illustrated in the fatality rate of older drivers. One underlying reason for this situation may be that current design and policy are not effectively representing the highway safety needs of the special p opulation subsets. In most cases, the design of highway facilities, traffic control devices, and corresponding policies and r egulations are based on the average characteristics of all population groups. Therefore, certain population subsets have a greater propensity to be involved in highway crashes, which . is expected to vary with time Due to the previously mentioned considerations, it is more helpful to develop separate forecasting for special populatio!l subsets, considering the time point (year) as the . single independent variable. Such models will enable the critical groups to be identified at any time point within the study horizon Depending upon the type of model, different crash rates were used as dependent variables. Fatality rates were considered for models at the 21


national level, while different crash involvement a!!cl injury rates were taken into account for Florida models. 4.2 Methodology in Model Building Process 4.2.1 Data Sources For national level models, the source of the crash statistics was the Fatal Accident Reporting System (F ARS), developed and maintained by National Highway Traffic Safety Administration of U.S. Department of Transportation: At the state level, crash statistics were obtained from publications of Florida Department of Highway Safety and Motor Vehicles. Population estimates were obtained from two major sources Statistical Abstract of the . United States and Florida Statistical Abstract. Estimates on driver licensed population by age group were obtained from Federal Highway Administration publications on highway statistics. The Average Annual Vehicle Miles Traveled (AVMT) by population subset," for young and older drivers, was obtained using results from the Nationwide Personal . Transportation Survey (NPTS) This survey has been conducted for years 1969, 1977, 1983, 1990, and most recently in 1995. For the int ermediate years, it was assumed that vehicle miles traveled varied linearly with the time between the two end data points. However, caution p1ust be taken in using the models developed using A VMt because of changes in survey methodologies at certain points. 4.2.2 The Approach Two sets of models were developed: one at the national level and the other for the state of Florida. Among the several different model formats tested, the format that provided the best fit for all population subsets was oflog linear type. A log linear model is obtained by taking the natural logarithm of the fatality or crash rate. The mathematical form of this model is given in Equation 1. Log, (X)=a T+ b (1) X= e

where, X cras h/fa tality rate, T independent variable associated with the time, and a, b = p arameters to be determined. This m odel, als o referred to as. negative exponential, has the advantage of having reducing gradients with time, whi ch is most suitable for the safety situation of the Using thi s . model format, regress i on analysis was performed on the historical crash data. In a study by . Haight (1989), a s imilar time series model was used to forecast to tal annu al traffic death totals. Haight's method was based on the assumption that the long term trend of fatalities was a representation of two distinct parts ; safety and exposure. As stated by E vans (1993), application of log linear models instead of simple linear regression heeds the warning of "Naive Extrapolation." In some cases l inear regression may provide a good fit to the data, but may not avoi d the unreal isti c possibility of a zero crash rate within a certain time period. 4. 3 RESULTS 4.3.1 M ode l s At Nationall&vel 'Three models were selected at the national level to forecast fatality rates for of older drivers, young drivers, and school-aged children. However, crash data were not available for the other s pecial population groups. Crash rates of international t ourists are not significant as they are spread throughout the counlly. The data range used in the model building process was extended from 1975 to 1996, resulting in 22 data points. As for young and older driv ers, several different crash rates such as fatality rate per I 00,000 populatio11, per 1 00,000 licensed drivers, and per 100 million-vehicle miles trave l e d wer e considered as dependent variables. . Summary of the negative exponential models developed for young and older drivers along with the correspond in g statistical parameters are given in Tables 4.1 and 4.2. For each populatio n subs e t, the model providing the best fit (highest R1 value) was se lected t o be th e f orecasting model. The model that consi d ered fatalities per 100 million vehicle miles traveled was the best model for young and old er drivers. In order to consider the ris k faced by school-aged children, fatality rate per 100,000 population was selected as the d ependent variable. The re sul ting mo dels are shown in Table 4 3 23


. TABLE 4.1: Models Developed for Older Driver s at the National Level . Dependent Variable The Model . !Value o f R (X) Value variable constant . Fatality Rate per I 00 X"'exp [0 .01425 (Year-1974 ) + 1.14064 ) -10.8 65.8 0.854 million vehicle miles . trav e led ' Fatality Rate per X-exp [0.00715 (Year1974.) + 2.64 ) 3.35 94.2 0.359 100,000 population . Fatality Rate per X G exp [0.0087 (Year1974) + 3.2405 ] -3.67 104.1 0.403 I 00,000 lieeased . drivers . ------... ... 24


TABLE 4.2: Models Developed for Young Drivers at the National Level Dependent Variable The Model tValue of JV. (X) ValuO' variable constant Fatality Rate per I 00 X exp [0.05733 (Year-1974 ) + 2.28204] -13 .47 40 8 1 0.901 million vehicle mile s traveled Fatality Rate per Xexp (0.01889 (Year-1974) + 3.7483] -7.84 118.4 0.754 100,000 popUlation Fatality Rate per X exp ( -0.01542 (Year-1974) +4.0457) -5.99 119 68 0.642 100, 000 licensed drivers -------------------------------------. 2S


illustrations of observed and estilnatd fatality rates of the three special population groups are given in Figures 4.1 through 4.3. 4.3.1 Models For the State of Florida For the state of Florida, four sepaiate forecasting models were developed for oider drivers, international tourists, young drivers and school aged children. Unfortunately, crash data for new immigrants and people with disabilities were not available. When considering smaller sub divisions, the use of fatality rate as a dependent variable is not recommended since the number of fatalities may be too small and therefore subject to considerable random fluctuations. As such, injuzy or crash involvement rates were used in addition to fatality rate . The reliability of these crash statistics may be less, since fatalities and f!!tal crashes are better defined and more uniformly reported than injuzy crashes. The availability of historical data , on injuzy does not extend the period covered by PARS, causing models developed. for Florida to be short term as compared to those developed on a national scale. Summary of the negative exponential forecasting models developed for Florida are given in Tahle 4.4. Similar to the national models, it was found that the model considering crash rate per 100 million vehicle miles traveled as the dependent variable was the best fitting model for both old and young driver population subsets. Several other models developed for this study were eliminated to save space. O.nce again, the models that best fit the available statistics were seleeted to be the forecasting models. Observed and estimated crash rates from selected forecasting models for the foilr population subsets are illustrated in Figures 4.4 through 4.7. It was observed that the R2 vaiues for the were lower than those or'tbc national models. Older driver models for Florida generated the lowest R1 values. This is observed due to a sharp rise in crash rates for this population subset during the last several years. More research is necessary to deteimine suitable means of ensuring the highway safety for Florida's elderly drivers. This could be achieved by changing the operational characteristics of the highway system and/or making the necessilry policy changes. 26


4.4SUMMARY The purpose of this study was t o develop models to forecast fatality or crash rates for selected special population subsets at different time points within a specified study horizon. Hence, the independent was selected as the time elapsed from the base year Howeve.r, the models should not be used to project crash rates for an extended period of time. A maximwn ofi0-13 years (or until the end ofyear2010) was the time period visualized by the authors in the developmental of the models. The forecastin g models at the national level were found to be more accurate than the Florida mode ls. The type of dependent variable employed in the models was the primary reason for the differenc es in model accuracy. The national level models used fa tality rates, whereas t,he Florida models consider ed all levels of crash severity. Statistics based on fatalities tend to be . more reliable because they are well defined and recorded in traffic crash reporting . In contrast to fatalities, injury and property damage only (PDQ) crashes are not recorded on a consistent basis not are they defmed with the variation of time: In the young and older driver models, model accuracy is reduced due to the unreliability of A VMT data obtained from the NPTS. When changes in survey methodologies occur, it is difficult to detennine if changes in travel patterns are genuine As a result, the model accuracy for the two population subsets depended heavily on the reliability of the A VMT data. The models developed in this study do not require the independent variable. to be forecasted; rather the year for which the crash rate is to be estimated is substituted into the model. In . comparison with models that require independent variables to be forecasted, uncertainty associated with the aforementioned models i s less. Each additional prediction of an independent variable the reliability of the final forecast The mOdels developed for the populalion subsets will identify the groups having critical safety problems, and also provide an opportunity to check whether the same population subset will remain critical in the future. This identification wiU assist researchers in the area of highway safety t o address the safety problems of most critical special population subsets. 27


TABLE 4.3: Summary of the Forecasting Models for School-A ged Children at the National Level . Dependent Variable (X) The Model tValueof R!Value variable Constant Fatality Rate per (a). Aged S-IS years: X oxp [0.0446 (Year1974 ) + 1.6659] -20.35 57.86 0.954 100,000 popu l ation Fatality Rate per (b) Aged less than 15 years: 100,000 population X oxp [ 0.0476 (Year1974) + 1.5611] -27.3 1 68.14 0.974 i -' 28


TABLE 4.4: Summary of the Forecasting Models for the State of Florida Population Dependent Variable The Model t Value of R' Subset N (X) variable Constant Value . Older Drivers 12 Crash Involvement Aged 65 + years: -3.99 11.43 rate per million VMT X=exp [0.05738 (Year-1984) + 1.20835) 0.784 International 10 Crash Involvem ent Tourists rate per million pop. X=exp [O.OSOn (Year1985) + 7.09387] -7.69 109.01 0.881 Injury rate per million X= exp [0.06287 (Year1985) + 5.68259] -6.82 99.39 0.853 population . Young 12 Fatality rate per Aged 1 5 to 24 years: Drivers millionVMT X=exp [0.07988 (Year-1984)-2.12 825 J -8.2 2 -29.77 0.871 Crash Involvem ent X=exp [0.08387 (Year-1984) + 2.4454] -6.29 24.92 0.834 rate per million VMT School Aged 12 Injury rate per (a). Aged 5 -15 years: Children I 00,000 population X": exp [0.02951 (Year 1984 )+ 5.62511] -6.51 168.68 0.809 . (b) Aged less tlian IS years: (Non motorists) X =-exp [ 0.03027 ( Year-1984) + 5.33544) -6.44 154.31 0.806 . . -Number of years of data considered in the model building process 29


. ,_ 3.5 ::; > 3.3 +-._Observed c 0 3.1 ... \ .. Estimated 2.9 . 0 2 .7 ... 0 .2. 5 . .... 8. 2 3 ............ ll 2.1 ..... _ --Ptojecud 1 9 { X = e:q>.[ 0.01425(Year) + 1 .141} .fl 1.7 {f. 1.5 . . . s s Year . Figure 4 1 : Observed and Estimated Fatality Rates for Older DriversU.S 9 +-Observed .. Estimated . .. . ... ... __ ---(X= (Year-1974 ) +2.28J} . Agure 4. 2: Observed and Estimated Fatality Rates for Young Driv ers-U.S . 30


6 0 +- Obselved 0 5 I" <> Es!lmated 0 0 ,.. c4 0 .. ,., ....... Q. '"3 S:i "'"' 0: 02 . ----. ;; 1 {X= exp.[-{),0446 (Year-1974) ;z 0 ' &-!0" !{;> !b" !!>., !!> ... ... Year Figure 4.3: Observed and Fatality Rates for School Children 7 U.S 31


4.5 _____ .;__ ___ r=::::=======::;--,' 4 +-ObseNed ., 3 5 Estimated t 3 I .. .... .!! 2.5 li 2 li 1.5 0;;; 1 E .o.s ........ ..... .. ___ ----Projected X; exp{-0.0574(Year)+1.21j i 8 Figure 4.4: Obsetved and Estimated Fatality Rates for Older Drivers-Fl 1200 +- Observed 8. 1000 .. ............ EsUmated .!! ,g 800 .. c:-eoo .... .. .. -... __ 400 -----.. g.g c:= Projected :C E 200 . .. X= exp.[ -0.0807(Year-1985)+7.084j e 0 0 ib "-1.0 !l> !:> '!'>>,. !"-!"> S !:> ".,.

. :: -t- Observed 12 . ... Estimated 8. 10 .. .s .. 8 ... ..:.... ........ 4 ... .. '.-. ----. . ---fs 2 -exp[.0.0836(Year-1984)+2.44J Projected J::. 0 . I! tJ !:> t\ !!> !l> ::. I< !:> !0 ::\ !!> P.l sf' Year Figure 4.6: Observed and Estimated Fatality Rates for Young Drivers-FL 280 260 .... ........ +- Obse.rved 240 ..... .. .. Estimated -...., ... ........ 0 220 ... 0 .. C!.. 200 0 . .. 0 ... --.. -. Projected CD 0 expi.0303(Year 1984)+5.33] 120 C' 100 !:> t\ !!> !!> ::. g,"> g,<'o !!> !!> " " " " " y" " " 'l) 'l) ear Figure 4.7: Observed and Estimated Fatality Rates for School ChRdren (5-15 yrs). FL 33


CHAPTER 5: IDENTIFICATION OF CRITICAL GROUPS AND ISSUES/ CONCERNS l.llntr'od u ction Identification of critical special population groups and their correspon ding highway safety issues/co n cerns was performed using the responses obtained in both of the previously mentioned surveys. The rate of response for the each survey is given in Table 5 1. Table 5.1: Basic Statistics about Survey Responses Survey Number of Number of Response Rate Survey Fonns Survey Forms Sent Out Returned Survey 1 149 92 62% Survey2 164 72 44% The composition of survey respondents with respect to theii profession is shown in Table 5.2. Apparent from this table, the number of respondents that could not be identified by profession increased in Survey 2 The reason this situation was that the mail-back option was made available to the respondents, which made it diffl.cult to personally identify the respondent's profession. In the first survey, most of the respondents faiced the survey form, making it possib l e to identify th:e respondent through a matching of fax numbers. As expected, the return rate was low in the s econd survey due to the in-depth nature of the survey, which required respondents to spend more time and energy in completing the survey form. The same professionals were consulted in both surveys, except for the case of an e mployee transfer where an additional person was surveyed. Therefore, irwas expected that the results of the two s urveys would be comparable to one another.


Table 5 2: Composition of Survey Respondents Profession Number (Percentage) Survey I Survey2 Engineers (Traffic, 52 (56.5 %) 29 (40.3 %) Transportatio!l, Safety etc.) Law Enforcement Officers 13(14.1 %) 11 (15.3 %) (Police, Highway Patrol etc.) Administrative Officers in II (12.0%) 7 (9.7 %) Safety Area University Academics 5 (5.4 %) 2 (2.8 %) Others and Unknown 11 (12 %) 23 (31.9 %) 5.2 Preliminary Identification of Special Population G r oups and I ssues The mean ratings as provided by the respondents in Survey 1, were used to identify an4 rank critica l special population groups by highway safety importance. The ranking together with the mean rating and standard deviation is listed in Table 5.3. Rankings of the identified highway safety issues are given in Table 5.4. It is worthwhile mentioning that the ranking of Survey 1 issues and concerns was not considered to be accurate enough to address 'or implement a corresponding countermeasure. Rather than emphasize one single factor, a multi criteriqn approach was taken whereby a second survey form Was designed specifically to sui t that It took into account several different attributes that were eXpected to be important in reaching the final decision about addressillg of the . issues. In light of the opportunity for .respondents to eXpress their on safety related matters, several other populations subsets such as pedestrians and cyclists were nained However, these subsets were not subjected to further investigation in this study. 35


. Table 5.3: R.ankillg of Importance among Sele<:ted Population Groups Rankin g Special Population Group Index Std. Deviation 1 Older Drivers 4.28 1.07 2 School Aged Children 4.12 0.87 3 Young Drive rs 4.00 1.01 4 Disabled People 3.53 1.23 5 Int ernational Tourists 3.35 1.28 6 New Immigrants 3.12 1.06 Table 5.4: Order of Importance of Safety Issues/Concerns (a) Older Drivers Ranking Issue/Concern Index 1 Night time visibility 4 .53 2 Perception reaction time 4.23 3 D imensions of signs and lettering 3.84 4 Loca tion of the traffic signs 3.53 5 Gap acceptance capabilities at th e stop 3.43 controlled intersections 6 Difficulties in understanding the 3.32 changeable highway messages 7 Deficiencies in driving know ledge 3.09 8 Start -up lost time and saturation 2.92 headway at sigoaliud intersections (b) School Aged Children Ranking Issue/Concern Inde x 1 High vehicular speeds in and around 4 .2 7 school ;woes 2 High traffic volumes around schools 4.05 3 Lack of pick-up and drop-off zones, 3.70 sidewalks, traffic lights etc. 4 Lack of traffic and bi cycle safety 3.60 education programs 5 Lack of proper signage, signals etc. 3.60 6 Not enough school crossing guard s 3.23 7 Lack of traffic devices such 2.98 as humps, raised pede strian medians etc. 36


(c) International Tourists Ranking Issue/Concern Index 1 Confusion of traffic guid e signs 3.83 2 Unfamil iarity with roadways 3.83 3 Recognition of traffic signs 3..55 4 Unfamiliarity with the driving system 3.25 (d) Young Driv ers Ranking . Issue/Concern Inde x 1 Risk taking behavior/aggressiveness 4.26 2 Speedin g 4.17 3 General recklessness 4.10 4 Impaired Driving(Aic:ohol an d Drugs) 4.09 5 General lack of driving experience 4.08 6 Lac k of driver training programs 3.73 7 Lack of safety belt use 3.71 8 Teen cruising 2.81 (e) People with D isabilities Ranking Issue/Concern Index 1 Lack of policies for tecalling driving 4.01 licens es from people with progressive disa bilit i es . 2 Uns afe access to transportation 3.73 faciliti es 3 Lack of standards for vehicle 3.44 modifications necessary for disabled 4 Lac k of crossin g devices for blind 339 people s Default designs in roadways and 3.28 inters ec tions 37


(f) N e w Immigrants Ranking Issue/Concern Index 1 Inadequate/ Improper driving 3.52 education 2 Language barriers 3 .51 . 3 Unfamiliarity with the driVing system 3.40 4 Differences in the transportation system 3.30 Cultural differe nces 5 3.16 . 5 .3 Final Ranking of I ssues/Conc erns The final ranking of issues was based on the results of the second survey, which considered six different attributes (factors) These factors included impact on crash rates, . effectiveness of apply in g roadway design changes, effectiveness of applying policy . changes, cost associated with the change, ease of implementing the change, and priority. to address the issue. By applying weights to the individual factor ratings for each issueiconcem, the index explained in the previous calculated for each special popu l ation group The weights estimated by considering three different methods are given in Table 5.5. One set of weights was obtained through a focus group while the other two sets were achieved by considering the responses in Survey 2, In two sets of model s weights were used for all of the special population gro u ps without differentiating between th e groups. However in the third set, weights were estimated for each facior in each special pop ulation group. Three sets of weights were applied t o the mean ratings and the issues were ranked in . o rd e r of safety importance. The resultant rankings are listed in Table 5 6 : It can be observed that for the special population groups o f international tourists and pe o pl e witli. disabilities _the rankings are exactly the same irrespective of the set of weights For other groups the r anking changes slightly bu t nevertheless the major priorities remain the same Hence, it can be concluded that the selec ted s ets of weights d id n o t significantly affect the ranking r esults. 38


5.4Summary Obtaining a considerably high response rate from the surveys was a challenging task in . this project. This was more evident in the case. of Survey 2, where a considerable amount of time and effort was expected from the respondent. As such, the achieved response . rates were considered to be satisfactory. A concern arises about the consistency and compatibility of the two surveys, Since the final ranking of issues was based on the second survey this may not pose any negative implications on the results. Three differ ent types .of weights were applied to the factors in Survey 2 to obtain the rankings for each special population group. It was observed that a similar ranking of the . issues was achieved irrespective of the set of weights employed. This could be considered as an indication of the reliability of the ranking for each group. 39


Table 5.5: Estimated Weights for Factors given In Survey 2 Weights (based on a scille of I -I 0 ) Factor Through the Through the Responses to the Survey Focus Group For all groups For Individual Special Population Group I. Impact on crash/accident rate 8.5 7:3 Oldu Drivers 7.2 ; International Tourists 72 School Aged Children 7.4 ; Young Drivers 8.4 New Immigrants 7 2 ; People with Disabilities6.2 2. Effectiveness of using roadway 5.0 4.9 Olde r Drivers 5.8 ; International Tourists 4.6 design changes School Aged Children 6.4 ; Young Drivers 3.2 New Immigrants 4 0 : People with Disabilities 5.2 3. Effectiveness of applying policy 5. 6 1 Older Drivers 5.2 ; International T ourists 5.6 changes School Aged Children 6.4 ; Young Drivers 6.6 New Immigrants 6 0 : People with Disabilities 6 8 4. Cost associated with implementing 6.5 6 6 Older Drivers 7.2 : International Tourists 6.6 design/policy change School Aged Children 6.2 : Young Drivers -6.0 New Immigrants 6.4 : People with Disabilities -7.0 5. Ease of implementing the change 6 5 5.1 Older Drivers -52 ; International Tourists -5.0 School Aged Children 5.8 ; Young Drivers 4.8 New Immigrants 4.8 ; People with Disabilities 5.2 6 Priority to address the issue 3 6.8 Older Drivers 6.4 ; International Tourists -6.0 School Aged Children 7.6 ; Young Drivers -82 New Immigrants 6.2 ; People with Disabilities 6.4 . . . 40


