5
Application of Risk Estimates: Illustrative Scenarios

In this chapter, three simple scenarios are used to illustrate how the results presented in Chapter 3 can be used in a quantitative risk assessment to analyze alternative policies and procedures regarding the transportation of children to and from school. The committee wishes to reiterate that the results of the risk analyses are based on national statistics and that care must be exercised in using those results to draw specific conclusions about local problems. It is important to keep in mind that the results of similar scenarios for any particular school district could be different from those presented here. However, the general insights gained from these types of analyses are applicable to many school districts and can serve as a starting point for discussion and for the application of the safety checklists presented in the previous chapter.

The three scenarios presented here illustrate the application of all four risk measures presented in Chapter 3: injuries and fatalities per 100 million student-miles and per 100 million student trips. The first scenario examines the effects of changing the minimum school bus pickup distances for a suburban elementary school and employs the per student-mile risk measures. The second scenario shows the benefits of adding a later after-school bus service to the pupil transportation services provided for a suburban middle school. The final scenario explores the impacts of doubling the size of a suburban/urban high school parking lot. These last two scenarios use the per student trip risk measures.

Though the risk estimates are based on national statistics, the three scenarios are set in well-defined but hypothetical school districts. In every case, realistic values for the schools were selected. The transportation mode distributions under different busing conditions were based on discussions with several school districts. It must be emphasized that an actual school would have to estimate these types of changes in travel mode usage through expert opinion or a survey sent to parents.

SCENARIOS

Scenario 1: Changing the School Bus Pickup Policy for a Suburban Elementary School

To examine how different school bus pickup policies would affect the risk to students, a hypothetical elementary school was created in a rural/suburban area. There are 250 students attending school. All students live within 10 miles of the school, although most live closer (20 percent within 1 mile, 50 percent within 3 miles). The students currently travel to school using a variety of modes. Table 5-1 shows the distribution of mode choices for students who live various distances from the school based on two minimum school bus pickup



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment 5 Application of Risk Estimates: Illustrative Scenarios In this chapter, three simple scenarios are used to illustrate how the results presented in Chapter 3 can be used in a quantitative risk assessment to analyze alternative policies and procedures regarding the transportation of children to and from school. The committee wishes to reiterate that the results of the risk analyses are based on national statistics and that care must be exercised in using those results to draw specific conclusions about local problems. It is important to keep in mind that the results of similar scenarios for any particular school district could be different from those presented here. However, the general insights gained from these types of analyses are applicable to many school districts and can serve as a starting point for discussion and for the application of the safety checklists presented in the previous chapter. The three scenarios presented here illustrate the application of all four risk measures presented in Chapter 3: injuries and fatalities per 100 million student-miles and per 100 million student trips. The first scenario examines the effects of changing the minimum school bus pickup distances for a suburban elementary school and employs the per student-mile risk measures. The second scenario shows the benefits of adding a later after-school bus service to the pupil transportation services provided for a suburban middle school. The final scenario explores the impacts of doubling the size of a suburban/urban high school parking lot. These last two scenarios use the per student trip risk measures. Though the risk estimates are based on national statistics, the three scenarios are set in well-defined but hypothetical school districts. In every case, realistic values for the schools were selected. The transportation mode distributions under different busing conditions were based on discussions with several school districts. It must be emphasized that an actual school would have to estimate these types of changes in travel mode usage through expert opinion or a survey sent to parents. SCENARIOS Scenario 1: Changing the School Bus Pickup Policy for a Suburban Elementary School To examine how different school bus pickup policies would affect the risk to students, a hypothetical elementary school was created in a rural/suburban area. There are 250 students attending school. All students live within 10 miles of the school, although most live closer (20 percent within 1 mile, 50 percent within 3 miles). The students currently travel to school using a variety of modes. Table 5-1 shows the distribution of mode choices for students who live various distances from the school based on two minimum school bus pickup