Table 5.6: Final Ran!dngs oflssues/Concems Through Different Sets of Weights (a) Older DriverS Rankin g Common Common Individual Issue/Concem Weights Weights Weights through through through Focus Group Survey2 Survey2 1. Night time visibility 2 2 2 2. Driving in congestion 9 7 7 3. Freeway driving 8 9 9 4. Maneuvering curves 7 8 8 5. Deficiencies in driving knowledge 5 5 5 6. Location and siZe of traffic signs and lettering I 1 I . 7. Perception-reaction t ime 4 3 3 8. Gap acceptance 3 4 4 9. Narrow lanes 6 6 6 (b) School Children Ranking Common Common Individual Issue/Concern Weights Weights Weights through through through Focus Grouj! Survey2 Survey2 1. High traffic volwnes around schools 9 9 9 2. Speeding in and around scho(!l zones 4 3 3 3. Lack of drop-off and pick-up zones 5 5 5 4. Lack of traffic calming devices 8 8 8 5. Lack of proper si gnage and traffic Signals 6 6 6 6. Lack of sidewalks and bike routes I I 1 7. Lack of traffic and bicycle safety programs 3 4 4 8. Lack of school crossing guards 7 7 7 9. Lack of parental guidance 2 2 2 4 1


(c) International Tourists Ranking Common Common Individual Issue/Concern Weights Weights Weights through through through Focus Group Survey2 Survey2 1. Unfamiliar with roadway system 3 3 3 2. Unfamiliar with driving system 5 5 s 3. Difficulty in recognizing and understanding I 1 1 traffic signs 4 Lack of information about driving Jaws and 2 2 2 customs in US 5. Confusion in translating English into metric 4 4 4 distance on signs (d) Young Drivers Ranking : Common Common Individual Issue/Concern Weights Weights Weights through through through Focus Group Survey2 Survey2 1 Lack of Driving Experience 3 5 4 2. Reckless/aggressive driving 2 2 2 3 .. Speeding 1 1 1 4. Iinpaired driving s 3 3 5. Lack of safety restraint use 4 4 5 6. Lack of driver training programs 6 6 6 7. Immaturity 7 7 7 . (e) People with Disabilities Ranking Common Common Individual Issue/Concern Weights Weights Weights through through through . Focus Group Survey2 Survey2 1. Unsafe access to transportation facilities 3 3 3 2 Lack of policies for issuing and recalling 1 1 1 s licenses 3. Lack of standards for vehicle modifications 2 2 2 required for disabled 4. Faulty designs in roadways and intersections 4 4: -4 42


(f) New Immigrants Ranking Common Common Individual Issue/Concern Weights Weights Weights through through Focus Group Survey2 Survey2 1. Unfamiliar with traffic Jaws and regulations 1 1 1 2 Unfamiliar with vehicle systems 5 7 5 3. Unf3lll11iar with road systems 2 4 2 4. Lack of driver education 4 3 4 5. Lack of safety education 3 2 3 6. Language baniers 6 5 6 7. CultUral differences 7 6 7 . 43


CRA.P1'ER 6: ANALYSIS OF SOME CRITICAL ISSUES/CONCERNS The critical issues and concerns identified through the use of the ranking model in Chapter 5 are analyze d in this section. An extensive literature search was performed to obtaip. past studies that addressed each area of safety concem. Accordingiy, this analysis is based on the information ascertained in those studies. The details and summazy tables provided whenever possible render a detailed into eacb issue and cbncem. 6.1 Older Drivers In terms of highway safety, older drivers were consistently ranked first. Some of the issues/concerns identified in several past studies relating to older drivers are discussed in this section. 6 . 1.1 Location and Size o f Traffic Signs and Lettering The issue of location, size, and lettering of traffic sigxis received the highest priority among all the issues/conce rns selected in the second survey. The size and location of both traffic signs and lettering should be in such a way that not only the average population, but also older drivers; who generally experience decreased mental and physical capabilities, have sufficient time to respond to a required action. . A study conducted in 1992 attempted to determine if different alphabet styles/fonts other than the existing standard letter series specified in the Manual on Uniform Traffic Control Devices on Highways and Streets (MUTCD) would increase legibility (Tan, 1992). The measure of . effectiveness used to determine the performance of a sign was legibility distance, which was : . defined as the farthest distance at which a sign could be read. The three types of signs used in the study were regulatory, warning, and guid. e signs. Legibility distances were measured for two age groups, older (65 years and older) and younger (16-25 years). Table 6.1 gives the . mean legibility distances of different fonts (in terms of legibility per inch of letter height) for the two age groups categorized by gender. '44


Table 6.1: Mean Font Legibility Distances in ftfm {a) For Regulatozy Signs Subjects Mean Legibility Distance (ftfm) Letter Series C Serif Gothic Helvetica Vag Rounded Dot matrix Older Females 54 50 50 57 34 Older Males 57 48 57 59 34 Younger Females 66 62 66 70 4 9 Younger Males 85 78 83 86 .64 All Sub jects 65 60 64 68 45 (b) For Warning Signs Subjects Mean Legibility Distance (ftfm) ; . Letter Series C Serif Gothic Helvetica Vag Rounded Dot matrix Older Females 72 71 71 76 54 Older Males 76 77 79 80 56 Younger Females 85 86 87 . 89 75 Younger Males 97 97 102 103 84 All Subjects 82 83 85 87 67 (c) For Guide Signs Subjects Legibility Distance (ftlin) Letter Series C Serif Gothic Helvetica Vag Rounded Dot matrix Older Females 47 46 48 46 46 01derMales 50 48 49 47 50 . Younger Females 51 51 52 51 52 Younger Males 58 55 60 57 58 All Subjects 51 50 52 so 52 4S


An analysis of variance (AN OVA) petfolllled on the data indicated that for all three types of signs the legibility distances were significantly for age and gender at the 95 percent confidence level. The results show that older subjects perfolllled substantiilly worse than their younget: counterParts and males fared better than females. However, most of the . . legibility values were greater than the 50-ftlin standard This is the value utilized in current highway sign design that asswnes a legtoility distance of SO feet per one inch of letter height. As such, the study suggested retaining the standard highway alphabet letter series. A study conducted in 1991 examined the trade-off between traflicsign size and background material on legibility and conspicuity distances (Khavanin and Schwab, 1991). Three . different sign sizes and two retro-reflective materials were tested by 48 subjects of the ages 60 and older. The legibility distances observed during daytime for three sign sizes are given in Table 6 :2. As expected, the statistical analysis indicated ihat the legibipty distance was negatively correlated with age and reactio n time. Table 6.2: Daytime Legibility Distances (feet) Bump Signs Sign Size Mean SID Min Max Median 18 inch 222 65.4 85 441 220 24 inch 288 82.0 169 569 294 30inch 338 83.4 127 524 340 The nighttime legibility distances observed during the same study for the two different retro reflective materials, referred to as types II and liT, are given in Table 6.3. As indicated, legibility distances increase coD.siderably with the size of the sign. Although this effect is not as pronounced during the daytime. The results of this suggested that the standard of 50 ft per inch of letter height used for highway guide signs was not adequate for older drivers. 46


. Table 6 3: Nighttime Legibility Distances (feet)Bump Signs Size(in), Type Mean STD Min Max Median 18,11 132 36.4 57 207 .133 24,ll 186 44.2 85 267 188 30,11 229 59 2 87 339 235 18,III 142 41.4 57 251 141 24,III 215 59.7 84 350 216 30, III 249 64.5 90 388 252 6.1.2 Nighttime V is ibility In terms of safety for o l der drivers, nighttime visibility received the second greatest emphasis The second study indicated that nighttime legibility distances are smaller than daytime legibility distances for older drivers (Khavanin and Schwab, 1991) Benekohal et a!. conducted a statewide 'study on highway design and traffic operation needs of older drivers in Illinois (Benekohal et al., 1994). A total of 664 senior drivers responded to a mail survey that asked questions about the importance of present day highway design features and difficulty of driving activities as compared to 10 years ago. The ma j ority of the elder!y respondents indicated that driving at night was the most difficult d ri ving task. As for . the im p ortance of highway design features, intersection lighting was identified to be of primary concern. Examination of these two fact ors will explain the problem of nighttime visibiliiy experi e nced by older drivers. The driving activities that have become more difficult . and th e design features that became more important together with their corresponding percentages are listed in Table 6.4 and 6 5 respectively. Statistical tests were carried out to analyze the driving d i fficulties not ooly by age but also by gender S i gnificant gender differences were observed for driving at night. Female drivers reported nighttime driving more difficult than male drivers. As far as the four age groups 47


(66-68, 69-72, 73-76, end 77+) considered, nighttime driving has become particularly difficult for the 77+ age group, more then for any other age group. Table 6.4: Proportion of Senior Drivers Who Indicated that Driving Activities Became More Difficult Now Compared to I 0 Years Ago Driving Activities Percentage Driving at Night 62.17 Driving in Heavy Traffic 52.16 0 Driving at High Speeds on Freeways . 31.68 Driving in Construction Zones 28.5 Reading Street Signs in Town 26.79 Reading Signs on Freeways 23.46 Driving Across an Intersection 21.17 Finding Beginning of Left Tum Lane 20.37 Making Left Tums 19.01 0 Following Pavement Markings 16.85 Driving in Daytime 13. 3 Responding to Signals 11.57 Table 6.5: Proportion of Senior DriverS Who Indicated that Highway Features Became More Difficult Now Compared to 10 Years Ago Highway Design Features Percentage 0 Lighting at Intersections 61.69 0 Pavement Markings at Intersections 56.57 Number of Left Turn Lanes 55.24 Pavement Markings on Hills and Curves 54.98 Width of Travel Lanes 50.63 Length of Freeway Entry Lanes 48.73 Concrete Lane Guides for Turns 46.61 Size of Traffic Signal at Intersection 41.5 48 : ..


6.1.3 Perception -Reaction Time In the ranking of highway safety issues of older drivers, perception reaction time was ranked third in importanCe. One of the most important applications of perception reaction time in highway design is in relation to stopping sight distance It is essential to provide adequate stopping sight distance at every point a roadway. As a minimUm, sight distance should be long enough to ensure that a vehicle traveling at or near design speed can come to a complete stop prior to reachirig a stationary object in its path. Stopping sight distance is the . sum of two componentsbrake reaction distance (distance traveled from the instant of object detection to the instant the brakes are app li ed) and the b r aking distance (distance traveled from the instant the brakes are applied to when the vehicle is decelerated to a stop) .. The component of brake reaction distance is calculated using perception reaction time. llte American Association of State Highway Officials (AASHTO) recommends a total percepJion reaction time of2.5 seconds for all design speeds . Highway facilities designed on the basis of2.5 seconds may pose safety implications to older drivers with longer perception reaction times Several studies have investigated this situation and attempted to identify the exi stence of such a problem A recent study used an in vehicle . instrumentation package to measure driver p e rception reaction times, braking distances and decelerations to unexpected and anticipated stops fo r different age groups (Fambro, 1998). Four different types of studies were conducted: Study 1 Was a pilot study, while 3, and 4 were actual experiments from which the conclusions were drawn. Studies : 2 and 3 where similar in regard to the use of a closed course, but study 2 employed a test . vehicle whereas study 3 used the vehicles belonging to the test subjects. Study 4 was performed on open road conditions using personal vehicles. The observed Perception Brake Reaction Times (PBR1) for unexpected and expected object scenarios are presented in Tables 6.6 and 6.7 respectively. 49


Table 6 6: Summary ofPBRT to an Unexpected Object Study Age Group No. ofTest MeanPBRT STD Subjects (sec) (sec) Study2 Older 1 2 0.82 0.159 . Younger 10 0.82 0.203 Study 3 Older 7 1.14 0.353 Younger 3 0 .93 0.191 Study4 Older 5 1.06 0.222 Younger 6 1.14 0.204 Table 6 7: Summary ofPBRT to an Expected Object Study Age Group Gender No. ofT est MeanPBRT SID Subjects (sec) (sec) Study 2 Older Female 7 0 .66 0 .216 Male 7 0.65 0 .228 Younger Female 6 0 .57 0.167 Male 6 0.48 0.088 Study 3 O lder Female 5 0 .67 0.252 . Male 3 0.65 0.345 Younger Female 2 0.49 0 .168 Male 1 0 .55 0.078 Additionally, a project conducted in 1996 studied inteisection decision reaction time for older drivers (Naylor and Graham, 1996). Statistics on d ecision-reaction times observed in the field are listed in Table 6 8 for this study. The multivariate analysis of variance conducted on the data revealed that decision reaction times for the age groups were significantly different at a 95 percent confident level. Table 6.8: Decision-Reaction Tim e s for Different Age Groups Statistic (sec) Younger Older Total Group Group Mean D R time 1.16 1.32 1.24 Standard Deviation 0.45 0.54 0.50 Minimum D-R time 0 .38 027 0.96 Maximum D-R time 2.48. 3.00 3.00 85doPercentile 1.66 1.86 1.73 . 50


6.1.4 Gap Acceptance Gap acceptance capabilities of older drivers at stop-controlled intersections was identified as the fourth issue in the order of highway safety importance. is reason for concern since crah statistics indicate that poor gap acceptance at intersections is the leadin g causal for more than forty percent of all crashes at such locations. One particular study on the gap acceptance of elderly drivers on rural highways estimat ed critical gap values for elderly and other drivers (Yi, 1996). Observed critical gap values for elderly and other drivers were 7.31 and 5.i9 seconds, respectively. This difference found to be significant at the 90 % confidence level. The study further estimated that on average an elderly driver rejected 18 % . of all usable gaps, compared to only 7 .8% for drivers in other age groups. A study conducted by the University of South Florida (USF) looked into the gap acceptance capabilities of di.fferent age groups for different turning movements (Lu and Dai, 1997). Three age groups, Older, Middle, and Young along with the turning movements of left turning and through were considered in this study. A summaiy of the critical gaps observed is presented in Table 6.9. For both movements, gap acceptance behavior of older drivers and. the other age groups was found to be significantly different at the 95 %confid ence level. For the through movement it was found to be statistically significant for all three age groups. Table 6.9: Observed Critical Gap Values for Different Age Groups and Turning Movements AgeOroup Critical Gap Value (seconds) Left -tum Movement Through Movement Older Driver Group 6.95 6.13 Middle Driver Group 6.04 5.58 Yo1111g Driver Group 5.89 5.42 All Driver Groups 6.81 6.09 $1 ;


6.1..5 Deficiencies in Drivin g Knowledge The fifth issue of safety concern as it relates to older drivers was the lack of driving knowledge. In 1992, McCoy et al. more than 2,000 driver license tests taken by older drivers who were renewing their Nebraska driving licenses. The study tried to Iden t ify whether there was a relationship between test scores and accident patterns. A score of 80 % was required to pass the test, but it most cases the average score of the older drivers was less than80%. For three age groups, the average stores were as follows: 55 64 yrs: 79 % 65 -.74 yrs: 76 % 75 +yrs: 69% The results of this study clearly illustrate that older d rivers were not versed in the various driving IaVlS and regulation.s. In addition, the study attempted to correlate scores with left-turning and right angle accident patterns. The percentage of questions answered correctly relatin g to these two types of accidents is given in Table 6.10. In each case, the percentage of correct answers decreased with increasing driver age. Drivers 75 years and older had the most difficulty answering the test questions. TSble 6.10: Driving Knowledge Test Scores Related to Right-Angle and Left-Tum Accidents Questions Age Group (years) 55-64 65-74 75+ Right-Angle Accident Related 83% 78% 72. % Left-Tum Accident Related 86% 82% 73% 6.2 School-aged Children Accorciirig to the findings of the two surveys and the feasibility for evaluation, five issues were selected for further examination. The five critical safety issues ide ntified were high S2


traffic volumes around school zones, speeding in and around school zones, illegal passing .of . school buses, Jack of proper signage and traffic signals, and Jack of drop-off and pick-up zones. In order to facilitate the development of appropriate countermeasures, it is essential to identify traffic and highway design characteristics that are prevalent in crashes involving . . motor vehlcles and school-aged pedestrians and bicyclists. Furthermore, to perform an . accurate safety analysis detailed information on the physical characteristics of roadways and environments surrounding the crash locations is needed. Consequently, it would be advantageous to utilize Geographic Information Systems (GIS) to facilitate the analysis and display of large quantities of data. There are various GIS software packa g es available in the market today. The most popular PC based software include ATLAS ms, GIS PLUS, and Maplnfo. In this study, Maplnfo was the software of choice because of its functions. 6.2.1 Speeding In and Around School Zones In order to evaluate this issue, hard-copy police reports of traffic crashes involving pedestrians and bicyclists were collected from th e Metropolitan Planning Organiza t ion (MPO) of Brevard County, Florida. In addition, data were compiled for the 1990 Census of Population and Housing Summary Data File 3A, and student enrollment and location of each private and public school in Brevard County. Lo n g-form reports involving pedestrians . and bi<;yc!ists were obtained for a 5-year period (1993-1997). Pedestrians and bicyclists of the ages 5 to 15 were elctracted from the crash reports and transcribed into electronic format file. A child's age could be easily determined from the birth date indicated in the reports. The 1990 U.S. Census of Population and Housilig Summary Tape File 3A was used to examine the demographic and socioeconomic characteristics of Brevard County in an attempt to identify factors that may potentially impact traffic crashes involving school-aged pedestrians and bicyclists. This summary data file was geographically referenced and used in this study on a census block group level. Also, information pertaining to the spatial distribution of schools and student population for public and private schools was !lcqUired from Brevard County Department $3 .


All three data sets collected were graphically depicted on the GIS coverage by geocoding. Geocoding is attached with a reference (known to the GIS coverage) to the traf!ic ciash . database. The GIS coverage is capable of interpreting latitude and longitude. If a traffic . accident record, school, or census block has a Oatitudellongitude) coordinate, it will also . have a position on the GIS coverage. As a result, all the information could be displayed on one map. Another advantage of using GIS for traffic crash analysis is the i.cherent ability of GIS to perform spatial analysis. Traffic crashes involving school-aged pedestrians and bicyclists in and around school zones can be analyzed in conjunction with demographic and socioeconomic characteristics as a means of identifYing crash patterns. The data analysis phase included analyzing traffic crash data for the statistical relationship between speed and traffic crasbes involving school-aged pedestrians and bicyclists. A linear regression model was selected to describe the mathematical relationship between speed .l!:lld severity of injwy of school-aged pedestrians and bicyclists. The severity of an injury Was provided in each crash report by the investigatin$ officer. Crash severity varied on a scale of 1 to 6, with 1 indicating 'None Injwy', 2 'Possible Injury', 3'Non-capacitating', 4'Incapacitating', 5-'Fatal with 90 days', and 6-'Non-Traffic Fatality'. In order to perform an accurate and comprehensive analysis of accidents involving solely school-aged pedestrians and bicyclists, five criteria were developed to select crash records in the crash database. The five criteria used to screen the data set were as follows: (I) under prevailing conditions a driver was driving at normal operali?& speeds instead of performing parking, backing, or turning maneuvers;' (2) records that contained information on estimated vehicle speed at . impact and posted speed limits; (3) records that contained information on iJUury severity level, (4) records that contained the location of a Crash; and (5) the location indicated in tlie police report was in a school zone. After using the five criteria to screen the aasb reports, a total of 41 crashes across Brevard County were analyzed (as shown in Figure 6.1 ). The mathematical relationship between pos ted speed and crash injwy severity is given .below: Average Injwy Severity= 2.313 + 0.019 x Posted Speed .. :


6.00.------------:----., .a. 5.00+--------------l 4 .00 t--"--t-----

Table 6.13: Significance of the Model Coefficients Model Unstandardized Coefficients Standardized t Statistic Significance . Coefficients Level B Std. Error Beta Constant 2 313. 0.681 0.153 3.396 0 002 Difference 0 019 0.020 0.957 0.344 It can be inferred from the data that higher posted speeds -result in more severe craslles . involving school-aged children. However, it is worthwhile mentioning that the .databaSe of school-aged traffic crashes was very J.llnited compared to other traffic safety databases 1bis may be due incomplete police reports and underreporting of crashes. Several studies in the area of highway sllfety have indicated that demographic and alcollol related variables couid be used in addition to vehicle SPeed to model the risk of traffic crashes. Unfortunately, these variables c o uld not be quantified in this study, The research team tried to model the risk of traffic with demographic variables such as universal population of school zone, school-aged children population, and student population of school zone; however, no significant statistical relationship could be established. 6..2.21Iigh Traffic Vol ume around School Zones For the evaluation of this issue, iraf!ic count data -were collected from Metropolitan Planning Organization (MPO) of Brevard County, Florida. This computerized data file maintained by Brevard Cotinty MPO contains annual average daily traffic for individual segments of count}-, state and local roads for the time period of 1993 to 1997. Within ihe county's boundary, a total of 457 road segments with available traffic count data were assigned to a corresponding road and graphica!ly displayed with the use of Maplnfo. Accordingly, data for roadway segment traffic volume, traffic crashes involving school-aged pedestrians and bicycllits, school" information, and socioeconomic-demographic S6 .


characteristics were depicted on one map. However, traffic count data were only available for major roads, thereby limiting the number of roadways analyzed. GIS technology played an integral role in :(Urther examining this issue. Twenty school zones in Brevard County we r e defined as study areas, A school zone was defined to be the area within one-mile radius of each school. If school zones were overlapping, efforts were then taken to combine zones together to form one school zone (as shown in Figure 6 2). Zone I Figure 6.2: Definition of school zone Several factors were taken into consideration prior to establishing the criteria for school zones. In the study Home-to-School Transportation Study (University of Florida 1993), 2547 students living within a 2-mile radius of a school were surveyed to determine the effect of home-to-school distance on .mode choice. As indicated in Figure 6.3, the home to school distance of 1-mile was the critical point where mode choice shifted from walk/bike to cars. Secondly, Florida Statutes and Rules

500 450 400 350 .:1 300 250 ., 200 en ISO 100 so 0 114 112 I >I MUes Figure 6.3: Effect of Distance on Model Split, However, there were a limited number of road segments with available traffic counts lhat could be assigned to a school zone. In some school zones, only 3 or 4 road segments with 0 0 traffic count data were available. In some cases, the defmed school zone did not have traffic crash data. As a result, a total of 10 school zones were analyzed in this study. This limited . data set affected the quality and reli ability of the data analysis. In addition to the traffic coilnt 0 data, data of traffic crashes involving school-aged pedestrians and bicyclists, schoo l 0 infomiatioo, and social and demographic characteristics were used in the traffic volume study. The traffic count data, witli the other data sets, were anal yzed using Maplnfo and SPSS to obtain a statistical relationship between the mean annual avera ge daily traffic of a schoo l zone and the risk of school-aged pedestrians or bicyclists. Of all road segm e nts with traffic countS, 55.4% were located inside 10 school zones. Although a total of20 school zones were defin ed, 1 0 school zones were without traffic crash data or count data. Available and valid data are graphically illustrated in Figure 6.4. 58


s.oo ?;> '6 > 3.SO .:l !;> -g b . 2.00 < so SOOO 9000 13000 17000 21000 29000 Average Daily Traflic Volume Figure 6.4: Statistical Relationship between Average Daily Traffic and Injury Severity The mathematical relationship between average daily traffic and .average injuiy severity involving school-aged pedestrians or bicyclists is: .. Average Injury Severity -:2.243 +2.814 x 104 x Average Daily T raffic Volume The statistical analysis results are given in Table 6.14, Table 6 .15, and Table 6.16 . Goodness-of-fit statistics are provided in Table 6.14. Results of the ANOV A test and significance of model coefficients are illustrated in Tables 6.15 and 6.16, respectively. Table 6.14: Model Summary (Volume-Injury Severity Model) Model R RSquare AdjustedR Std. Error of Square the Estimate 1 0.227 0.052 .067 0.9591 Table 6.15: AN OVA Results Model Sum of Degrees of Mean F Significance Squares Freedom Square Statistic Level Regression 0.400 1 OAOO 0.435 0.528 Residual. 7.359 8 0.920 Total 7.759 9 59


Tab l e 6.16 : Significance of the Model Coefficients . Model Unstandardized Coefficients S tandardized t Significanc e C oefficients Statistic Leve l B Std Error Beta Constant 2.243 0.802 0.227 2.798 0.023 Difference 2.8 1 4*E S 0.000 0.660 0.528 . F r om the goodness fit-statistics, i t can be seen th a t there is a poor statistical r elationship between the average traffic volume and the injwy severity This o c currence may be due in part to th e quality of the d ata sets. In order to establish a stronger relationship betWeen average traffic volume and injwy severity, it is suggested that law enforcement and governme n tal agencies collect m ore traffic crash data of school-aged pedestrians/bicyclists. . Likewise, a traffic crash database containing accurate and complete information should be developed and maintained for school-aged pedestrian and bicyclists. 6 .2.3 1ll egal P assing of School Bu ses In this section a literature review was undertaken to obtain information regarding illegal passing of school buses The literature review was primarily based on two studies conducted by the Texas Transportation Institute (TI1) a t Texas A&M University (1984), and the Cent e r for Urban Transportation R e search (CUTR) at University of South F l orida (1996). Although there is a ten-year interval between the two studies, the research methodologies and study findings were nonethelesS similar to one another. At loading and unloading zones, there is a major concern for the safety of children boarding and exiting the school bus. This coneem stems from the high {requency of school buses being illegally passed. The main purpose of these two studies were to: (I) determine the number, location, and circumstances of illegal passes occurring on s chool bus routes; (2) obtaili. a greater knowledge and understanding of driver characteristics on the roadway network; and (3) identify and recommend areas for furthe r r esearch. The research methodologies and findings of the two studies are discussed 'in the following s ection. 60


. In these two studies, a school bus driver or an observer aboard a school bus was asked to record each individuai illegal passing. In the Texas study, researchers recorded illegal passing by direction, type of roadway, type of passing, and time of day (morning and afternoon) for a three-day peri_od. Observers also noted the manner in which vehicle passed stopped school buses (e.g., stopped .then passed, slowed then passed, or passed without slowing). Iri the Florida sll!dy, illegal passing data were recorded for two typical schools in 58 school districts. In addition to collecting similar information as the Texas study, the Florida study improved the method of investigation by compiling more detailed and useful . information. The follo\ving information was recorded in the Florida Study. time of occurrence (recorded in both Florida and Texas studies), number of students at the school bus stop, whether the vehicle passed traveling in the same or opposite direction as the school bus (recorded in both Florida and Texas studies), whether the vehicle passed on the left or right of the school bus, type of vehicle that passed-by, type of roadway (recorded both in Florida and Texas studies), whether the pass-by occurred in ao urban or rural locale, whether the pass-by occurred on a paved or unpaved roadway surface, and whether the strobe light was in use at the time of the pass-by. . . As recorded in the Florida Study, 10,590 vehicles illegally passed a total of 3,427 school buses. The total number of illegal passes (10,590) represented data from 58 of 67 school districts in Florida. Analysis of the data showed that approximately 89 percent of all school buses illegally passed during the field study were Type C buses, as displayed in Tap1e 6.17. Of the buses identified as Type C, approximately 10 percent were equipped with wheel chair lifts. This fmding might suggest that motorists illegally passed these particular buses more frequently due to the extra time it took for a student requiring a wheel chair to board or exit a school bus. As indicated by the illegal pMsing data, no explicit trend existed between the time of day and frequency of illegal passes. A close examination of data showed that illegal passing was. 61


almost evenly distributed between the morning hoius of 6:00AM to 10:00 AM (483 percent) and the afternoon hours of2:01 PM to 6:30PM (49 percent). The midday period from 10:01 AM t o 2:00 PM accounted for only 2.7 percent of th,e total recorded illegal passes, as indicated ijl Table 6.18 . Table 6.17: Illegal Passes by School Bus Type and Presence of Wheel Chair Lift (CUTR,J996). Type of School Percent without Percent with Total Bus wheel chjlir lift wheel chair lift A 0.3% 0.0% 03% B 1.8% 1.0% 2.8% c 79.5% 9.8% 89.3% D 7.2% 0.4% 7.6% .. Total 88.8% 11.2% 1000/o Table 6.18: Time of Illegal Passing (CUTR, 1996). Time Frequency Percent 6:00AM to 10:00 AM 4,727 48.3% 10:01 AMto2:00 PM 273 2.7% 2:01 PM to 6:30PM 4 ,796 49% Valid Observations 9,796 10()0/c, Missing Observations 794 NIA Table 6.19 implies that the majority of illegal passing occurred on two-lane roadways; nearly 56 percent of all recorded illegal passing occurred on this type of roadway. Furthermote, . 23.5 percent of all illegal passes occurred on four-lane roadways with two:way left-tum Janes. Tbis was an important finding since only a small percent of.bus stops in Florida are located on this type of roadway. Four-Jane and more. than four-Jane roadways with medians accounted for 15.1 and 5.6 percent of all Ulegal passing, respectively. 62


. Table 6.19:Type of Roadway (CUIR, 1996). TyPe of Roadway Frequeocy Percent 2Janes 5,836 55.8% 4 lanes, no median 2,456 23.5% 41an es, with median 1,583 15.1% More than 4 lanes with median 590 5.6% Valid observations 10,465 100% Missing observations 125 N/A The majority of illegal passing occurred when the vehicl e in violati on was travelling in the direction opposite the stopped school bus, accounting for 65.7 percent of all illegal passes . Vehicles traveling in the same direction as the stopped school bus accounted for the remaining 34.3 percent of the illegal passes. Table 6.20 shows the results rdated to the direction of the illegal passing. Tabl e 6.20: Vehicle Passed from the Same or Opposing Direction (CUTR, 1996). Direction of illegal passing Frequency Percent Opposing 6,691 65.7% Same 3,487 34.3% Valid observation 10,178 100% Missing observation 412 N/A . Swprisingly nearly four-percent of all reeorded illegal passes occurred on the loading and unloading side (right Side) of stopped school buses. While the remaining 96 percent occurred on the left side, as depicted in Table 6.21. A cross tabulation of the data indicated that approximately 78 percent of right-side illegal passes occurred on two-lane roadways as opposed to 13 percent on four-lane roadways with two-way left -tum Janes. While the 63


remaining 10 percent of illegal passes occuned on four or more lane roadways with median According to school bus drivers, passenger carS most often flagged for violation of school . bus stopping law. Table 6.22 shows that passenger cars accounted for 75.4 of the . . 10,590-recorded illegal passes. While light and heavy trucks accounted for 20.5 and 3.8 percent of all illegal passes, respectively. Interestingly enough, two police cars, one fire truck, and one school bus were observed passing a stopped school bus during the field study. Table 6.21 Vehi c l e Passed in the Left or Right of the School Bus (CUTR, 1996) Side Frequency Percent Left 9,887 96.0% Right 415 4.0% .. Valid observation 10,302 100% Missing observation 288 N/A Tabl e 6.22: Type of Vehicle that Illegally Passed (CUTR, 1996). Type ofVehicle Frequency Percent . Cat 7,900 75.4% L ight truck 2,149 20.5% Heavy truck 401 3.8% Van 18 0.2% Motorcycl e 3 <0.03% Police car 2 <0.02% Fire truck 1

Additionally, approximately 68 percent of all illegal passes occurred at school bus stops. . Most frequently, the range of 1 to 5 students were observed boarding or exiting the stopped school bus at the time of illegal passing, as shown in Table 6.23. Of the 3,427 stopped school buses illegally passed during the field study, 78.4 percent {2,687) had a roof-mounted white strobe light activated at the time of illegal passing as shown in Table 6.23. Table 6.23: Number ofStuilents at School Bus Stop (CIJTR. 1"996). Number of Students Frequency Percent 1 to 5 7,128 67.9% 6 to 10. 1,705 16.3% 11 or more 1,662 15.8% Valid observations 10,495 100% . Missing observations 95 N/A Table 6.24: Strobe Light in T:Jse at Time of Illegal Passing (CUTR, 1996) Strobe in Use Frequency Percent Yes 7,428 78.4% No 2,045 21.6% Valid observations 9,473 100% Missing observations 1,117 NIA 1n the Texas T r ansportation Institute stlidy, a driver's lack of understanding of Texas traffic . regulation law was identified to be the leadilig cause for the illegal passing of stopped school buses The survey results are as follows: 1. There is no requirement for vehicles to stop for yellow warning lights or hazard lights, yet . nearly half ( 48%) of the drivers surveyed said that they would stop for a yellow warning light and an additional31 percent for hazard lights. These inisinterpretations are errors of 6S


taution (i. e., stopping when it is inditating that a significant portio n of the driving public does not fully understand tb.e meaning of various signal configurations. 2. There is no requirement for traffic in opposing lanes of a divided highway to stop for a school bus displaying red loading lights; nevertheless 95 percent of tb.e drivers surveyed stated that they would stop. Once again this is an error of taution, clearly showing that the majority of drivers are not aware of state law on for school bus es on multi lane facilities. 3. With only red loading lights, 6.2 percent of the drivers indicated that tb.ey would proceed without stopping. This type of misunderstanding is termed as an error of recklessness (i.e., not stopping when netessary) Surprisingly, when the red loading and hazard lights are used simultaneously the error percentage actually intreased to over 10 pertenl This confirms that many drivers are confused with the meanings of various signal configurations . 6.2.4 Lack of Proper Sign11ge and Traffic Signals This section provides a general overview to the problem of lack of proper traffic signage and signals. Traffic signs and signals are important to the safety of school-aged pedestrian s and bicyclists because they aid in the safe interaction of yoWJg pedestrians/bicyclist s and motorists. M;any factors contribute to the effectiveness of traffic contro) devices (TCDs), including understandability, recognition time conspicuity, legibility distance, and leamability. The Manual on Uniforrn Traffic Control Devices for Streets and Highways (MUTCD) sets forth several basic requirements for TCDs (Federal Highway Administration, 1997). Motorists' understanding is an important element in the overall effectiveness of TCDs, since they will serve little purpose if ere misunderStood. A review was conduc ted on several studies relating to pedestrian/bicyclist traffic control devices. These studies provide an overview of the problems associated with pedestrian/bicyclist Ielated traffic control devices by examining: (1)' driver understanding of alternative pedestrian/bicyclist related traffic sigoage and signals; (2) pedestrian and bicyclist 66


understanding of alternative traffic and signals; and, (3) effectiveness of alternative pedestrian or bicyclist signs and signal mes.Sages. In 1995, Tidwell and Doyle conducted a nationwide survey to identify . misinterpreted traffic control devices. In this study, researchers evaluated a respondent's knowledge of pedestrian/bicyclist traffic control devices. Respondents were asked for their interpretation of 'WALK' and 'DON'T WALK' pedestrian traffic signals Also requested were the interpreted meanings of crossing and advance crossing signs. The survey questionnaires were distributed at driver's license examination centers in 48 states across the u.s. Also in 1995, Picha et al conducted a similar sutvey to identify frequently misinterpreted traffic signs and signals, and to further develop alternate device designs. The three-pha5e study included an evaluation of issues on the basis of a focus group, initial survey, and a follow-up survey. In addition to these studies on existing pedestrian/bicyclist traffic control devices, research was conducted to determine possible improvements in these signs and signals Dutt, Hummer, and Clark (1995) conducted a survey to evaluate the use of strong yellow green pedestrian crossing signs (the color of strong yellow green is an unassigned color listed in MUTCD). The surveys were perfonned at four locations with strong yellow green warning signs at mid-block pedestrian crossings Of primary concern is the effectiveness of pedestrian/bicyclist related traffic control devices. In .a study aimed at measuring the effectiveness of tra.ffic control devices, data were collected on pedestrian/bicyclist behavior, traffic and pedestrian volumes, pedestrian vehicle conflicts, and site characteristics (Zegeer, Cynecki, and Opiela, 1984). Video-recording and manual collection techniq11es were used to obtain the data A summary statistical analysis must be perfonned to analyze the data from the returned questionnaires. In Tidwell and Doyle's study (1995), the chi-squared test was used to analyze survey responses. This statistical test was chosen because of the categorical, non-67


normal nature of the. survey questions. The statistic was used to determine if significant differences existed between groups. In the study of "Evaluation of Innovative P edestrian Signalization Alternatives" (Zegeer, Cynecki, and opiela, I984), the z-test for proportions was used to determine whether the p roportion of occurrences in one group was significantly . different from that of a second group. PedeshianJBicyclist Related Traffic Signals The answers of many respondents were wrong when asked about the meaning of flashing 'DON'T WALK', 'wALK', and flashing 'WALK' messages (Tidwell, and Doyle, 1995, and Zegeer, 1984). Pedestrians and bicyclists are often given a false sense of security by the presence of a steady WALK' message. Pedestrians do not realize that conflicts With turning vehicles may still occur for a 'WALK' message. The 1978 MUTCD allowed the use of. !I flashing WALK message to warn pedestrians of tumingvehicles. However, this practice .was eliminated in the 1988 MUTCD because it was determined to be unclear. In 1984, Zeeger conducted a study to evaluate alternate pedestrian signals. The findings are as follows: I. The steady 'DON'T START' clearance indication (a three-lens pedestrian signal) resulted in a significant improvement over the flashing 'DON'T WALK' diSplay in tenns of pedestrian violations and associated clearance-related conflicts at the test sites . 2. The steady 'DON'T WALK' for the clearance interval provides no improvement over the flashing 'DON'T WALK' signal. 3. The 'WALK WITH CARE' display was tested in conjunction with a 'WALK' indication to warn pedestrians of turning vehicles. The 'WALK WITH CARE' message was found to be effective in reducing tum-related conflicts as well as pedestrian violations. Further analysis showed that these displays were effective for moderate to high right-tum volumes. 68


6.Z.4.i Pedestrian/B icyclist Relate d Traffic Sign s End School Zone Sign -In Tidwell and Do yle's study (1995), many participants realized that they c ould resum e normal vehicular speed after travelling through a school zone. All of the participant s indicated a strong prefer ence for the use of the current 'End School Zone' sign (SS-2) either alone or in combination with a speed limit (R2-l) sign. Sch_ool Advance Sign -Also in Tidwell and Doyle, for the standard sign (S 1 1 ), survey results indicated that 32 percent of the participants associated this sign with a school area and 46 percent associated i t with an advance warning of a school crosswalk. When the scho ol plaque was added below the standard sign the association of the "School Crosswalk increased significantly. Yield SJgn -In Zegeer's study (1984), a yield sign stating 'YIELD TO PEDESTRIANS WHEN TURNING' was found t o be eff ective in reducing turning conflicts, particularfy right tum conflicts The s i gn would be most appropriate on the right side of interseciion approaches, particularly where right turning motorists fail to yield righ t -of -way to Warning S ign Also in Zegeer's study, the pedestrian warning sign statin g 'PEDESTRIANS WATCH FOR TURNING VEIDCLES' was found to be effective in I: . reducing right tum conflicts. This sign would be appropriate in place of or in conjunction with the 'YIELD TO PEDESTRIANS WHEN TRUNING' sign discussed preViously. The 'PEDESTRIANS WATCH FOR TURNING VEIDCLES' sign could be applicable to sites with frequent incidents of right-tum pedestrian accidents. . 6.2.5 Lack of Drop off and Pick-up Z o nes School site design is critical to ensure a safe environment for bicyclists, pe_destrians, and motorists. On the school site, parent drop-off and pick-up points are areas where there are numerous vehicular movements and conflicts. In cases where busses are provided the problem is more acute. Although the issue of lack of drop-off and pick-up zones was identified as an important issue, there have been few studies conducted to address this concem. H owever, there are several guidelines for the design of new and improvement of 69