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment TABLE 5-1 Scenario 1: Distribution of Travel Modes (percent) for 1-Mile and 2-Mile Minimum School Pickup Distances and No School Bus Service Miles from School Walk Bike School Bus Adult Driver Teen Driver Other Bus (a) 1-Mile Pickup Distance Less than 1 60 30 – 10 – – 1–1.5 30 20 35 15 – – 1.5–2 8 8 49 35 1 – 2–3 3 8 49 40 1 – 3–4 – – 54 45 1 – 4–5 – – 59 40 1 – 5–6 – – 59 40 1 – 6–7 – – 59 40 1 – 7–8 – – 59 40 1 – 8–9 – – 59 40 1 – 9–10 – – 59 40 1 – (b) 2-Mile Pickup Distance Less than 1 60 30 – 10 – – 1–1.5 50 35 – 15 – – 1.5–2 36 26 – 37 1 – 2–3 3 8 49 40 1 – 3–4 – – 54 45 1 – 4–5 – – 59 40 1 – 5–6 – – 59 40 1 – 6–7 – – 59 40 1 – 7–8 – – 59 40 1 – 8–9 – – 59 40 1 – 9–10 – – 59 40 1 – (c) No School Bus Service Less than 1 60 30 – 10 – – 1–1.5 50 35 – 15 – – 1.5–2 36 25 – 37 2 – 2–3 22 16 – 60 2 – 3–4 5 10 – 82 3 – 4–5 – 5 – 92 3 – 5–6 – – – 97 3 – 6–7 – – – 97 3 – 7–8 – – – 97 3 – 8–9 – – – 97 3 – 9–10 – – – 97 3 –

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment distances—1 mile and 2 miles—and the case of no school bus service. In all cases, students who live close to school predominantly walk. As the living distance from school increases, more students take a school bus or are driven by someone else. As the minimum school bus pickup distance increases, more students have to rely on other, non–school bus modes. For example, it was assumed that 49 percent of the students living between 1.5 and 2 miles from school would take a school bus if it were available. However, if bus service were eliminated for these students, a sizable number would shift to walking or bicycling, and a few more would be driven by their parents. Tables 5-2 through 5-4 convert these distributions of children by travel mode into average numbers of children/day and total student-miles traveled/year by mode for the three school bus pickup policies. Fractional numbers are permitted because students may use different modes on different days. To compute total distance traveled by students living at the various distances and taking the different modes, an average trip-length multiplier was used. It was assumed that a student walking to school would take a fairly direct path and would travel a distance closest to the direct distance; the multiplier for a walking student was estimated to be 1.25 times the direct distance. School buses, with their more complex routing, had the largest multiplier (1.75 times the direct distance). Bicycling and the other motor vehicle categories had multipliers of 1.30 and 1.50, respectively. For an actual school, these numbers would most likely have to be approximated. Using these multipliers and the number of children using each transportation mode at each distance from school, the total annual miles for each mode can be calculated. These results are shown in the Tables 5-2(b), 5-3(b), and 5-4(b) for the three different policies. The number of total miles may, at first glance, appear to be surprisingly large. For the 1-mile minimum school bus pickup distance [Table 5-2(b)], the students log more than 285,000 miles riding school buses, nearly 175,000 miles in parents’ cars (drivers age 19 and older), more than 12,000 miles walking, nearly 11,000 miles bicycling, and more than 4,000 miles in cars with drivers younger than 19. The overall total is just under 0.5 million student-miles per year for this relatively small elementary school. Using the hazard rates from Chapter 3 (Tables 3-4 and 3-5, using mean values for ages 5–10) and these annual student-miles by travel mode, risk estimates for the school can be calculated [Tables 5-2(c), 5-3(c), and 5-4(c)]. For example, from Table 3-4, the average injury rate per 100 million student-miles for a child aged 5–10 bicycling is 2,625. In this hypothetical school, the students bicycle 10,962 miles per year under the 1-mile minimum school bus pickup distance [Table 5-2(b)]. Thus, the expected number of injuries per year for children who bicycle is This number appears in Table 5-2(c) under the bicycle column. Similar calculations are done for the other modes, for fatalities, and for the different pickup distances. These values can be summed across all modes to obtain an overall injury and fatality risk for the school.