' existing scho. ol sites. In addition, each state has its own statutes governing the safety of children in and around school zones. It is recommended that site design with an evaluation of the location of the facility to intersections and arterials, commercial and residential centers, while co nsidering the potential for safe and egress by bicyclists; . pedestrians, and mot orists (University of Florida, 1993). Safe pedestrian and bicycle access to a school site must be evaluated in light of volume, characteristics, and classification of traffic. At parent drop off points, it is recommended that continuoUs sidewalks be constructed so children do not have 'to disembark roadways or parking lots. In school site design, there are two key safety principles that should be strictly :followed: (I) Adequate phySical space should be providedfor all modes of transportation to the degree that each is found on the school or planned for in the future, and (2) The physical routes provided for the basic school modes of transportation should be separated as much as possible from each other (Arizona Department of Transportation, 1983). Also, student-loading zones should be completely separated from bus zones. Increasing popularity of neighborhood based schools and pedestrian/bicycle friendly paths deerease the demand for buses. Bus loading zones should be physically separated from biking and walking routes, as well as parent pick-up and drop-off areas. 6.2.6 Summary In summary, vehicle speed has an impact on the safety of school-aged Pedestrians and . bicyclists. As posted speed of a roadway increases, so does the average injuiy severity of crashes. The results of the data analysis show that high traffic volume in and around school zones significantly impact the safety of school-aged pedestrians/bicyclists. Du e to a limited number of school zone traffic counts, the statistical relationship was found to be insignificant. . During the loading and unloading of students illegal passing of stopped school buses (with stop arms extended and red lights flashing) can jeopardize the safety of school-aged children. In evaluating the i ssue of lack of traffic signage, it was shown that pedestrians and bicyclists . . 70


. were often given a false sense of security by the presence of a steady 'WALK' message. As for t!ie issue of lad of pick-up and drop-off zones, arc few studies addressing the safety aspects 'of this topic. However, several agencies have published guidelines for the design of parent/school bus pick up and drop-off zones. 6.3 Young Drivers 6.3.1 Ge'!eral Lack of Driving Experi ence Most officials claim that inexperience and immaturity are the leading causes for crash involvement of young drivers, rather than alcoh ol consumption and teenag e risk taking: While young drivers learn basic skills to operat e a vehicle, they are often unprepared for difficult situations encountered during rush hour, inclement weather, and traffi c emergencies . Nighttime driving is especially a problem for young drivers; five times as many crashes occur during night hours t han during the day. According to Brown (1986), drivers undergo a three-step maturing process. The fttst stage, "car craft", young drivers learn maneuvering, braking, steering, shifting gears, changing . lanes, and choosing pr oper Janes. In the second stage, perceptual skills are developed which include proper visual scanning, knowledge of driving limitations due to weather and road conditions, ability to r ecognize hazardous situations, and anticipation of other drivers' actions. The final Stage drivers develop decision-making skills. These skills, B rown argues; are the most complex and take the longest time to develop. They involve judgement, reasoning, logic, and deciding when or how not to drive. As compared to olde r drivers, . mistakes in perception and decision-making are the common cause of crashes involving 16-19 year-old drivers 6.3.2 Lack of driver training programS An important factor contributing to the teen driver problem is the shrinking number of education programs in high schools. In 1975, 84% of all high schools in the U.S. offered driver education. By 1984, tbatnwnber had decreased to 68%, as of 1995, only 52% of 71


high schools in the U.S. had such prol!ramS. Aceording to the T ampa Tnllgpe. (10{01197) only 9"/o of students in Florida for driver education are actually enroll ed Even when driver educ11tion is available, the material discussed may not be entir e l y usefui to tod a y's young driver s. The c urriculum in most programs has not changed much over the last SO years. {Saunder s, 1998) Very little effort has been made to determine which topics are most appropriate for modem driver education programs. Another problem is that many of the current driver education teachers are not prope rly trained to instruct drivers 1n many cases, the teacher's area of expertise is in a completely unrelate d subject. Some instructors may not have had any training for driver skill instruction, while others only teach it part time. Often money is the major factor that governs the existence of high school driver education When education budgets are c ut, funds for driver education programs are usually among the first to get the ax. Allen Robinson, CEO of the American Driver and Safety Education Association, questions why parents are so willing to pa y for music and sports lessons, yet reluctant to cover the costs of formal driver training. Robinson points out that a lifetime of safe driving can save an individual $50;000, according to economists' . estimates, but school s, governm ents, and parents are unwill ing to spend $300 per studen t for driver education. 6 .3 .3 Lack of Seat Belt Use T raffic crashes impose a tremendous cost to society in terms of bodily injury, loss productivity, and most importa:ntly loss of life Each year, it is estimated that over 3 motorists are injured 41,000 killed, and ISO billion dollars lost as a res ult of tr affic crashes . Seat belts are a car occupant's easiest and most effective way of prev enting injury and . premature death. I f a seat belt is worn, front seat passengers are 45% less likely t o suffer fatal injuries in traffic Approximately 9,500 lives are saved each year from use of a scat belt. Their risk of moderate -to-critical injury is cut in half when sea t be lt s are worn. Drivers fall into one of three categories according t o the of seat belt usag e. Full-time users are more likely to be female, and claim that their primary reason for buckling up is to 72


avoid jnjury. While part-time users believe that seat belts lessen injury severity, they Usually don't use them on short, familiar, low-speed trips. Often they claim to be full-time users, since they do wear a seat belt when they feel they are at greater risk. Non-users constitute 510% of drivers. This group typically resents government regulation of their behavior, and believes that in a crash, seat belts would cause inore injury than they could prevent. Non users are unlikely to feel responsible for social and economic consequences of their (non-) action, and are more likely to drink and drive. The typical non-user is an unmarried male under the age of 30 with little college education. Traffic injuries are the leading cause of death among people age 6 to 27. Teens are especially vulnerable to injury and death in crashes because they use seat belts less than other drivers. Young people buckle their seat belts only 35% of the time, compared to an overall usage rate of 68%. Infrequent use of seat belts by teens is commonly attributed to peer pressure, feelings of invulnerability, greater risk-taking and tluill seeking, and other emotional stresses. Another possible exp11Uiation for low seat belt use among young people is that they often drive older vehic les. These cars make seat belts harder to use because of less sophisticated restraint systems. 6.4 International Tourists In the considerat ion of highway safety importance, international tourists received the fourth . highest ranking. The number of studies conducted in regard to this driving population subset is very limited. Available literature. in terms of critical issues/concerns of international . tourists is discussed in this section. 6.4.1 Difficulty in Recognizing and Understanding Traffic Signs In addressing the highway safety of international tourists, the most critical issue identified was the difficulty of tourists to recognize and understand traffic signs. The most important and probably the only study conducted and published in this area is "Evaluation of International Signing conducted by Wilbur Smith Associates for the Florida . Department of Transportation (FDOT, 1994). In this study, problems faced by international 73


tourists were identified by directly contacting them at the international aizports of Orlando and Miami. It was found that most international travelers' problems were related to guide signs, rather than regulatory or warning signs. Under two headings, Freeway or Interstate signs and State Route or Surface Arterial si,gns, the problems mentioned most frequently by international visitors are listed in Table 6.25. The problems are listed in the order of frequency of response. Table 6.25: Reported Problem Areas in Relation with Signs Category Problems Limite d Access Routes I. Advance warning of upcoming exits Exit Signs 2. Destination or route signing on exit signs 3. Destination signing to airports 4. Destination signing to car rental agencies Limited Access Routes I. Route signing to ocher interstates or routes General Signs 2. Consistency of lhe use of either a route Dallle cr a number for the same route 3. Use of cardinal directions on gulde signs 4. Route confirmation sigos State Route Signs 1. Trailblazing or destination signing on other routes or intersllltes 2 Destination signing to airports 3. Consistency of the usc of either a rout e name or number for the same route 4. Use of cudinal on guide signs S. Route confumation signs 6. Destination signs to car rental agencies 7. Use of S)I!Dbols In sign age 8 Use of cliagrom1110tie sips It was also discovered that each of the problems cited by international visitors was caused by one or more of the following six reasons: 1. The proper information was not available on the sign. . 2 The sign message was visually inadequate. 3. The infoiDlation on the sign was ina4equate. 4. The information on the sign was misleading. 5. The information on the sign was confusing. 6. Th e proper sign to convey the information was non-existen t. 74


The six underlying fact ors mentioned above Were divided into sub categories and the problem locations analyzed. In another study, Hawkins et .a!. (1996) dealt with Mexican driver understanding of traffic control devices in T e xas. For the purpose of this study, Mexican drivers were identified to be residents of Mexi co driving either automobiles .or commercial trucks in the bord e r areas of Texas. Even though this target population can not classified as international tourists they do have similar characteristics. The assessmen t of driver comprehension of traffic control devices in the study by Hawkins has provided a clear pictur e of the current issue. The results of the S\ll'Vey condu cted to determine the level of comprehension ofregulatory, warning, and other signs are given in Table 6.26. Table 6.26: Surve y Results for the Understanding ofTraffic Control De vices (a) Regulatory Signs Device R esponse Rate (Percent) Sample Size Corre<:t Partially Incorrect Not suro Unlcnown correct Stop sign 601 98 .7 0.0 o.s 0.2 0.7 Yield sign 603 63.9 0.0 21.S 13.6 1.0 Day/Night speed limit sign 601 82 .3 1 5.2 1.3 o.s 0 .1 Do not enter sign 582 90.1 o.o 4.6 3.4 1.2 One way sign 560 83 3 0.0 13.8 1.6 1 .3 Stop for school bus sign 557 51.9 24.2 14.6 3.3 0.0 Fasten seat belts law 587 33.2 23.2 s.s 36.6 I.S 75


(b) Other Signs . . Device Response Rate Sample (Percent) . Size Correct Partially Incorrect Not sure Unknown comet Road work ahead sign 519 803 1.0 2.9 14.9 0.9 Right lane ends sign 502 8.4 11.2 16.9 62.3 1.2 . (c) Warning Signs Device Response Rate Sample (Percent) Size Come! Partially Incorrect Notsure Unknown correct . . Curve sign with speed plat e 528 65.4 30.8 2.5 1.0 0.4 Two-way traffic sign 538 93.3 0.6 4.1 1.9 0 .2 Advance railroad warning 602 39.1 40. 5 6.S 12.8 1.2 School crossing sign 545 52..3 34.3 11.6 1.S OA 6.4.2 Lack of Information about Drivi .ng Laws and Customs in US The second most important issue when considering the highway safety of ipternational tourists is the lack of information concerning driving laws and customs In the aforem:entioned study by Wilbur Smith Associates lack of information was cited as the second most frequently mentioned factor among non-signing related issues. A video shown on flights prior to arriving in the U.S. was identified by the study as an importan t method of addressing this problem. This video developed by D ollar-Rent -a-Car discussed Florida driving and safety issues. 76 .


6.4.3 Other Issues As far as international tourists were concerned, the issues of unfamiliarity with the roadway system, confusion in translating English units into and unfamiliarity with the driving sysiem received third, fourth, and fifth rankings, respectively. However, these issues were not addressed in detail in any of the studies. It is also not possible to provide a completely familiar roadway and driving system for visitors coming from vastly different countries A coordinated effort is therefore required to educate and familiarize international visitors about the roadways, driving systems, laws, customs and conditions in Florida. 6.5 People with Disabilities The full participation of people with disabil.ities is important to society by les sening the . economic burden (societal costs) that would otherwise be used to take care of them . full participation is also essen t ial to their independent living and overall well being. H owever, the mobility needs of the disabled requires access to the transportatio n system be in a manner that is both safe and barrier-free. Several issues that are important to the highway safety of people with disabilities !U'e addressed in the following: Are intersections appropriately designed so that pe ople in wheelchairs .can cross streets in time? Is street crossing facilities acjequate for people with blindness? -How is safety ensured for people in wheelchairs to get around highway work zones? Are there standard methods of recalling licenses from people with disabilities? How safety of adapted for drivers with disabilities is ensured ? progressive The issue of private vehicle access for people with disabilities is furlher addressed in this study. lbis issue is related to not only highway safety of people with disabilities but also for tho traveling public in general. Four aspects of the issue are subsequently discussed. 77 ..


6.5.1 Persons with Driving Disabilities In accessing the market potential for vehicle adaptive equipment, TRB (1993) estimated the number of persons whose disability"either prevented driving or caused driving problems. Table 6 27 shows theses estimates in various degrees of driver disability. Over 60 percent of all people with driving disabilities drive without the need for special equipment. An additional 23 percent represent consumers who cumntly drive, but for whom adaptive equipment is either necessary or beneficial. The other 15 percent represent people for whom driving requires major vehicle modifications and adaptive equipment. These estimates do not include persons who do not drive and those who would not be able to drive even \vith the use of special equipment. Nor do these estimates include those who do not drive because of age, preferences, or other factors preventing them from driving. A more complete picture of the t otal driving population and its sub-groups is given in Figure 6.5. Table 6.27: Persons with Disabilities by Degree, 1987 US. Estimated Number Percent Degree of Driving Disability of Persons Total Drive Now without Special Equipment 12,000,000 63 But Would Benefit from Enhanced Design Drive Now without special equipment 3,820,000 20 but would benefit from.such equipment Drive now with special equipment 525,000 3 Health or disability prevents driving but 2,815,000 15 special equipment and design would enable to drive Total . 19,160,000 100 Source: TRB (1993). 78 .


. Toto! Orivlog . P.,uladon bocs Drive a Drive a Vcbiclc Vehicle 4<,Prcl0mlccs Prevmts Drivtr,a .......,, DriYin& . Special Spc

NCHS estimates that approxitriately 211,000 persons with disabilities used some type of adaptive equipment with their motor vehicles This estimate is based on a 1990 survey of the American public on a number of national health issues. The 1992 Truck Use Survey of the Bureau of Census obtained data from a sample of reSPondents. on the use of adaptive equipment in trucks. Based on this sample, NHTSA estimates that approximately 184,000 trucks opera t e with some type of vehicle adaptation. Every year, NASS/CDS collects elctensive information using in-dep!h investigations of a sample of approximately 6,000 crashes, involving light vehicles (passenger cars, light trucks, and vans) towed due to damage caused by the crash. Based on NASS/CDS for the period of 1995I 996, NHTSA estimates that 382,907 registered vehicles were fitted with some type of adaptive equipment. The . NCHS and Census estimates of the number of vehicles with adaptive equipment fall within the range of the 95 percent confidence limits of the NASS/CDS estimate. As a result of the Americans with Disabilities Act, NHTSA expects that the number of vehicles with adaptive equipment will continue to increase, as a larger proportion of the population begins to age and as access to employment, travel, and recreation continues to improve for persons with disabilities 6.5.3 Wheelchair Users Injuries and Deaths NCSA recently estimated the number of injuries and deaths of wheelchair users involved in motor vehicle crashes during the period 1991-1995 (Table 6.29). The estimation is based on the National Electronic Injury Surveillance System. (NEISS) of the Consumer Product Safety Commission. NEISS collects data on a nationally representative sample of persons treated in . hoSPital emergency rooJ!lS each year for injuries related to consumer NEISS is a 3-. level system consisting of monitoring of persons treated for injuries in hospital emergency rooms, follow-up telephone interviews with injured persons or witnesses, and comprehensive investigations with injured person s and/or witnesses. NEISS obtains data from a sample of 91 of the 6,127 hospitals nationwide with at least six beds that provide emergency care on a continuing 24-hour basis. NCSA used the NEISS data on injuries to wheelchair users to develop national estimates of the number of injuries and deaths to wheelchair users associated with situations involving motor vehicles . 80


Table 6.29: Wheelchair Users Injured or Killed from All Causes, 1991-1995. Estimated Percent Type Number of Total Persons Involving Motor Vehicles 7,121 2 Others 292,613 98 Total 299,734 100 Source : NHTSA (1997b) During the five-year period (1991-1995), a total of 299,734 persons in wheelchairs were estimated to be injured. More than 2 percent (7,121) of these persons were injured or killed in incidents involving motor vehicles Table 6.30 shows the estimated number of wheelchair users injured or killed by the type of vehicle involved. Vans were involved in almost half of the injuries to wheelchair users during the five year period. Pa5senger cars were involved in another 30 percent, with the remainder involving buses, ambulances, and trucks. Table 6 30: Wheelchair Users Injured or Killed by Type of Vehicle, 1991-1995. Type of Vehicle Estimated Number Percent Total of. Persons Van 3,410 48 Passenger Car 2,153 30 Bus 856 12 Ambulance 506 7 Truck 196 3 Total 7,121 100 Source: NHTSA (1997b) Table 6.31 shows the number of wheelchair users injured or killed by the type of activity . The five categories include improper or no securement of the wheel chair (in the motor . . 81 .


vehicle); collision a wheelchair and a motor vehicle; wheelchair lift maU\mction; to or from a motor vehicle; and falling onto or off of a ramp. Improper or no securemen t was involved in more than one-third of the cases. Another quart er was the result of a collision between a wheelchair and a motor vehicle . Tab le 6.31: Wheelchair Users Injured or Killed by Type of Activity, 1991-1995. Type of Activity Estimated Number Percent of Persons Total Improper or No Securement 2,494 35 Collis i on with Motor Vehicle 1,819 26 Li ft MaU\mction 1,366 19 Transferring to or from Motor Vehicle 1,035 15 Falling On/Off Ramp 407 6 Total 7,121 100 Source: NHTSA (1997b) Table 6.32 shows the number of wheelchair users injured or killed by the type of activity and . typ e of vehicle. None of the wheelchair users injured in situations of improper or no securement were associated with passenger cars. Of the 2,494 wheelchair users whose jJUuries were related to improper securement, 65 percent involved vans, while I 8 percent involved ambulances and 17 percent involved buses. Meanwhile, 83 percent of the wheelchair users whose injuries were rela ted 1o a collision with a motor vehicle involved p assenger cars. Trucks were involved in 8 percent while vans and buses were involved in 7 and 2 percent, respectively. Table 6.33 shows the number of wheelchair users injured by injury severity. The estimated number of wheelchair users killed i s 4 3, representing 0 6 percent of the wheelchair . users. All of these estimated fatalities were associated with collisions involving a motor vehi cle. 82


Table 6.32: Wheelchair Users Injured or Killed by Type or Vehicles and Selected Activity Type Type of Selected Activity Type Vehicle Improper or No S e curement Collis i o n with a Motor Vehi c le Estimated Number Percent Estimated Number Percent of Persons of Persons Van 1,616 65 122 7 Passenger Car 0 0 1,511 83 Bus 422 17 34 2 Ambulance 455 18 0 0 Truck 1 0 1 52 8 Total 2,4 9 4 100 1,81 9 100 Source: NHTSA (1997b) Tab l e 6.33: Wheelchair Users Injilred or Killed by Severity and Medical D isposition Estimated Number Percent Se y erity of Persons Total Death 43 0.6 Injury 7,078 99.4 Total 7,121 100 ; Source: NHTSA (1997b). 6 .5 4 Type of Vehicles Adapted and A d aptations Through NHTSA's web s i te, users (drivers and passengers) of vehicles with adap tive equipment are invited to complete a brief survey on the type of vehicle modified fo r use and the equipment and/or modifications in use. R e sults for the p eriod from May 19 9 7 to March 1998 are available. A total of 59 visitors to the NHTSA web si t e completed the s urvey O f these visitors, 36 were drivers of the adapted v ehicle. Ten of the visitors were passengers in 8 3 ._


the adapted vehicle, while 13 used !lie vebicie as both passenger and driver. .Table 6.34 . shows the type of vehicle modified for use by persons with disabilities. An equal number of respondents, 20 out of the 59 respondents, indicated that the adapted vehicle being used was either a passenger car or standard van. Another 17 respondents indicated that the modified . vehicle was a minivan. None of the respondents indicated that the modified vehicle was a heavy truck (over 10,000 pounds) or a motor home. Table 6.34: Type of Vehicle Adapted for Use by Persons with Disabilities Vehicle Type Number of Percent Respondents Total Passenger Car 20 34 Standard Van 20 34 Minivan 17 29 Sports Utility 1 2 Pickup Truck 1 2 Heavy Truck 0 0 Total 59 100 Source: NH:fSA (1998). Table 6.35 shows the equipment and modifications in use in their vehicles. The responses ranged from modifications for accommodating wheelchair users to vehicle control adaptations. A list of 25 specific vehicle modifications or adaptations is shown in the survey. Respondents could select as many choices from the list as were applicable. Only the five most often reported types of modification are included in the table 84