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment TABLE 5-2 Scenario 1: Travel and Risk Measures for 1-Mile Minimum School Bus Pickup Distance     Walking Bicycling School Bus Adult Driver Teen Driver (a) Number of Students by Transportation Mode Miles from school   Less than 1 28.5 14.3 – 4.8 –   1–1.5 7.1 4.8 8.3 3.6 –   1.5–2 1.4 1.4 9.2 6.6 0.2   2–3 0.9 2.8 18.4 15.0 0.4   3–4 – – 17.6 14.6 0.3   4–5 – – 16.2 11.0 0.3   5–6 – – 13.3 9.0 0.2   6–7 – – 10.3 7.0 0.2   7–8 – – 7.4 5.0 0.1   8–9 – – 4.4 3.0 0.1   9–10 – – 1.5 1.0 0.0 Total students 38 23 107 81 2 % of students 15 9 43 32 1 (b) Student Miles per Year by Transportation Mode Miles from school   Less than 1 6,413 3,463 – 1,283 –

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment   1–1.5 4,008 2,886 6,546 2,405 –   1.5–2 1,107 1,196 10,129 6,202 177   2–3 1,055 3,417 28,941 20,250 506   3–4 – – 38,698 27,641 614   4–5 – – 45,998 26,730 668   5–6 – – 45,998 26,730 668   6–7 – – 42,281 24,570 614   7–8 – – 34,847 20,250 506   8–9 – – 23,696 13,770 344   9–10 – – 8,828 5,130 128 Total miles/year 12,582 10,962 285,961 174,960 4,227 % miles 3 2 59 36 1 (c) Risk Measures   Injuries/year 0.09 0.29 0.04 0.13 0.11   % of injuries 14 44 6 20 16   Fatalities/year 0.0017 0.0023 0.0003 0.0005 0.0006   % of fatalities 32 43 5 9 12   Total injuries per year 0.659           Total fatalities per year 0.0054        

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment TABLE 5-3 Scenario 1: Travel and Risk Measures with 2-Mile Minimum School Bus Pickup Distance     Walking Bicycling School Bus Adult Driver Teen Driver (a) Number of Students by Transportation Mode Miles from alchol   Less than 1 28.5 14.3 – 4.8 –   1–1.5 11.9 8.3 – 3.6 –   1.5–2 6.8 4.9 – 6.9 0.2   2–3 0.9 2.8 18.4 15.0 0.4   3–4 – – 17.6 14.6 0.3   4–5 – – 16.2 11.0 0.3   5–6 – – 13.3 9.0 0.2   6–7 – – 10.3 7.0 0.2   7–8 – – 7.4 5.0 0.1   8–9 – – 4.4 3.0 0.1   9–10 – – 1.5 1.0 0.0 Total students 48 30 89 81 2 % of students 19 12 36 32 1 (b) Student Miles per Year by Transportation Mode Miles from school   Less than 1 6,413 3,463 – 1,283 –

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment   1–1.5 6,680 5,050 – 2,405 –   1.5–2 5,316 4,146 – 6,556 177   2–3 1,055 3,417 28,941 20,250 506   3–4 – – 38,698 27,641 614   4–5 – – 45,998 26,730 668   5–6 – – 45,998 26,730 668   6–7 – – 42,281 24,570 614   7–8 – – 34,847 20,250 506   8–9 – – 23,696 13,770 344   9–10 – – 8,828 5,130 128 Total miles/year 19,463 16,076 269,286 175,314 4,227 % miles 4 3 56 37 1 (c) Risk Measures   Injuries/year 0.14 0.42 0.04 0.13 0.11   % of injuries 17 50 4 16 13   Fatalities/year 0.0026 0.0034 0.0003 0.0005 0.0006   % of fatalities 36 46 4 6 9   Total injuries per year 0.842           Total fatalities per year 0.007        

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment TABLE 5-4 Scenario 1: Travel and Risk Measures for No School Bus Service   Walk Bike School Bus Adult Driver Teen Driver (a) Number of Students by Transportation Mode Miles from school Less than 1 28.5 14.3 – 4.8 – 1–1.5 11.9 8.3 – 3.6 – 1.5–2 6.8 4.7 – 6.9 0.4 2–3 8.3 6.0 – 22.5 0.8 3–4 1.6 3.3 – 26.7 1.0 4–5 – 1.4 – 25.3 0.8 5–6 – – – 21.8 0.7 6–7 – – – 17.0 0.5 7–8 – – – 12.1 0.4 8–9 – – – 7.3 0.2 9–10 – – – 2.4 0.1 Total students 57 38 – 150 5 % of students 23 15 – 60 2 (b) Student Miles per Year by Transportation Mode Miles from school Less than 1 6,413 3,463 – 1,283 – 1–1.5 6,680 5,050 – 2,405 – 1.5–2 5,316 3,987 – 6,556 354 2–3 9,281 7,290 – 30,375 1,013 3–4 2,559 5,528 – 50,369 1,843 4–5 – 3,007 – 61,479 2,005 5–6 – – – 64,820 2,005 6–7 – – – 59,582 1,843 7–8 – – – 49,106 1,519 8–9 – – – 33,392 1,033 9–10 – – – 12,440 385 Total miles/ year 30,248 28,325 – 371,807 11,998 % miles 7 7 – 86 3 (c) Risk Measures Injuries/year 0.22 0.74 – 0.29 0.31 % of injuries 14 48 0 18 20 Fatalities/year 0.0041 0.0060 – 0.0010 0.0018 % of fatalities 32 46 0 8 14 Total injuries per year 1.555         Total fatalities per year 0.013        