. Table 6.35: Top Five Modifications/Adaptations in Use by Persons with Disabilities Type of Modification Percent Respondents Use Hand Control 56 Wheelchair Securement 42 Lift 34 Automatic Door Opener 31 Steering Control Device 29 Source: NHTSA (1998). 6.6 New Immigrant s Immigrants are coming to this coun!Iy in record numbers. Many of these peop l e come t;om environments that vastly differ from that of this country. In addition to traditional factors related to highway safety (i.e., alcohol use, weather conditions, driver age, driver gender)" a set of unique issues exist for new immigrants. These 'include unfamiliarity with the transportation system, language barriers, cultural differences, and inadequate driver education. 6.6 1 Background Upon conta ct with a new culture, individuals may undergo a change in any or all of six areas of psychologicai function: language use, cognitive style, personality, identity, attitudes, and stress (Marin and Marin, 1991). This process of learning and adaptation can include an . initial stage of crisis or coo.flict that is often followed by the acceptance of an adaptation strategy. In terms of attitudes, an individual can adapt by assimilation, integration, or rejection of the attitudes prevalent in the new culture and environment' In terms of language, an individual may completely shift to English, become bilingual, or maintain their native language. The proces s of adaptation has been shown t o affect new immigrants' inental health status, levels of social support, levels of social deviance, alcoholism, and drug use; political


and social attitudes; and health behaviors such as the consumption of cigarettes and the use of preventive cancer screenillg practices. The process of adaptation may also affect the ease and rapidity with which new immigrants . . . become accustomed to the U.S. highway system. This ease and rapidity of integration by new immigrants may in tum affect their safety of highway travel as drivers, passengers, or pedestrians (NJITSA, 1995). 6.6.2 The Issues Nc\v immigrants ate up.famlliar with the transportation system because of vast differences in transportation practices of the U.S. as compared to that of their home countrieS. In some instances, walking or bicycling may have been their primary mode of transportation in their counby. As pedestrians or bicyclists, they may not be accustomed to highspeed traffic . They may be forced to learn to drive in the typically low--density environment. This task of learning to drive can be more difficult for them than for riatives, since immigrants face language barriers and cultural differences. Often, new inunigrants do not have the opportunity to observe the manner in which people drive safely. To make thing s worse, many of new immigrants are taught by fellow peers, who themselves are new drivers Furthermore, many immigrants are either teenagers or elderly. These barriers, differences, and difficulties further add t o the already high risks experienced bY teenagers and elderly. 6.6.3 Evidence Several of these issues emerged in a recent study of highway safety needs of Hispanic communities iJi the United States (NJITSA, 1995). High concentrations of Hispanic population were selected from California, Colorado, Florida, New Jersey, New York, and Texas and the District of Columbia. The study involved telephone and onsite discussions with representatives from highway safety, law enforcement, emergency medical services, liealth, education, and general service agencies and organizations in the study areas. In addition, the study conducted nearly 50 focus groups of community members from the study . 86


areas. Among others, the study identified segments of the Hispanic communities with speeia! highway safety problems. One of these segments included new immigrantS. New immigrants were second to young males in the frequency with which study participants . . mentioned vehicular and pedestrian safety problems. Often these two segments of the population were mentioned together. Some agency and organization repres entatives stated that new immigrants could actually be the segment of the community with the greatest need, since they are not accustomed to new driving conditions, language, customs, and laws. New immigrants were most frequently mentioned in areas that experienced or is expected to expe.rj.ence a large number of new arrivals particularly Te>eas and Calif ornia. Representatives of highway safety organizations reported that one of the major for new immigrants who come from rural areas is the disorientation they may feel as they try to become familiar with life in a large city. The sheer volume of traffic, signs, and roadway configuratio n s are unfamiliar and confusing even to those who have substantia l driving experience Part i cipants attributed unsafe driving behavior of new immigrants to habits brought from their countries of origin. A participant in Florida indicated that unsafe driving behavior is due to the view that driving is more of a social event than a means of transportation. Language barriers and cultural differences create problems in a number of aspects rdated to highway safety. These include outreach efforts by highway safety organizations, law enforcement activities, emergency medical services, and efforts by educational inStitutions. 87


CHAPTER 7: MATRICES -. lbis chapter presents countenneasures for issues and concerns identified earlier. These counten:neasures are intended to address the problems that were identified for each special population group. Identification of these countermeasures was based on the results of a literature review and the current practices related to the methods of improving safety. The suitable countermeasures are summarized and presented in the form of tables or matrices for clarification. 7. 1 Older Dri v ers lbis section discusses countermeasures to the safety problems of o lder drivers. Driver education is one of the countermeasures suggested here and in many other studies conducted earlier. This is based on the evaluation of the driving knowledge of ol der drivers. The evaluation has indicated that they have deficiencies primarily in righ t of way rules, procedures for crossing and making left turns at intersections, safe following distances and lane positioning in freeways, driving in congestion, backing out and parking procedures, etc. Therefore, driver education is necessary t o address the identified critical issues directly or indirectly. Previous researchers have ruled out the necessity for developing new driver education programs in the presence of three nationally prominent driver ed ucation programs (McCoy et al. 1992). They are (I) Coaching the Mature Driver (National Safety Council), (2 ) SS Alive/Mature Driving (American Association of Retired Persons) and (3) Safe . Driving for Mature Drivers (American Automobile Association). All the three programs have been recogniml as those addressing the driving deficiencies or older drivers adequately. Another countermeas ure that bas been identified i s physical and perceptual therapy. They can be used to improve the performance of older drivers. The study by McCoy et al., mentioned earlier, employed physical therapy. This involved exercises designed to improve trunk rotation, shoulder flexibility and posture. It was observed that overall improvements. iii physical fitness and flexibility would lead to improving reaction time, cognitive performance, and energy levels. With regard to the perceptual therapy, they employed visual perception 88

PAGE 100

exercises designed to improve spatial orientation, visual discrimination, figure ground perception, visual closure ana visual memory. These exercises provided significant improvements to the older drivers. Anothe r counte rmeasure towards improving safety of older drivers is employing licensing a:nd screening measures, so that those drivers who experience certain incapab ilities are kept away from these activities. Actions such as age-based renewal procedures would increase not only the safety of older drivers but also that of the overall driving population. However, careful consideration and further studies are necessary because the counterme asure reduces the mobility of older drivers. This leads to issues of individual rights versu s benefits to soci ety. I n the light of advancements in Intelligen t Transportation Systems (ITS), vehicular . improvements can be employed as another countermeasure capable o f addressing several critical safety issues and concerns identified for older drivers. Vehic les equipped with sophisticated instruments that would make "hands-off' navigation possible will be the u ltima te solution for special population groups like older drivers. The role of safety devices l ike seat belts and air bags are considerable too. Among the other driving optio ns available, t o make driving easier is automatic transmission, power steering, press button controls for windows, adjustab l e steering wheel and seats. The countermeasures mentioned earlier like driver education programs therapies, and vehicular impro vements are effective. only for those people who can afford th e service. Unlike those mention ed above, engineering countermeasures are applicable to all road users . making them mor e effective. Several engineering countermeasures suggested addressing the critical issues and concerns identified for older drivers are improving roadway, signage, signal timing and pavement marking are. The roadway improvements suggested primarily involve intersection design improvements since the older drivers experi ence greater difficulties in processing information and making quick decisions at intersections. I nters ection design impro vements, that could be considered in reducing the older driver crashes at these locati ons, include providing left-tum lanes, increasing sight distances, 89

PAGE 101

simpufying intersection configurations, providing proper roadway lighting and xe-evaluating pedestrian walking speeds to provide sufficient timings. In addition to roadway design improvements mentioned earlier, signage improvements on highways are suggested for further consideration. Highway signs are vital in passing information to the driver. Most of the information drivers receive from the highways' information system is received from the signs. They provide route guidance, warning, regulatory, and advisory information. Improving highway sign visibility is one of the major parameters that strongly impact the driving performance of older drivers. Visibility of highway signs can be improved in several ways: (1) increase Jetter size, (2) improve conspicuity of signs, (3) multiple signage, and ( 4) use of symbols on signs. Improvements in pavement markings can be implemented to achieve better safety for older drivers. All drivers use markings and delineation to help track or follow the road. Some forms of marking and delineation are reflectorized, thus helping drivers track the road both during day and night. Thus, the emphasis should be on the installation and maintenance of good pavement marking and delineation system. The system would include the need for solid edge lines, longer dashes, a shorter stripe-to-gap cycle, painted and reflectorized guardrails, and delineator posts. Emphasis should be on the maintenance procedure where they must be regularly inspected to ensure that they meet minimum standards and specifications. This is particularly true for degree of brightness and contrast. Signal timing improvements may be necessary, in particular, changes to the yellow int erval to address the slower perception reaction times exhibited by older drivers as a result of their decreased mental and physical capabilities. Re-evaluating the perception reaction times of older drivers at intersections is necessary and the currently accepted values of perception brake reaction time in computing stopping sight distances (2.5 seconds) needs to be checked for sufficiency. Reducing speeds by road design and traffic control, in areas that require complex maneuvers, may also improve the safety situation of older drivers. Complex situations in dense or high speed iraffic can be eliminated wherever possible as a countermeasure. Such measures can 90

PAGE 102

be accomplished by application of one way systems, prohibiting dangerous Janes, by avoiding or e limin ating level junctions and by separ ating flows. All the countermeasures mentioned earlier and their capability to address each of the issues . an d concerns identified are cri tical for older drivers They are summarized in Table 7 .1, in the form of a matrix. 7 .1. School-Aged Children In this section, the countermeasures for the issues that were addressed previously are presented, based on extensive review of publications related to school-aged children and transportation safety Authorities from transportation engineering safety enforcement, education, and o ther organizations throughout the nation prepared these publications. Although each individual issue has its specific countermeasures, it is strongly that when facing any of the complex issues of highway safety of school-aged children, a systematic analysis of th e problem be performed, and all the possib le countermeasures be considere d These procedures may include identification of all possible defic i encies of a child's route with a clear understanding of the prevailing transportation characteristics, se lection of ro ute improvements and control measures, implementation of route improvemen ts and periodic evaluation of routes . Definite procedures and a step-by-step p rocess needs to be established to help the various local agencies and groups to work t ogether. These local and groups involve police traffic engineers (Engineering), schoo.l administra!ion (Education), and parents. 91

PAGE 103

Table 7.1: The Countenneasure Matrix for the Safety Issues of Older DrivelS Countermeasure Physical Licensing Vehicular Roadway Speed Signal Signing Pavement Driver and and Improve-lmrove-Related Timing Improve-Marking Education Perceptual Screening ments ments Measures Im provements Improve. Therapies Measures ments ments Issue/C o n cern NightTime X X X Visibility Driving in X X X Congestion Freeway Driving X X X .. Maneuvering X X X Curves . Deficiencies in X X X X Driving Knowledge Location and Size X X X X X X of traffic S igns X . and Letteri n g Perception X X X X X X X X Reaction Time . Gap Acceptance X X X X Nan'Ow lanes X X X . 92

PAGE 104

7 .2.1 Speeding in and around School Zones The countermeasures for. the issue of speeding in and around school zones can be discussed under three main .categories. They engineering, education and enforcement. Engineering Countermeasures InterucJion Improvement-Intersections, in particular signalized intersections, are the most . dangerous part of the road network for pedestrians. Most pedestrian fatalities iii Florida . occur at intersections. There are 32 possible vehicle-pedestrian conflicts at a 4-legged interse ction (Florida Department of Transportation, 1992). It is recommended in the Florida Pedestrian Plan (Florida Department of Transportation, 1992) that intersection areas (conflict zone) be kept to a minimum to reduce the exposure of a pedestrian/bicyclist to vehicles, -reduce pedestrian crossing distances and increase sight distances. These practices make the vehicle path clearer and reduce the relative speed between opposing movements. Also, channelization with medians, and right tum slip lanes with channelization islands can reduce the conflict zone and provide safe refuge for pedest;rians. Furthermore, prohibited turns that are dangerous to pedestrians can be blocked. As for residential neighborhood, streets, can be an effective treatment for reducing pedestrian/vehicle conflicts and vehicle speeds (Florida Department of Transportation, 1992). fu crossing the roundabout, pedestrians only need to cope with one direction movement. Traffic Calming -Traffic calming aims to reduce the dominance and speed of motor vehicles. It employs a variety of techniques to cut vehicle sPeeds. Measures include physical . alterations tO. the horizontal and vertical alignment Of the road and Cblll\ges in priority (Cyclists' Touring Club; 1991). Special facilities for bicycie traffic are rarely seen in areas subject to .traffic restraint or low speed limits. This offers a measure of one of the many inherent benefits traffic calming holds for: cyclists. They can move around in relative safety when riding in slower traffic. These measures may be more appropriate in older and . narrower streets, which have Jess scope for satisfactozy cycle paths or other special cycling facilities. A wide range of people are benefited by traffic calming -residents, pedestrians and 93 .

PAGE 105

cycliststhus obviating the need t o wrestle with financial justifications for cycle facilities, such as segregated provision and special engineering treatment at junctions. The following general design guidelines should be followed in the implementation of traffic calming schemes (Cyclists' Touring Club, 1991): (1) Where possible, provide cyclists with an alternative that by-passes physical obstacles such as chicanes or ramps, (2) Where a reduction in carriageway width is employed as a speed control measure, careful consideration should be given to how motorists and cyclists can safely share the remaining space, (3) Surface materials, particularly on ramps, should have a good skid resistance, while . teKtured surfaces should not be so rough that they endanger the stability of cyclists or cause severe grazing if the cyclist should come off, ( 4) A smooth transition on entry and exit to ramps should be provided. Inclines should be .. clearly indicated, (5) If the traffic-calming feature is to be installed on a road with an overall gradient, it must be noted that cyclists are likely to approach it at different speeds uphill and downhill. Tbis should be taken into consideration while designing the feature. In addition, three general observations could be made from successful traffic calming schemes (Cyclists' Touring Club, 1991): (1) Where consistent low speeds Qess than 20 mph) are required, such as in residential areas, physical traffic calming features shoul d be positioned sufficiently close to each other to deter unnecessary acceleration and braking, . (2) The use of appropriate signage is'important to remind drivers that they are entering a traffic restraint area Public awareness campaigns facilitate the acceptance of lower speed, (3) Sympathetic speed limits are used to reinforce the physical speed control measures Possible Policy Change -Medians are effective measures to reduce exposure of the pedestrian to traffic while crossing a roadway. Refuge islands and medians help to eliminate 94

PAGE 106

. pedestrian/bicycle-vehicle traffic conflicts by providing a "refuge" or safe retreat from oncoming traffic. Medians are recommended in the Florida Pedestrian Plan (Florida Department of Transportation, 1992), w herever the crossing distance exceeds 60 feet, to provide a refuge for slow or late crossing pedestrians. Pushbuttons should be installed in the median. Also, handicap ramps or a full cut should be provided through the median. Preferably, the refuge islands should be a minimum of 6 feet in width and in no case less than 4 feet wide t o reduce potential danger to the island users. Poss ible Design Chang e -Many signalized intersections are pedestrian unfriendly bec;ause of the speed and complexity of vehicle movements and the number o f lan es added for capacity of the roadway As recommended in the Florida Pedestrian Plan (Florida . Department of Transportation, 1992) one-way pair streets, slip lanes and medians may be used to reduce the number of lanes that need to be crossed. Also, consideratio n should be given to roundabouts at intersections in residential neighborhoods as they effectively reduce vehicle speed and pedestrian/bicyclist-vehicle conflicts . Education Count ermeas ures Children's Education -Research of nationwide school-aged pedestrian/bicyclist crash data reveals that the most significant factor is the irresponsible and unpredictable behavior of the pedestrian or bicyclist. Typical.examples of unsafe behavior of school-aged pedestrians and bicyclists includ e playin g in the roadway, not crossing at intersections, and walking or biking at night without prOPer reflective clothing . Hence, the education for sch ool-aged pedestrians and bicyclists, about proper attire, traffic regulations, and safety issues, is vital. Information campaigns on safety are substantially inadequate in teaching children life-iong skills in assessing gap acceptance, route planning, and other vital traffic sldlls. Regular, repetitive and a cumulative education program that offers a minimum of ten lessons per year t?etween kindergart en aild grade 2 is essential to develo p minimal traffic competency. As indica t e d in the Florida Pedestrian Plan (Florida of Transportation, 1992), the traffic safety education program should be supported in these ways: (1) implementation of traffic safety 95

PAGE 107

education program for 5-9 year olds iii school, (2) conducting mass media campaigns, through family and child-orientated companies such as AAA, food, drug, fast food, and toy stores, to promote safe walking skills, and (3) training of crossing guards, bus drivers, and school liaison officers to reinforce the classroom safety lessons. This could be done by county Sheriff's and Police Chiefs' Associations. AAA also has material for adult crossing guard training. Other Roadway Users' Education -The general public need to be educated about the severity and the nature of pedestrian safety problems. Pedestrian safety education should be included in driver education classes. It is essential to gain public backing and support. Such campaigns should include secondary school and college students. They should target areas with high incidence of school-aged pedestrian/bicyclist problems. The campaign could use various mass media such as television, newspapers, pamphlets, video, and other materials . 7 .2.1.3 Enforcement Countermeasures Enforcement of pedestrian and bicyclist safety issues will be an important factor in bolstering pedestrian systems. Strict monitoring of the actions of the pedestrian and the automobile driver reinforce the view that pedestrians are a vital part of the transportation system. Enforcement of the basic safety strategies would make society more aware of other modes of travel. Warnings should be issued to those who do not follow these safety precautions. Enforcement methods would help the population realize that the transportation system should be equally accessible and safe for all users. The population should also understand that all . . transportation modes are required to follow the laws of the state. . PedestriansPedestrians need to follow the rules of the road and should be given warnings if they disregard them. A widely unknown practice is that pedestrians. should travel against traffic and should not travel in the same direction as traffic along a roadway. Florida Statute 316.130 (4) states "Where pedestrian facilities are not provided, any pedestrian walking al
PAGE 108

EnforcemenJ Officers -It is recoriunended in the Florida Pedestrian Plan (Florida Department of Transportation, 1992) that officers receive more training in completing crash report foims and gathering evidence concerning pedestrian-vehicle crash reconstruction The . evidence includes speed of the vehicle location of impact, and weather conditions It is also recommended that law enforcement officers involve themselves in community programming. Summary of the countermeasures to speeding in and around school zones is given in Table 7.2. 7.2.2 High Traffic Volume around School Zones Engineering Countermeasures Higher motor vehicle traffic volumes represent greater risk for bicyclists. More frequent . passing situations are less comfortable for school-aged pedestrians and bicyclists unless special design treatments are provided. In a study conducted by the Federal Highway Administration (1994), the traffic operations and design recommendations (as shown in Table 7.3, Table 7.4, and Table 7.5) are given based on three ranges of Average Annual Daily Traffic (AADT)-under 2,000 AADT; 2,000 to 10,000 AADT; and over 10,000 AADT. 97

PAGE 109

Table 7.2: Countenneasures for Speeding in and around School Zones. 12 (a) By Category Engineering Countermeasures Roadway section I. Build traffic calming area around schools to reduce vehicle dominance and speed. 2. Build refuge islands and medians on roads with high traffic volume and speeds to reduce exposure of the pedestrian when crossing a roadway Signalimi intersection I Build one-way pair streets . 2. Build medians to reduce number of lanes to be crossed. Residential neighborhood 1. Build roundabouts at intersections in residential neighborhoods to reduce pedestrian-vehicle conflicts and vehicle speed. Education Countermeasure . Children l. Implement traffic safety education program for S-9 year olds. 2. Conduct mass media campaign. 3. Implement training programs for crossing guards, bus drivers, and school officers. Drivers 1 : Pedestrian and bicyclist safety education should be taught in drivers' education classes. 2. Conduct mass media carnpaigo_ . Enforcement Cotmtermeasure Pedestrians I. Give warnings or citations to pedestrians or bicyclists if they do not follow the road rules. Enforcement Officers I. Give enforcement officers more training in completing traffic crash reports and in gathering evidence concerning pedestrian and bicyclist crash reconstructioit 98

PAGE 110

7.2 (b) By Type of Countermeasure Engineering Countermeasures I. Provide highway signage in school bus loading and unloading zones 2. Improve visibility of bus stop signal ann 3. Provide lettering on rear of school bus. Education Countermea sures I. Develop and distribute materials related to the school bus stop law. 2. Analyze lhe basic school bus curriculum and make necessary changes. 0 Enfortement Countermeasures 1. Promote awareness and focus efforts in the direction of school bus stop law enforcement wilhin the state law enforcement . community . 0 0 0 99