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment For the 1-mile school bus pickup distance, the total injury rate is approximately 1 injury every 1.5 years1 and 1 fatality every 185 years.2 The majority of these risks (more than 70 percent) involve students who walk and ride bicycles, even though these students account for just 25 percent of the trips and log only 5 percent of the student-miles traveled. If the minimum school bus pickup distance is increased to 2 miles (Table 5-3) or if school bus service is eliminated altogether (Table 5-4), more students will depend on transportation modes that are disproportionately riskier, and injuries and fatalities per year will increase. To perform this analysis, it was assumed that students who no longer took a school bus would adopt new transportation modes in rough proportion to those already using the various modes. This impact can be seen in Figure 5-1: increasing the minimum pickup distance from 1 to 2 miles would increase the student risk by more than 27 percent, while eliminating school bus service would more than double the student risk as compared with a 1-mile pickup policy. Scenario 2: Adding School Bus Service for Students Attending After-School Activities A hypothetical suburban middle school with 750 students was used to demonstrate how additional after-school bus service might affect total transportation FIGURE 5-1 Scenario 1: effects of minimum school bus pickup distances on student injuries and fatalities per year. 1 This injury rate includes all severities of nonfatal injury (Types A, B, and C). Most injuries are Type C and do not require significant medical treatment. 2 Looking at the risk from another perspective, given 100 schools of this size, two-thirds of the schools will have one injury every year, and only 0.9 schools will have a fatality every year.

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment risk. The distribution of the transportation modes for the base case with no after-school bus service is shown in Table 5-5(a). This distribution includes all students for the afternoon trip from school. Most students either take the bus (35 percent) or are driven by a parent (35 percent); nearly 20 percent walk, and 10 percent bicycle. Because of its suburban location, no public transit service was assumed for this school. Using the per student trip risk measures derived in Chapter 3 (Tables 3-6 and 3-7, using mean values for children aged 11–13) and the expected number of trips per day by mode from Table 5-5(a), and assuming 180 days per school year, annual risks were calculated. For bicycle trips, for example, there are 2,057 injuries per 100,000,000 trips for those aged 11–13 (Table 3-6), and 75 students ride bicycles every day [Table 5-5(a)]. Therefore, the calculation for bicycle injuries is TABLE 5-5 Scenario 2: Effects of Adding Bus Service for Students Participating in After-School Activities, Showing Afternoon Trips for All Students (Students 11–13 Years Old)   % of Mode No. of Students Injuries per Year Fatalities per Year (a) No After-School-Activity Bus Service School bus 35 263 0.05 0.0001 Other bus 0 – – – Passenger vehicle (adult driver) 35 263 0.20 0.0006 Walking 19 143 0.09 0.0010 Bicycling 10 75 0.28 0.0016 Passenger vehicle (teen driver) 1 8 0.04 0.0003 Total 100 750 0.66 0.0035 Years between events     1.52 283 (b) After-School-Hours-Activity Bus Service Added School bus 48 360 0.07 0.0001 Other bus 0 – – – Passenger vehicle (adult driver) 30 225 0.17 0.0005 Walking 15 113 0.07 0.0008 Bicycling 7 53 0.19 0.0011 Passenger vehicle (teen driver) 0 – – – Total 100 750 0.51 0.0025 Years between fatalities     1.98 396 (c) Net Effect of New Policy Change in risk (%)     −23 −29