PAGE 111

. Table 7.3: Traffic Operations and Design Recommendations for New Adult and Teenage Bicyclists on Urban Section without Parking. A verage motor Average annual daily traffic (AADT) volume vehicle Less than 2,000 2,000-10,000 Over 10,000 operating speed Adequate sight Inadequate Adequate sight Inadequ ate Adequate sight Inadequate distance sight distance distance sight distance distance sight distance Lessthan30 wcl4 truck bus, RV wcl4 wcl4 Truck, bus RV wcl4 .blS truck, bus, RV bl s mph wc14 wcl4 wc14 wcl4 blS bl5 i . 30-40mph bl5 bl5 bl5 bl5 bl5 bl6 bl6 bl5 bl s bl6 bl6 bl5 ' bl5 bl5 bl5 blS bl6 bl6 bl6 bl6 bl6 bl6 bl6 bl6 I Over 50 mph bl6 bl6 b16 bl6 b16 bl6 bl6 b16 bl6 bl6 bl6 bl6 Key: wc=wide curb l ane**, sh=shoulder, slshared l ane, bl bik e l ane, na'"llot applicable we numbers represent "usable width" of outer lanes, measured from lane stripe to edge of gutter pan, rather than to face of curb. If . no gutter pan is provided, add a minimum of 1-ft for shy distance from face to curb. BL numbers indicate minimum width from the curb face. The !?ike lane stripe should lie at least 4 ft from the edge of the gutter pan, . unless the gutter pan is built with adequate width to serve as a bike lane by itself . . ' 100

PAGE 112

Table 7.4: Traffi c Opetations and Design Recommendations for New Adul t and Teenage Bicyclists on Urban Section, With Parking. Average motor Average annual daily traffic (AAD1) volume vehicle Less than 2,000 2,000-10,000 Over 10,000 operating speed Adequate sight Inadequate Adequate sight Inad equat e Adequate sight Inadequate . distance sight distance distance sight distance distance sigh t distance tess than 30 wc14 truck, bus, RV wcl4 wcl4 truck, bus, RV wcl4 bl5 truck, bus, RV bl5 mph wcl4 wcl4 wcl4 wcl4 bl s bl s 30-40mph bl 5 bl 5 bl5 bl5 bl 5 bl6 bl6 bl s bl6 bl6 bl6 bl6 41-50mph bl 5 bl s bl5 bl5 bl6 bl6 bl6 bl6 bl6 bl6 bl6 bl6 Over 50 mph oa na na na na na na na na na na na Key: WC"'Wide eurb lane .. sh=
PAGE 113

. Table 7oS: Traffic Operations and Design Recommendations for New Adult and Teenage Bicyclists on Run!! Section 0 0 A v erage annual daily traffic (AAD'D volume 0 0 Average motor Less than 2,000 2,000-10,000 Over 10,000 vehicle Adequate sight Inadequate Adequate sight Inadequate Adequate sigbt Inadequate operating speed distance sight distance distance sight distance distance sight distance Lessthan30 sh4 truck, bus, RV sh4 sh4 truck, bus, RV sh4 sh4 truck, bus, RV sh4 mph 0 sh4 sh4 sh4 sh4 sh4 sh4 0 0 30-40mph sh4 sh4 sh4 sh4 sh4 sh6 sh6 sh4 sh6 sh6 sh6 sh6 0 41-SOmph sh6 sh6 sh6 sh6 sh6 s'h 6 sh6 sh6 sh6 sh6 sh6 sh6 Over 50 mph sh6 sh6 0 sh6 sb6 0 sh 8 sh8 sh 8 sh8 sh8 sh8 sh 8 sh 8 Key: wc=wide curb lane, sh=sho.ulder, sl=shared lane, bl-bike lane, na"''lot applicable ., 102

PAGE 114

Four basic facilities are used to accommodate bicyclists These facilities are: {I) shared lane: shared motor vehicle/bicycle use of a "standard" width travel lane, (2) wide outside lane: an outside travel lane with a minimum width of 14 ft, (3) bike lane-a portion of the roadway designated by striping, signage, and/or pavement for preferential or e)Cclusive use for bicycles, and (4) shoulder-a facility physically separated from the ro adway and intended for bicycle use. It is also recommended that at school crossings with high traffic volumes, students be provided with home-school route map (Institute of T ransportation Engineers, 1984). The map should include location. of crosswalks, crossing guards, traffic signals, sidewalks, and bike paths 7.2.3 Illega l Passing of School Buses In the study entitled "lllegal Passing of Stopped School Buses in Florida" performed by the University of South Florida (CUTR, 1996), countermeasures were recomme nded :_"and organized into actions at both the state and local levels At the state level; the Florida Commissioner of Education, the Florida Lcgislatute, and other relevant groups representing law enforcement, planning and local school districts should work together to develop and implement practical and effective countermeasures to this problem. Their responsibilities include: {I) Promotion of awareness and fOC1JSSing of efforts in the direction of school bus stop law enforcement within the state law enforcement community, (2) The development of high impact public service announcements in television, radio, and newspapers to educate private motorists about the school bus stop law and graphically remind them ofthe potential consequences the'law; (3) The development and distribution of additional material related to school bus stop law. Information pertaining to traffic stopping for a stopped school bus should be included in automobile license tag notices, rental car conlnlct signoffs, and ?n billboards; (4) Clarification of the section in the Florida Driver's Handbook that pertains to the 6Chool bus stop law; 103

PAGE 115

(5) Providing highway signage in school bus loading and unloading zones, advising motorists of the law pertaining to sto.pping for stopped school buses, and to inform motorists of areas where school buses make frequent stops; (6) Conducting research on possible changes to the School Bus Specifi cations by the State Board of Education or pilot testing to include higher visibility of school bus stop signal arms, lettering on the rear of school buses advising motorists of school bus stop l!lw, or other safety measures that may be proposed such as video cameras mounted on school buses to record the vehicles that illegally pass; (7) The Department of Education analyzmg the Basic School Bus Driver Curriculum and making necessary changes t o standardize the school bus stop warning procedure; and . (8) The Department of Education identifying best practices and m aking recomme ndations for school districts regarding the establishment of safe school bus routes and stops. In addition to the above-recommended countermeasures,, it is suggested in the Florida study tha t the existing law pertaining to illegal passing of stopped school buses be amended to better reflect the current driving environment. The slUlllilary of the poss ible is shown in Table 7.6. Table 7.6: Countermeasures ofl!legal Passing of School Bus Bngineering Countermeasures Provide highway signage in school bus loading and unloading zones Improve the visibility of stop signal ann, put lettering on rear of school bus . Education Countermeasures Develop and distribute materials related to the school bus stop law. Analyze the basic school bus driver curriculum and make necessary changes. Bnforcement Countermeasures Promote awareness and focussing of efforts in the direction of school bus stop law enforcemen t . within the state law enforcement community. 104

PAGE 116

7.2.5 Lack of Proper Traffic Signage and Traffic Sigilals Traffic Control Signs-The fundamental purpose of traffic control signs is to convey a message to motorist or pedestrian. Many studies h
PAGE 117

. (7) Beacons, activated dwing periods when a sign is applicable, may be used to improve the effectiveness of one or more of the abOve mentioned signs by drawing . motorists' attention to these devices (Institute of T ransportation Engineers, 1984); and, (8) Parking SiSJ;tS are intended to inform lllotorists of parking regulations that )lave been established to aJfeviat e hazards that would exist if parking were permitted. Stopping and . parking of vehicles should be prohibited at locations where it will limit visibility of pedestrians using crosswalks It should also be prohibited if school grounds are very near to the curb or roadway edge such that careless activities of the children may come in c onflict with passing traffic (Institute of Transportation Engineers, 1984). For lack of proper signage, the law enforcement countermeasures are as follows (I'idwell, and Doyle, 1995): (1) Drivers who do not yield to the right-of-way of pedestrians at the appropriate time. sho\lld -be given a citation; and (2) Pedestrians who exhibit unsafe behavior should be given a citation, particularly in corridors that have traditional pedestrian safety problems; Traffic Control Signals In the long term, traffic signals are more cost effectiv e compared to crossing guards. These signals may be coordinated to provide efficient traffic flow. In some cases installation of a school-crossing signal may not eliminate the need for a crossing guard (Institute of Transportation Engineers, 1984). & suggested by many studies, the following traffic engineering countermeasures are provided for lack of effective traffic signalization: (I) Mid-block signalized school crossings should not be considered until all other alternatives have been studied; and (2) Mid-b l ock traffic signals for school crossings should always be pedestrian actuated unless the signal is designed to assist in preserving platoon discipline through a progressive signal system (Institute ofTransport ation Engineers, 1984). At intersections, right-tum-on-red traffic in
PAGE 118

drivers focus their attention to the left prior to negotiatiDg the tum (Florida Department of Transportation, 1992) The following recommendations were made t o improve traffic signals: (I) Pedestrian signalizati on should be implem ented at urban signalized interse ctions where field studies warrant. Pedestrian signal heads should be mounted on poles that suppo rt the push buttons so that they relate to the signal display. All s ignal heads should be brought up to current MUTCD standards; (2) Where diag0nal spans supporting traffic signal heads prevent pedestrians from seeing the current vehicle phase, existing span wire installation should be removed and a new traffic signal should be installed using pole/mast arm signals or box spans; (3) When the view of a driver approaching an int ersection is restricted, the comer should be cleared of all sight obstructions by using joint-use poles to support traffic signals, street .' names, and lighting. All other items like trees and shrubs should be relocated or removed; and (4) Whenever feasible, l eft tum movements at signalized interseciions in downtown or commercial zones should be restricted by protective left tum phasing, Pavemmt Markings-Pavement markings are used to mark pedestri an crossings arid emphasize traffic control by supplementary warning and regulatory s igns. All markings must confom to the MUTCD. Strict adherence to the following procedures is suggested while using pavement markings (Institute of Transporta tion Engineers, 1984): (1) Paveinent marking s should never be used as the sole protection for children in the vici.llity of schools markings may become ineffective due to wet or snowy . conditions on pav ement, l ack of adequate maintenance, or when hidden from the vi e w of motorists b y vehicles ahead. Markings should always supple ment standard traffic signs; and (2) Special school pavement markings should be limited to those r oadways that are in the vicinity of the sch oo l grounds The marking of roadways at considerable distances from the school grounds tends to weaken the effectiveness of such installations. 107

PAGE 119

7.2.6 Lack of Pick-up and Drop-off Zones The parents and school bus pick-up and drop-off zone is a critical element to the school site design. A good school site design should stress a maxnnum separation of the three . basic modes of transportation: {I) school bus; (2) cars and motorcycles, and (3) pedestrians and bicyclists. It is recommended in the School Trip Safety Program Guidelines published by the Arizona Department ofT ransportation that the two key safety principles to be rigidly adhered to are: (I) Provision of adequate physical space for all modes of transportation to the degree that each is found on the school site or planned for in the future, and (2) Physical routes be provided for the basic modes of transportation to school and they be separated as much as possible from each other. As suggested in these guidelines, it is absolutely essential that bus zones be separated from all other transportation activities. An estimate of the number of buses expected to be within the loading and unloading zone at any given period of time will dictate the dimensions of the school bus loading zone. The estimation should include the total number of buses used at the . school site, the number of drop-off and pick-up cycles run by each bus during its pick up and drop off period, an estima te of the duration buses will be parked in the zone during its longest period (loading or unloading), and the length of the buses to be used at the location. Two exit lanes (one each for right and left turning buses) should be provided at locations where . bus loading and unloading driveways reenter the street system. Sufficient length of storage should be provided from the street exit back to the location where buses are lined up for . loading and unloadirig .. The bus loading and unloading zones should be designed for one way movement, with the passenger door on the building or curbside. It is highly recommended that this be a counterclockwise movement Likewise, the movement of cars in loading and unloading zones should be in a one-way, counterclockwise direction. Cars should be parke>! parallel so that they do not need to back up. This counterclockwise motion guarantees that students eKit the vehicle on the curbside. . . The student loading zones should be comple_tely separated from the bus zones. There should never be a need for parents' cars to mingle with buses. 108

PAGE 120

7.2.6 Summary The counteimeasures presented in this chapter for: each of the evaluated issues identifi ed through an extensive literature review of ppblications oil school transportation "3 Es" (Engineering, 'Education, and :Enforcement) were d_eveloped in the safety program for school-aged children. Although each issue has its specific countermeasure, it is suggested that when facing any of the complicated issues of school transportation safety, a . systematic analysis of the problem .should be performed and all possible countermeasures involving the 3 Es be considered. 7.3 Young Drive rs 1bis section discusses countermeasures applicable for the safety concerns of young drivers. Most countermeasures are education or enforcement related than engineering related for this population subset 7 .3.1 Education and Expanded Programs The curriculum of the driver education programs to be overhauled before reintroducing them to the nation's high schools. In addition to covering basic driving techniques, the goal of the program shou ld be to help reduce risk-taking behavior of drivers and improve safety deci.sion_-making skills. Safety officials have called for the development of teaching and testing software, and the use of interactive simulators. . Parents need to assume more responsibility for training young drivers. National Highway Traffic Safety Administration (NHTSA) senior researcher, Michael Smith, believes that more parental involvemen t is crucial to novice drivers. They gain a Jot more experience by practicing on the road with parents thaii they c:Ould ever get by attending classes only. Parents may or may not "be good drivers, ac!mitS Robert Foss, a highway safety research . ." associate at the University of North Carolina, "but ... they can keep their young drivers focused and serious" (Smith; 1998). Community groups, national organiutions, and businesses are helping to improve communication between parents and teens. 109.

PAGE 121

Focus on the Future, a plan developed by the National Safety Council, seeks to consolidate teen safety efforts of many different organizations. A part of the plan, aimed at 16-24 year old traffic offenders is called DDC: Alive at 25. This National Safety Coun9il-sponsored defensive-driving course does not include many lectures; It is a 4;hour program that encourages young people to get involved in the discussion of various issues. By the use of videos, case studies, and discussions this course helps young drivers realize and understand the consequences oftakibg risks. Focus on the Future .ilso sponsors 'safe-driving' events at schools. Lessons on drinking and driving, safe-attitud,e workshops, and semi-t:(iuler blind. spot demonstrations are some of the activities at these events. Other public symposiums feature sess ions on teen viol ence, traffic safety, impact of the media, and cooperation of parents, teachers, students, and law enforcement in issues of traffic safety. . General Motors Corporation SpoDS?rs Partners in Safety, a program that helps ease : tlie anxiety teenagers may face when <4iving with their parents. Using videos, booklets, and other materials, the program aims to increase the time parents spend in the car with the teens. . The program is designed for use within graduated licensing, and is already in place at ov7r 13,000 schoolS. Many states link high school and other driver-education programs to graduated licensing policies 7 .3.2 Grad uated Licensing Programs Graduated licensing policies are probably the most popular weapon in the battle to prevent . accidents of teen drivers. Th .ese. policies reinforce the notion that driving is a privilege and not a right. Over 20 states, including Florida, have already adopted graduated licensing policies . Several others are seriou51y considering adopting such measures. Nearly every major highway safety orgaoi:retion in the U.S. eodorses graduated licensing For example, American Association (AAA) advocates the instillation of graduated licensing in . aliSO states by the year 2000. The agency makes several suggestions and recommendations for states to consider during the development o f such policies. l]nder graduated licensing programs, drivers must complete a 3-step process before they are granted a full license. AAA guidelines suggest that the first level of licensing allow a 16 110

PAGE 122

year-old to obtain a learners permit after passing a written test At this level, teens may drive during daylight hows only and must have a driver over 21 present in the car. AJso; driver is not permitted to transport any teen:aged passengers. After six monthS of no or the driver is eligible to road test for an intermediate levei license. This license allows the teen to drive at night as long as a driver over 21 is present in the car. The driver is . also allowed to carry a maximum of2 teen passengers. Upon reaching the age of 18, a full license is issued to those drivers who have not had any accidents or traffic convictions during the last 12-month period. Other agencies, such as NHTSA and the National Safety Council, suggest similar rules for teen licensing. ManY graduated licensing policies currently in place resemble those of AAA and other models. Graduated licensing policies typically incoxporate . other countermeasures such as curfews, limitations on number of passengers, zeio-tolerance alcohol rules, mandatory supervised periods, and more parental involvement. . 7.3.3 Drivin g Restrictions: 7.3.3.i Curfew According to the Insurance Institute for Highway Safety, 42% of the teen driving deaths occur between 9 p.m. and 6 am. Most graduated licensing programs, including Florida's program, place nighttime curfews on teen drivers. Usually, the only exception allowed to the curfew is when the novice driver is en route to or fr9m work According to Florida's curfew policy, a 16 year-old driver may not be on the streets between 11 p.m. and 6 am. The restricted hours for a 17 year-Old are I a.m. to S am. Since 'the law took effect in 1996, number of nighttin{e accidents of drivers aged between 15 and 17 has diopped by 23% . Other states such as North Carolina have similar curfew rules. North Carolina broke new ground by setting its curfew at 9 p.m. According to the University of North Carolina. research scientist, Rob Fos.s, the curfew hour Is b!!Sed on the finding that 77% of noVice driver crashes occur between 9 p.m. and midnight., Foss also pointed out that "recr eational driving" usually started by 9 p.m. The novice driver is easily distracted when several other teens are in the car. 1bis situation becomes worse when combined with lack of experience 111

PAGE 123

. and poor nighttime driving skills. For this reason, most graduated licensing programs also include passenger restrictions. 7 .3.3.2 Passengers . The Insurance Institute for Highway Safety pointed out that passengers accounted for about 66% of all teens killed in crashes involving young drivers. Many graduated licensing programs specify the age and number of passengers a teen driver may carry. Depending on the level of the license, the presence of an adult is required at least part of the time. Also, the . level of license limits the number of teen-aged passengers allowed in a car with a teen driver. As part of its guidelines for graduated licensing programs, AAA suggest s that level-one drivers have at least one adult over 21 in the vehicle at all times. This stage forbids the driver from carrying any teen passengers. Once the driver attains the intermediate level, he/she carry a maximum of 2 teen passengers. The presence of an adult is required for nighttime driving. Upon receiving a full license, these restrictions are dropped. Zero Toleranc e Over 37 states have enacted zero-tolerance legislation. Zero-tolerance is a concept designed to prevent persons under 21 from driving even if small amounts of alcohol are comumed. This is based on the fact that it is illegal for anyone under 21 to purchase, possess, or consume alcohol. There is no leniency for under-age drivers found to have alcohol in their system. Most zero-tolerance set a blood alcohol content (BAC) Jitqit at 0.02%. Penalties include fmes, community service, license suspensions, and completion of substance . abuse programs. 7 .3.3.4 Seat Belt Use The rate of seat belt use in the U.S. was about 15% through the early 1980's. Due to the passage of seat belt legislation by 31 states, the rate of seat belt use climbed to 42% by 1987. By 1992, national campaigns to increase awareness and enforcement pushed the rate to 62%, and the current rate lies around 68% As mentioned earlier, seat belt use among young 112

PAGE 124

dri":ers is much l ower than th e national Most efforts to illcrease seat belt use usually illcorporate messages specifically aimed at yoUDg drivers. The Presidential Initiative for Increasing Sear Belt Use Natforrwide, a plan illvolvillg ll!any agencies, seeks t o raise seat belt use rates to 85% by 2000 and to 90% by 2005. Achievillg these goals would help America save thousands of lives and billions of dollars; as well as prevent hUDdreds of thousands of illjuries. Many government officials do not understand the link between seat belt use and social burdens. The Initiative sees this as one of the largest barriers to illcreasillg seat belt use. Legislators must be made aware that the combined costs of illjuries preventable by seat belt use is shared by the government, health care systems, businesses, and illdividuals. The Initiative is a fourpoint p l an that seeks to build public/private partnerships to help save lives, pass tougher laws, make law enforcement more effective and visible, and improve public education about seat belt use. While teen seat belt use is not the sole focus of tjle p lan, parts of it do address this issue. The Department of Education is one of the key agencies participating ill the Presidential Initiative. Its role is to encourage school health and safety officials to illclude proper seat belt use in their health and wellness programs. It is also charged with working ill cooperation with NHTSA to develop a traffic safety curriculum for all grade levels, as well as promoting district seat belt use policies: State and local enforcement agencies have also focused on teen seat belt use. While most seat )?elt Jaws are considered secondary, because a citation can be issued only if the driver is . stopped for another violation, II states have passed primary enfor cement measures. This policy allows police to stop motorists solely for seat belt use violations. Many states have illcluded prinwy enforcement ill teen driver legislation. 7 .3.3.5 Enforcement/Punl sb ment Stricter law enforcement and creation of innovative punishment are some of the methods being used t o reduce teen accidents. There are many supporters for these coUDtenneasures, but arc no t always universally accepted. Promoters of young driver safety support more 113

PAGE 125

frequent traffic stops t o verify licensing status, check curfew violations, ensure seat be l t use, and enforce zero -to l erance limits. They lllge police officers t o enforce th ese tules strictly Unfortunately, many p olice officers are reluctant to make stops and issue c itations because they feel judges are l eliient with young people. In Aprill995, a 16 year-old driver struck and killed a construction worker on Interstate 95 in South Florida The teen, who fell asleep at the wheel, did not realize he had hit the worker until othe r motorists flagged him down over a mile later. Police charged the youth with vehicular homicide and leaving the scene of an accident. The presiding judge withheld a judgement of guilt, and instead suspended the young driver's license for 6 months fined him . $500, and ordered him t o attend driving school. Many other alternate forms of punishment have been used to change the behavior of teen drivers. In Los Angeles and other cities, the court orders some young offenders to t our the morgue. They must also watch a graphic video and relate their exp eriences. Similar measure s i n clud e mandatory tours of emergency rooms and rehabilitation clinics. Another idea is to have teen juri es decide cases involving t een offenders. Case Studies Ontario, Canada began a very specific graduated licensing in 1994. Young drivers . start with a Gl permit after passing written and vision tests. Under this pemlit, a fully licensed driver with at least 4 years of experi e nce is required to be in the front seat at all . times. This person must have a BAC lower than 0.05%, while the drivers must be zero. 7.4 Internationa l T ourists Several ixoprovements are suggested as to the identified critical safety i ssues of internatio nal tourists. Most of the critical issues are however inter-related. H ence, instead of discussing countermeasures separat ely for each issue and conc erns, they are . discussed in general. It was also noted that most o f the problems and hence the countermeasures are reiated to traffic signs. 114

PAGE 126

. 7.4.1 Sign age Improvements Exit Signage on Limited Access Roads: Exit signage on limited access roads needs to be improved in several ways. considering signage on limited access roads, the most . prevalent problem experienced by international towists was the need for better advance warnings of upcoming exits. An international tourist traveling on the interstate system looks for an exit to take him/her to their intended destination. fu such situations, European visitors are familiar with three wamiitgs. Often m Florida there is just one advance guide sign indicating the exit and, hence creates a safety problem. This situation is more serious in lef't exit conditions as international tourists, travelliog slower than the rest of the traffic, in the right-most lane are put at risk. Hence, more advance signs are recommended especially in places with high percentages of tourist travel and presence of left exits. . . Better Destination S{gning: Rather than cardinal directions i.D.temational tourists depend heavily on destination keys. Since tourist attractions and locationS are not typically listed on the signs on the limited access routes, international tourists have problems matching route numbers or name and cardinal directions with their intended destination. This problem is amplified when faced with just one opportunity to assimilate the necessary informatio n on the interchange advance guide signs. Also, particular attention should be paid to providing better destination signage to airports and car rental agencies. Improving Consistency in the use of Route Name/Number: Inconsistencies in the usage of route name and number for the same route has proved problematic for international tourists, . particularly on tolled roads. Using the same number or name in all applications, such as on . . guide signs, maps and also veroally is particularly important in ensuring the safety of international drivers. Use of More Symbols within the Road Signs: This problem was suggested to be easily anticipated with the international signage practice of usirig symools and the U.S. reliance words. The current Manual on Uniform Traffic Control Devices (MUTCD) does use more symbology, in particular with the warning, general information, and recreational and cultural interest signs. International tourists, particularly Europeans are more accustomed to symbols 115

PAGE 127

. on road signs. Hence deployment of additional symbols within the road signs will assist in achieving better understandillg of road signs among international tourists. This step will essentially help those tourists who are not familiar with the English language. Mo;e Use if Diagrammatic. Signs: The international sign system in relation to its dependence on symbology rather than words uses diagramnuitic signs for guidance on all classes ofroads. International tourists are therefore accustomed to navigating by such signs Diagrammatics have the added benefit of showing the general direction and geometry of the upcoming route in relation to the current travel route .and hence are suggested as a col.\Otermeasure in increasing the' safety of international tourists. 7.4.2 Educational Countermeasures The other type of countenneasure that will assist international tourists when traveling on unfamiliar roadways is education.. Even though it is not possible to provide a completeiy fanuliar driving experience for visitors from throughout the world, a coordinated effort has . been recommended by Wilbur Smith Associates to educate the international tourists about. driving laws customs and conditions in Florida. This will require a multi -disciplinary approach in coordination with the other state and local agencies as well as the tourism industry. A professional video should be produced that will i.o.temational visitors to Florida roadway COJ)ditions, traffic contro! procedures (signs, signals, and markings), CUStoms; traffic. laws and saf ety tips. This video may be produced in several languages and could be distributed for viewing to those airlines that provide direct i.o.temational service to Florida. . ,Also, developing and operating official tourist ulformation centers at airports and rest areas in the interstates will be extremely u,-;eful to the tourists. Available technology should be used to install interactive computers that provide navigational services and information about tourist attractions, hotels, and restaurants. 116

PAGE 128

The aforementioned COuntermeasures are Sllllpilarized in a matrix fonn for a ,better understanding as one countenneasure is sometimes capable of addressing seveial identified issues. Thus, a countenneasure matrix for international tourists is given in Table 7.7. 7.4 Pe .ople with Disab!.l.itles The Committee on Specialized Transportation of the Transportation Research Board held a targeted workshop in 1991 designed to identify a framework of key policy issues and research and development needs. The workshop identified a total of 13 key policy issues from fou: policy areas and related research and development needs to address these issues. These is sues and countermeasures are discussed below and swnmarized in Table 7.8. Information, Legislative, and Regulatory_ Issues: ; (1) No standard and scientifically sound method of license recall exists for people with progressive disabilities. We need to be able to identify when progressive disabilities are likely to interfere with safe driving. We also need to identify accurate and equitabl e methods of evaluating the deterioration in people's driving skills, to standardize them, and to make them available to policy makers of driver licensing. (2) Information conceroing private vehicle transportation for people witli disabilities is not . available to consumers, suppliers, government, and others. Research needs to be conducted to determine missing information and the effects of making this infollDation more widely available. Widely available information would allow government agencies, . suppliers of equipment, and other organization s to benefit from the experience of consumers, suppliers, evaluators, and others. {3) The private vehicle modification industry is largely an unregulated m arket. A sound profile of the vehicle modification industry needs to be developed, to document the . . quality and perfollDance of its services and products, and to determine the necessity of regulatory control. 117

PAGE 129

Table 7.7: The Countennea.swe Matrix for the Safety Issues of International Tourists Countenneasure Better Jnfonnation Development Size and Avoid I Infonnation Sessions of new. Signs Location Confusing/ on Signs Brochures etc. Changes of Misleading Signs Text on Signs Issue/Concern I Unfamiliar with x X Roadway System Unfiuniliar with X Driving System Difficulty X X X X Recognizing and Understanding Traffic Signs . Lack oflnfoiiilation X X X about Driving laws and Customs In US Confusion in X X X Translatilig English . into Metric Distance on Signs '" 118

PAGE 130

Table 7.8: Summary ofPolicy Issues and Coimtermeasures . Areas -Issues Countermeasures Information, Legislative and No standard method of recalling Research and develop methods Regulatory licenses from people with for evaluation and license recall : proJUessive disabilities Poor information flow between Research and _develop methods to consumers, suPpliers, facilltatc communication governmeot Lack of regulatioo of safety Research ways to develop standards in the produetio11 of sta11dards adaptive equipmeot Driver Assessment, Education, Insufficient standardizat!oo of Research and develop and Licensing driver assessment and testiag systematized tests and evaluation criteria for people with methods disabilities No standardized training or Researeh and develop a standard course materials for instructors syllabus and training system who conduct driver assessments for people with disabilities Driver evaluation and assessment Research and develop a plan to techDology has not kept pace with bring these techDologies into ; bio-medicalmd vehicle balance technology Vehicle Design, Adaptive Insufficient standardization of Develop standards and Equipment Standards and the adaptive equipment and specifications Driving Environment specifications for v e hicle modification Lack of disability related design Research meth o ds to mal
PAGE 131

(4) Driver Assessment, Education, and Licensing: (1) Standardization of methods and criteria is insufficient for testing the driving ability o f . . people with disabilities. An appropriate basis for standard methods and criteria needs to be determined, both in terms of techniques and the ability. of these functional areas measured by the techniques to reveal an individual's ability to drive safely. (2) Standard syllabi and training materials for instructors, who assess drivers with disabilities, needs to be developed. These materials would take into account standard methods and criteria for testing the driving ability of people with disabilities . (3) In order to narrow the gap between driver evaluation and assessment of bio-mediclil. and vehicle technology, useful medical and vehicle advances needs to be identified: 'to evaluate the potential in improving our driver assessment methods. Vehicle Design, Adaptive Equipment Standards, and the Driving Environment: . (I) Few standards exist for the equipment to be used in vehicles driven by people with disabilities. As equipment for private vehicle adaptation is often provided by a competitive tendering process, bidders have an incentive to reduce the cost of equipment to win the bid. Existing standards would ensure that the equipment meets the specified safety requirements. Further, accident history data would be very useful to the agencies that establish standards. (2) Potential size of the market needs to be established to encourage vehicle manufacturers to consider the needs of people with disabilities. It is also necessary to defme features that could be incorporated into new vehicles that would benefit people with disabilities either directly or indirectly facilitating later vehicle modification. (3) Crash dummies, representing able-bodied people, are used to test the safety of new vehicles. The exten t of deficiency in the final products offered t o people with disabilities 120

PAGE 132

needs to be detennined. If tjle aaSh dninmies cause deficiencies, the parameters of test devices that would provid e more appropriate testing of products needs to "be defined . (4) There is a need to a national emergency aid network for drivers using adapted private vehi cles. This service .would assist in rescuing stranded individual s and.'render ' those services to the specialized vehicles that the normal automobile service networ k would not be able to provide. 7.6 New I mmigrants The three basic types of countermeasures to address the unique highway safety issues of new ' immigrants are improved data collection, education, and roadway sign changes. The standard forms used to col)ect highway crash data do not identify whether persons involved are new immigrants. Extensive data is needed to examine the seriousness of safety associated with new immigrants. In areas where a significant portion of the population are ne w immigrants, changes in the .language of roadway signs may be a while such changes are inappr opriate elsewhere. The sole study concerning highway safety needs of Hispanic immigrant communities identifies education as a primacy countermeasure (NHTSA, 1991). The following discussion is based on the strategies recommended in that study. Focus group participants and telephone respondents of government organiza tions e x pressed support for promotion al programs that are highly personalized, family-orient ed, culturally . sensitive and relevant, and non-confrontational. Mos t often, focus group participants ' mentioned personal cont act and relation.Ships as important strategies. They pointed out that graphic and expli cit depictions of motor vehicle crashes are an effective method of conveying the message to families. Also emphaSized was the importance of developin_g and maintaining cultural sensitivity t o the di verse groups of new inimigrants. Promotional efforts can be carried out in various ways, including community-based ozganizations, churches, schools, families, and the tnedia that target immigrants. .. 121

PAGE 133

Organizations: .. Participants mentioned seve!lll features of community-based organizations that are good for safety promotion. Community-based organizations often provide services t o low or modetate income househqlds at little or no cost, have staff who speak the language of the and are culturally sensitive. Based on the opinions of telephone and focus group participants, it was found that outreach activities SJionsored by community-based organizations are successful. For example, the Century Council in Los Angeles has organized traffic safety campaigns around traditional and popular Hispanic festivities. The direct outreach strategies included setting up boothS with traffic safety information as well as distributing pledge cards that individuals sign promising to stop unsafe behavior. Churches: Participants of all ages are receptive to traffic safety programs through churches because they are trusted and accessible. Highway safety organizations in Florida and Texas have engaged in partnerships with the church. The Community Alliance for Training and Safety in San Antonio, Texas has enlisted priests to include safety themes in masses. In many areas, church-related organizations are one of the principal agencies that provide service to recent immigrants. Schools: School-based programs are oftm effective in promoting highway traffic safety becai)SC both parents jllld children respect teachers and the educational system. Some of the examples from the focus group participants are listed below: (I) According to participants in McAllen, Texas schoolteachers do traffic safety education by taking the children to see the court system and the jails, and teachers are assisted in this effort by the police department. (2) Women participants in the Rio Gtando Valley, Texas, pointed out that community schools are good for educating adults in traffic Currently, the only places that 122

PAGE 134

', offers driving classes are specialty-driving schools, at a price of $200. All the participants agreed the price was beyond their means and that they would be willing to pay a lower price for the classes. Also, they expressed an interest in taking driving lessons t o improve driving skills and l earn traffic laws. Such c lasses would be a good fo.ium for them to leam from each other by sharing their experlences behind the wlieel. (3) Adolescent male participants in the District of recommended that the driver education programs be made available at schools. Traffic safety professionals need to be recruited to teach young people and others in the COl!llllunity. Driver education courses should be offered to those who are interested in learning traffic safety. One of the method s used to ensure active partic i pation is presentations by people who have had crashes in the past. Family: The family is one of the most powerful symbols in the immigrant commwtities Anything that is viewed as a danger to the family is more effective in reaching the entire commwtity. Participants emphasized that any public awareness campaign for these communities must feature the family. For example, grandmothers have been u5ed in child safety messages by some organizations because they are often the true family educators. Instead of presenting statistics alone, the participants suggested that the data be presented with an emphasis on its unplications on the family unit. Media: . In areas with a high concentration of immigrants, there is often a variety of print and elecfronic media that target the iJllmigrants. Study participants agreed that the creative use of both English media and the media of the immigrants' language is essential to any effort to promote traffic safety. Study participants vieWed television as the best medium for disseminating traffic safety Information to immigrant populations, especially in urban areas. The Century Council in Los Angeles has used television extensively in its traffic safety campaigns. Channel 32, a . 123 .

PAGE 135

Spanish Network in Los Angeles, broadcasts a tt.U'Iic safety segment on a morning talk s how at least once every three months. In a segment on io.fant safety seats, they explained the consequences of not having childr en in safety seats' whil e driving. Also, they showed correct ways to buckle children in safety seats and offered io.formation on obtaining child safety . seats: In addition, participants viewed Rescue 911, COPS, and other reality-based shows as ideal teaching mechanisms. Television is often seen as the medium with the most potential for pub lic service announcements on traffic safety. While radio may not be as eff ective as t elevision in promoting traffic safety, it is however more widespread in terms of the number and ways people can access io.formation. The El Guardian program in San J ose, California uses radio on a weekly basis to discuss an array of traffic safety i ssues. In addition to te le vision and radio, newspapers also play a role in promoting traffic safety for immij!l'ant communities. El Observador in San Jo .se publishes a . monthly supplement providin g information on traffic safety issues, in addition to health and other community issues. Study participants also indicated that the government should place safety announcements and stories in dailies. 124

PAGE 136

CHAPTER 8: CONCLUSIONS AND RECOMMENDATIONS 8.1SUMMARY Selection. of special population groups and their corresponl!ing issues/concerns were achieved from results of a statewide survey of transportation safety . Obtaining a high response rate for both surveys was a difficult task, but one that was accomplished through persistent follow-up. Highway safety experts ranked (in order of highway safety importance) older drivers, school"aged children, young drivers; disabled people, international . tourists, and new immigrants as the special population groups requiring further attention. Each special population group was studied and their critical issues and concerns identified through a multi criteria decision making approach. The factors taken into consideration were: 1) impact of the issue on crash rate; 2). effectiveness of roadway design changes on addressing the issue; 3) effectiveness of policy changes on addressing the issue; 4) cost. of .. implementing roadway design or policy change; 5) ease of implementation; and 6) priority to address the issue. Three different approaches were used to assign weights to each factor. The final ranking of the issues and concerns was performed on the basis of considering an indel( of weighted averages. According to the findings, the most critical issue/concern for older drivers was identified to be 'location and size of traffic signs and lettering'. Nighttime visibility and perception reaction times were the second and third most important issues, respectively. These were followed by gap acceptance capabilities, deficiencies in driving knowledge, narrow lanes, driving in congestion, maneuvering curves, and freeway driving in that order. Similar rankings were obtained for other special population groups as well. .. Detailed investigation was conducted on the identified critical highway safety issues of the special population groups. Most of the analysis was based on past studies, using relationships between speed, injury severity, and traffic volume. Injury severity for school aged children was established by collectilig data O!l pedestrian/bicycle crashe:s. Countermeasures were then provided and discussed in detail for the critical issues/concerns identified J;rom each special population group. Since some of the countermeasures are capable of addressing more than one issue, the finiiings were summarized in matrix form whenever if was possible. 125

PAGE 137

Projection models were developed in this stUdy, so thiu crash rates for S
PAGE 138

8.3 RECOMMENDATIONS For each identified issue/concern, countermeasures were suggested in this study. However, it is necessary to identify the most effective countermeasure in addressin g each of the issues/concerns. It would be also useful if knowledge of the benefitS of a specific safety countermeasure could be obtained prior to implementation. Thus, the second phase of this project evaluates some of the suggested countermeasures. A performance parameter is selected for each individual issue by field measurement, under two different conditions ('with' and 'without' the countermeasure), allowing for a final evaluation. This evaluation assists the decision-maker (highway safety engineer) in determining the most effective approach. In addition, the effects of certain countermeasures are evaluated using : the Florida Traffic Crash Database. This is achieved by analyzing crash statistics 'with' and the countermeasure when a).1 other conditions are expected to remain the same. :. . The database used in the model building for school-aged children was found to be unsatisfactory. In order to obtain better models of the relationship between vehicular speed, . traffic volume, and crash severity of school-aged children pedestrian/bicyclist, it is essential that the database is expanded and the necessary information obtained. In the future, greater attention should be paid by relevant authorities to modify crash reports so that more detail ed information could be gathered. However, such changes should only be . carried out under careful consideration and detailed investigation. The projection models developed in this study were based on the compilation of statistical data obtained from various publications If current crash databases were searched for the same figures, more accurate values would have been achieved. The authors recommend repeatin g the modeling process with values from an' actual crash database to determine whether the accuracy of the. models could be improved. 127

PAGE 139

REFERENCES Baits, M.R., Measuring Motorist Comprehension of Florida's School Bus Stop Law and School .Bu.s Signalization Devices; TRB Preprint, Transportation Research Board, Washington D.C., January 1998 Benekohal, R. F., Michaels, R.M., Resende, P. T.V., and Shim, E., Highway Design and Traffic Operations Needs of Older Drivers, TRB Preprint, 940722, Transportation Research Board, Washington, D. C., January 1994 Benekohal, R. F., Michaels, R.M., Shim, E and Resende, P. T. V., Effects of Aging on Older Drivers' Travel Characteristics, TRB Preprint, 940722, Transportation Research Board, Washington, D. C., Januazy 1994 Better Living for Seniors, Florida Department of Health and Rehabilitative Services, 1989 I., Predisposing factors in the Alcohol and Drug-Impairment of Young Drivers;:. . Performance. Yormg Drivers Impaired by Alcohol and Drugs, 1986 Cerrelli, E C., Crash Data and Rates for Age..Sex Groups of Drivers, I 996, Research . Note, National Highway Traffic Safety Administration, U.S. Department of Transportation, Washington D.C., January 1998 Chu, X., The Effects of Age on the Drivi11g Habits of Elderly: Evidence from the 1990 NPTS, Center for Urban Transportation R;esearch, College of Engineering, University of South Florida, October 1994 Dickinson F. 0., Special Report on Driver Age Differences in. Traffic Crashes, Florida Department of Highway .Safety and Motor Vehicles, March 1995 Estimating the Number of Vehicles AdC1]1ted for Use by Persons with_ Disabilities. NHTSA, U.S. Department ofTransportation, 1997 Estimating the Number of Vehicles Adapted for use by Persons with. Disabilities, . Research Note, National Highway Traffic Safety Administration, U.S. Department of Transportation, Washington D.C., December 1997 . Evaluation of International Signing Practices, Final Florida Department of Transportation and Wilbur Smith Associates, Volume 1 and 2, 1994 128

PAGE 140

Evaluation of International Signing Volume 1. Final Report. Wilbur Smith . Associates, and Florida Department of. Transportation, Office of Traffic Operations, 1994 Evans, L, Future Predictions and Traffic Safety Research, Transportation Quarterly, Vo!ume47, No.1, January 1993, pp. 3-18 . Fambro, D. B., Kappa, R. J Picha, D. L., and Fitzpatrick K., Driver Perceptlon-]Jriike Response in Stopping Sight Distance TRB Preprint No. 981410, Transportation Research Board, Washington D.C., 1998 Florida Pedestrian Safety Plan, Florida Department of Transportation, 1997 Florida School Bus Occupant Safety, Center for Urban Transportation Research, University of' South Florida, 993 Florida Statistical Abstract 1997, Bureau of Economic and Business Research, College . of Business Administration; University of Florida, 1997 Florida Traffic Crash Facts, Years 1985-1996, Florida Department of Highway Safety and Motor Vehicles, Tai.Jahassee, Florida Florida Visitor Study 1996, Florida Tourism Industry Marketing Cozporation, 1997 Graham, J. R., Fazal, A., and King, L. E., Highway Sign Minimum Luminance Requirements for Older Drivers, TRB Preprint 970079, Transportation Research Board, Washington D C., January 1997 Haight, F.A. A Method for Comparing and Forecasting Annual Traffic Death Totals, Transportation Research Record 1238, TRB, Natioruu Research Council, . Washington, D.C., 1989!pp.3 1 -36 Hakim, S. D., Shefer, A.S., Hakkert and I. Hocherman, A Critical Review of Macro . Models for Road Accidents, Accident Analysis and Prevention. Vol.23, No. 5, 1991, pp 379-400. Hall, H W., Highway Engineering Improvements to Accommodate Older Drivers, ITE Compendium of Technical Papers, Institute of Transportation Engineers, 1990, pp. 283-286 Hawkins, H. G., Picha, D. L., Mann, B. L., Mcilroy, C. R., Womack, K. N., and Dudek, . C. L., Assessment of Mexican Driver Understanding of Existing Traffic Control 129 ->

PAGE 141

. . .. Devices used tn Texas, Reseatcll Rejl<)rt No. 1274-1, Texas Traospo"l!tion Institute, The Texas A & M University System, November 1996 Highway Safety Needs of U.S. Hispanic Com11flmftles: Issues and Strategies. NHTSA, U.S. Department ofTransportation, Washington, D.C.: 1995 Highway Statistics Summary to 1995, Federal Highway Administration, U.S. Department of Transportation, Washington, D.C. . /fome-to-School Transportation Study, Center for Urban Transportation Research, University of South Florida, 1993 Khavanin, M. R, and Schwab, R. N., Traffic Sign Legibility and Conspicuity for Older Drivers. ITE Compendium of Technical Papers, Institute of Transportation Engineers, 1991, pp. 11-14 Lerner, N. D., Age and Driver Perception-Reaction Time for Sight Distance Design requirements, ITE Compendium of Technical Papers, Institute of Transportation Engineers, 1995, pp. 624-628 Lerner, N. D.; Older Driver Perception-Reaction Time and Sight Distance Design Criteria, ITE Compendium of Teclmical Papers, Institute of Transportation Engineers, 1991, pp. 6-10 Lu, J. J., and Dai, j., The Impact of Driving Behavior of Older on Intersection Capacity, Research Report by Department of Civil and Environmental Engineeri.Qg, University of South Florida, July 1997 Manual on Uniform Traffic Control Devices, Federal Highway Administration, 1988 Marin, Gerardo and Barbara VanOss Marin (1991), Research with H'IS]Janic Populations. Vol 23, Applied Social Research Methods Series, Newbury Park, CA: Sage Publications. McCoy, P. T., Tarawneh, M. S., Bishu, R. R., Ashman R. D., and Foster, B. G., Evaluation of Countermeasures for Improving the Driving Performance of Older Drivers, TRB Prep rint No. 930290, T ransportation Research Board, Washingto11 D.C., Januazy 1993 National Transportation Statistics I997, Bureau of Transportation Statistics, US Department of Transportation, Washington D.C., 1997 130 ..

PAGE 142

Nationwide Personal Transportation SUriley: 1990 Databoolr, Volume 1, U.S. Department ofTran.sportation, Federal Highway Arlminimation, pp. 3-18 Nayor, D. W., and Graham, J. R., Intersection Design Decision! Reaction Time for Older Drivers, TRB Prepdnt No. 970181, Transportation Research Board, Washington D.C., January 199 Owens, A. D., and Andre, J. T., Selective Visual Degradation and the Older Drive r IATSS Research, Vol. 20, No. 1, 1990 Partyka; S. C., Simple Models of Fatality Tre!!ds using Employment and Population Data, Accident Analysis and Prevention, Volume 16, No. 3, 1984, pp. 211-222 Partyka, S.C. Simple Models of Fatality Trends Revisited Seven Years Later. Accident Analysis and Prevention, Volume 23, No. pp. 423-430 Private Vehicle Access for People with Disabilities; Policy Issues, Options, and Research. Transportation Research Circular 415, National Research Council; Transportation Research Board (1993). Safety Isspesfor Vehicle Adapted for Use by Persons wlth Disabllities. Research Note, NHTSA, U.S. Department ofTransportation, 1998 Sivak, M.A. 1975 Forecast of the 1985 Traffic Safety Situation: What Did We Learn From an Inaccurate Forecast. Road Users and Traffic Safety, 1987, pp. 13-25 Sivak, M. Society's Aggression Level as a Predictor of Fatality Rate. Journal of Safety Research, Volume 14, 1983, pp. 93-99 Smeed, R. J., Some Statistical Aspects of Road Safety Research, Journal of the Royal Statistical Society, A, Volume 112, 1949, pp. 1-23 Statistical Abstrac.t of the United States, The National Data Book, Economics and Statistics Bureau of Census, U.S. Department of Commerce, 1997 Tan, T., Increasing Highway Sign Legibility for Older Drivers, TRB Preprint No. 920711, Transportation Research Board, Washington January 1992 Transportation in an Agif!g Society: Improving Mobility and Safety of Older Persons, Transportation Research Board, National Research Council, 1RB Special Report 218, Volume 1 and 2, 1988, Washington D.C. Transportation Statistics Annual Report 1997, Bureau of Transportation Statistics, U.S. Department of Transportation, 131 .

PAGE 143

Wheefchair Injuries and Deaths Associated with Motor Vehicle &lated Incldmts, Research Note; National Highway Traffic Safety Administration, u : s. . Departnient of Transportation, Washington D.C., September 1997 Yi, P., Gop Acceptance for Elderly on Rural Highways, ITE Compendium of Technical Papers, Institute of Transportation Engineers, 1996, pp. 299-303 132

PAGE 144

Appendix 1 133

PAGE 145

Special Population Safety Survey Survey Forni A (One page) SiX special population groups have been tdenUfled.Other population groups may be added at the bottom. Please Indicate the imparlance of highway safely concerns for each population by only marking one box for each group. Less. lmporlant No 1 1 Older Drivers Opinion D 0 2 lntemadonat Tourists D D 3 School Aged Chnclren D D 4 Young Drivers D D 5 .New Immigrants D D 6 Disabled People D D 7 Others : (please specify) D D P:' 1 2 0 D D D D D D 3 D D o D D D D . More lmporlant 4 5 D D D D D D D D o D D o o D

PAGE 146

. Survey Fonn B ( 6 pages ) For each of !he six special populaifon groups, please Indicate the rating orrmportance for Identified Issues regarding the highway safety. (Mark only one rating for each Issue) Less More Important Important 1. OLDER DRIVERS No 1 2 3 4 5 1 Night time VlslbDlUes Opinion D D D D D D 2 Difficulties In-uildeiSlandlng the D D D D D D Changeable highway messages 3 Deficiencies In OrMng Know1edge D D D D D D '4 Dimensions of Signs ond Lettering D D D D D D 5 Location of the Traffic Signs Cl D D D D D s Perception Reaction TJmeD D D D D D D D D D D D .,.. start-up l osttlme ancfsabirauon Headway at Slgnallzed Intersections sGap acceptance capabmties D D D D D D at stop controlled Intersections 9 Other Important Issues/concerns: D D D D D D (Please specify them) D D D D D D If there Is an obstruction to the movement of the vehicle, the Ume taken by, the driver to see i t and take necessary action is known as perception reaction time . -At a signalized Intersection, time lost In lnlllating lhe movement of vehicles when the signal turns from red to green is the start up lost time, -Minimum possible headWay between two SllCCe$Sive vehicles Is the saturation headWay. At a stop controlled intersection, mlhor street drivers will have to walt for a suitable gap between successive vehicles traveling on the major street. The gap size acceptable for each . Individual explains the Gap Acceptance CapabRitles P: 2 .,

PAGE 147

2. INTERNATIONAL TOURISTS 1 Unfamillarlty with Roadways 2 UnfamiUarity with the driving system (e.g Right hand side/left hand side) s Recognition of Traffic Signs 4 confUsion of traffic Guide signs 5 Other lmpoitant Issues/concerns: (Please "l'4lCify them) Less More Important Important No 1 2 3 4 5 Opinion D D D D D D D D D D D D D D D D o D -o D 0 D D D D D D D D D D D D D D D Although the traffic signs are uniform In US, international tourists may no! be able to correctly interpret or promptly response to some traffic signs. Typical problems Include understanding of words and symbols abbreviations and connotations of words, order of Information cardinal directions etc. Traffic guide signs In large cities may confuse International tourists. Becaus e of deficiencies in eXlstlng guide signs, international tourists may become lost or confused while driving In . major citles. Pt '3

PAGE 148

Less Mete Important lmpo'rtant 3. SCHOOL AGED CHILDREN No 1 2 3 4 5 1 High traffic volumes aroun d school zones Opinion D D D D D D 2 High vehicular speed s in and around school zones D D D D D D 3 Not enougll school crossing guards D D D D D D 4 Lad< of tralftc and bicycle safety D D D D D D education programs 5 Lack of "Trafllc Calming devices such as D D D D o o . speed humps, raised pedestrian medians etc . 6 Lack of pick up and drop off zones side 0 D D D D D walks, traffic Oghts etc. 7 Lack of proper slgnage, signals D o D D D D (walk don't wall<) 8 other Important lssues/eoneems: D D D D D D (Please specify them) D D D D D D Pr 4

PAGE 149

Less M or e 4. YOUN G D RIVERS I mport a n t I m porta nt N o 1 2 3 4 5 1 General lack of driving experience Opinion D D D D D D 2 General reckl essness -----D D D D D D 3 Lack of driVer train ing program s 0 0 0 0 0 0 4 Risk taklrifb ellaVior/agieSSiveness D D D D D o 5 Impaired driVI ng (Aiqohol & Drugs) -D D D D o D 6 Speeding ---D D D D D D 7 teen crui sing D D D D D D 8 _lack o f safety belt use D D D D D o 9 An y o t her Iss u es/concerns: D D D D D D (Please specify them) D D D o D D p; 5

PAGE 150

S. NEW IMMIGRANTS 1 U rifamlnarily with the driving system {eg.Left/Righl handslde driving 2' Differences In the transportation system 3'' language b an1ers ----4 Cultural differences 5 Inadequate/improper driving education {eg.learing through another Immigrant) 6 Other lssueslconcems: (Please spedff them) '< Less M ore Important Important No 1 2 3 4 5 Opinion D o D 0 0 0 D 0 0 0 0 0 D 0 0 0 D D D D D D D D D D D D D D 0 D o D D D 0 D D D D D . New immigrants may be used to roadways with low lralfte speed. and higher vehicle mixes. Higher ve h icular speeds might bring a fear to new Immigrants a n d hence ca n reduce the self confidence. Non EngDsh speaking Immigrants face language ban1ers In an aspects of tflelr daily life including their interaction with the transportation system. This may cause them to take longer Ume to recognize t raffic signs or warnings or even fall to understand them. These barriers may also preve. nt public campaigns on safety Issues from reaching them .., New immigrants usually come from very different cultures. Some cultures may not place high priority on traffic safety Issues as In America ': P: ;

PAGE 151

6. P E OPLEWinf DISABilmES 1 Unsafe access to transportation faciUUes 2 Lack of policies for recalling driving licenses from people with progressive dtsabiliUes 3 Lack of standards f or vehicle mocliflcatlon that are need to be used for d isabled 4 Defau lt designs In roadways and lntersecUons 5 Lack o f crossing d evices for blind 6 other lssu e s/c o ncems: -(Please spe cify them) Less More Importan t Important No 1 2 3 4 s Opinion D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D 0 0 0 D D D D D There is n o standard or scientifically sound (and fair) metho d f or driving license recall for people with progressive disabilities. 7

PAGE 152

Appendix2 141

PAGE 153

.. ..-.u. v Jj .J. v .J. .a. .J. Vl.l. .1. J.v v.nJ. .c. .1. J. J. 'IL..C.LI.:> v J..' u.t" .C.vUi.L r V J:" U.l...h 11 U \..:JKV U.t::) 11'1 r LV l'U.UA he purpose of this survey is to conect infonnation on highway safety needs of special population groups in Florida. This information is itical in helping to shape new safety design standards, engineering practices, and safety policy. A survey conducted by the Center for rban Transportation Research (CUTR) and the Department of Civil and Environmental Engineering in July 1997 Identified the following Jecial population groups and corresponding issues specific to each group. W e need your assistance In rating the importance of each I ss u e lecific to the special population group. Please take the time to complete the survey. You can fax your completed survey .to Dr. J. John Lu (813) 974-5168 or mail it to CUTR, 4202 E. Fowler Ave., CUT 100, Tampa, Florida 33620. Thank you for your assistance! 5. itbeiouso crutlyWould iibe 10 I easy wc<.ld it be 10 on ntc/ I ro.'1<1Wl!)' """tn el11111&es 10 I policy chnges 10 dd= impltment desir,n or implement dtc dtiii1CC1 (Cbcck one) odd= each ohhesc .. ch ofthesc issues? policy chn&c to oddtoss (Chcclc one) ues/concc:ms the...rcty Impo= T......,ct effedive dfectiVc cosdy cosdy llMY cffcc:tivc: dfcctiVc . %So time'Oisibilily 01 CJl CP o os o CJl 0' o o o CJ1 CP o o o CJl 0' o o o Cl'CP o CJS a CJ1 a o CP 0 1 O t 01 o a a ot Cl' a a a CJ1 a> a as a CJ1 a> a a a CJ1 0' a a a o CP a a -"""">' dtiving a at a> a CJS a l a al o os at a 0' o os at 0' al o CJS at a 01 a as at a CP o as muvetingcurves. at Ot CP a o o CJ1 01 a as at o CP a as at a CJS a a s at a Ol o a o a a' a CP 6d .. at QtCP o as at a 01 o a at 0' a> a o s at CJ1 a> o as at CJ1 a> a as at CJ1 CP a as .. lling lalowl o a s I o CJt os o 1 a CJ1 CP a as 1 o a a> a a s 1 a 0, a> a as I 0, 0, 0, 0, O I ocoeptmce o 01 0, a os a cz CP' c c 1 o oz 01 c o 1 c c 01 c CP 1 a c 0 c as 1 a 02 0 a o rging into lr.l!lic (<12. I a at CP a CP I c CJ1 cP o CJS 1 a CJt a' o a 1 o a CP a a 1 c CJ1 CJl a CJl 1 a CJ1 CP a 0 x : there is 2!10bsttuclion to of the vehicle, the time tJken bylhe dtiverto ste it :md t:lkt necess:uyoaion is known os perception/re:tttion time. . . . (

PAGE 154

on aashl occident r:tte? lsmes/ con= j{Chcclc one) lddn:sseach of or these issues? I polkychonge to :iddt..s I (cbcclc one) oflecting dtc s:tf e< c> o CJ2 c> o c> o CJ2 c> o os 1 o o IJl o o 1 o c> o o os 1 o CJ2 c> o o """""'.a.ooiZOil o o 1 o o o o o 1 o O' o o o 1 o c: o o o 101 0' o o o. l a c: 01 a os oroundschool20nes l..ockodtopoffond 1 Cl' o o o os 1 at 0' o o as 1 0' a o o o 1 Cl' c: o o o 1 a o o o as 1 Cl' CJ2 o a a pldt up :tOilCS Lode of ttoflic 1 a c: c> a o 1 o a> o o o 1 o a a _o o 1 a a> a o o 1 o o> CJI o os 1 o o> os o os c:olmiogde9i= Lode of I""P"" 1 a CJ2 CJI a os 1 a. a> fJ' a a> I a. a> CJI a os I a CJ2 c> o o I a c: CJI o os I a CJ2 os o as sigao&e and ttoflic signoh Lock of sidew:all a a I at o a o a I a a' a a a I o CJ2 a a a I o a CJ2 a as 1 at o CJ2 a a one! bike routes Lode ofttoflicond. I a c: CJ2 a a I at 0' a o as I o a> a a a I a CJ2 a a as I at a CJ2 a a 1 c c: a a os bieydesoky pn>fpiDS I o CJ2 C. o os 1 o a> .c:P a e< 1 a a> CJ2 a CJ2 1 o CJ2 a a as 1 a t CJ2 a a as 1 a c: .a a os r..d CJ2 a os I a a> a> a a I o a a a a 1 a a CJ2 o os 1 o c CJ2 o o 1 0 o CJ2 o a

PAGE 155

. . In )'OUt opinion, l>ow 5. In your Oj)l1bOGI,,... r ... 1"'" ........ .... ----11m: I it be to use I eCc.c.M: -..add it be to use' CX>Stly-..add it be 10 easy would it be to much oh pOority is it to oddtcss irnplemaJt or clunge? eoch issue/conczm? ssucs/ conoems . . one) (CIJeclc one) .ll'cdingthesofety gdriv= driving I 01 CP 0' 0' 0> r CP OJ o 0> r OJ CJ.1 o 0> r OJ 0> o QS r !Jl CP o 0> I !JI 0' 0> o o o oz 01 o os o o o o o o oz 01 o o o 01 o o o o o 01 o as 1 o oz 01 o !='' I o 01 o o os I o o 01 o a> I o oz 0' o o j o o 0' o !JS I o 01 0' o 0' I o 01 01. o a mpWd clriYing OJCP0'0'0>010'0>Cl'O>OCPO>OO'O>OCPO'O I o 01 01 a osl a oz os Cl' as I a CP oi o as :..X of sofcty 'tStl2intuse of driver 1 a 01 01 o o 1 a (Jl (Jl a a 1 a a D' o as 1 o a a) a as I a a a' a a 1 a 0' o> o as nining p:ogr:um lrnm>.turity Ia CP a a o 1 o CJ2 !Jl a o 1 o CP o a a 1 o a a a o 1 o o a o as 1 o a o a o (Cbecltone) to one) (Cbcdconc) with e:uy =to or CP 01 o o 1 or CP 01 o as 1 o CP o> o o I a 01 01 o o 1 o 01 01 o os 1 or 01 o a a !JI !Jl (Jl o D'ID' !J2 01 !J< (JS,!JI (Jl (JI a < OliO' CP !JI 0' a 1 a CP o o o o oz (Jl !JI CJS I Cor !JI CP (Jl as ID' 0' !Jl o !JI I !JI modilic:lliom a o o a 1 o a a a. os 1 o a o> a as 1 o (Jl o> o o Hor dis::d!l
PAGE 156

($$Ues/COS>ccms trn&miliotwilh tt:Uliclow$ ond Ununi!iorwith on 'cmsbloccidenc rn1e? (Checlt one) Ia. CP 01 a CP Ia 01 a a a I a'With lOOd I a a c a cc 02 a cla ;ysrcm C.:.ckof cbMt 1 a CP a> a a> 1 a CP CP a CP I []I :ducotion s.!ecy !duCllion I a az a a o l a oz C!' o o lo a Ol a a 1 a c a c a 1 c 01 a> a a 1 a CP a a a s 01 c1 a c1 Ia 02 o o 01 1 c c2 c o o l a a e> c o CP CP c CJl I a 01 CP a a 1 c 01 01 o CPI []I CJI or c 0' o z a o ola 0 1 o o os IO 01 01 o a> I o o o o a bmi... I a oz c> a c> I d 01 a> a os 1 o o a> a ol a oz a a o 1 a o 01 o o o oz os a ::ll!nddill'= a a> os CJt o a a1 c> a a a a1 c> o a a a 01 a as os c:p 01 o a1 o oz o 0< a