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment As in the previous scenario, this calculation is repeated for all modes and the fatality risk measures. The results are then summed to determine the injury and fatality rates for the entire school over 1 year. To put these small decimal values in perspective, the number of years between injuries and fatalities (the reciprocal of the annual rates) is also shown. For this hypothetical school, a student injury can be expected to occur slightly more than once every 1.5 years and a fatality to occur on average once every 280 years. In this hypothetical example, it was assumed that 120 students (approximately 15 percent of the student population) participate in some type of after-school activity (e.g., sports, clubs, band). With the addition of a new bus service in the late afternoon, some of the students who would previously have walked, bicycled, or received a ride from a parent now take the bus. The new distribution of transportation modes is shown in Table 5-5(b). With the shift away from the riskier transportation modes to the relatively safer school bus category, the students’ overall risk is reduced. In this hypothetical case, the risk of injuries decreases 23 percent per year and the risk of fatalities decreases 29 percent per year with the new bus service. Scenario 3: Increasing Student Parking at a High School The third scenario involves a hypothetical suburban/urban high school with 2,400 students. Mass transit is available for some of the students (7.5 percent), but most commute by school bus (30 percent) or by passenger vehicle with adult driver (22.5 percent) or teen driver (22.5 percent). It is assumed that the neighborhood around the school has been affected by student parking on local streets and that a permit-parking program is now strictly enforced to prevent unauthorized student parking. More students would drive if they could find legal parking. Because the risk measures in Chapter 3 vary with the age of the students, two separate calculations are necessary: one for those aged 14–15 and one for those aged 16–18. Table 5-6(a) shows the initial distribution of transportation modes for these two age categories. As in the second scenario, the per student trip risk measures are used to calculate annual injury and fatality rates and times between injuries and fatalities. Once again the risk measures from Chapter 3 (Tables 3-6 and 3-7 for those aged 14–15 and 16–18, respectively) are used in conjunction with specific information about the school and the students’ travel modes. Following the same process as in the previous scenario, risks per trip and trips per year are used to calculate an estimate of the total risk for the school. These values are shown at the bottom right of Table 5-6(a). In this case, the student population will average nearly eight injuries per year and have a fatality once every 20 years. Of all the scenarios, this one has the greatest negative effect on school travel safety. If the school encourages student driving by doubling the number of student parking spaces, the risks faced by the student population increase considerably. Travel shifts from the relatively safer school bus and other bus modes

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment TABLE 5-6 Scenario 3: Effects on Travel Risk of Increasing Student Parking     % of Mode No. of Students Injuries per Year Fatalities per Year (a) No After-School-Activity Bus Service 14–15 Years Old           School bus 35 420 0.21 0.0004   Other bus 10 120 0.02 0.0000   Passenger vehicle (adult driver) 20 240 0.45 0.0011   Walking 15 180 0.18 0.0023   Bicycling 5 60 0.28 0.0010   Passenger vehicle (teen driver) 15 180 2.76 0.0184   Total for age group 100 1,200 3.90 0.023   Years between events     0.3 43.1 16–18 Years Old           School bus 25 300 0.45 0.0002   Other bus 5 60 0.03 0.0000   Passenger vehicle (adult driver) 25 300 0.99 0.0048   Walking 12 144 0.18 0.0019   Bicycling 3 36 1.00 0.0032   Passenger vehicle (teen driver) 30 360 1.38 0.0152   Total for age group 100 1,200 4.02 0.025   Years between events     0.2 39.4 School total     7.92 0.05 Years between events for school     0.13 50.56 (b) 600 Student Parking Spaces 14–15 Years Old           School bus 25 300 0.15 0.0003   Other bus 5 60 0.01 0.0000   Passenger vehicle (adult driver) 20 240 0.45 0.0011   Walking 12 144 0.14 0.0019   Bicycling 3 36 0.17 0.0006   Passenger vehicle (teen driver) 35 420 6.43 0.0428   Total for age group 100 1,200 7.36 0.047   Years between events     0.1 21.4 16–18 Years Old           School bus 15 180 0.32 0.0001

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment     % of Mode No. of Students Injuries per Year Fatalities per Year   Other bus 5 60 0.01 0.0000   Passenger vehicle (adult driver) 20 240 1.24 0.0038   Walking 7 84 0.14 0.0011   Bicycling 3 36 0.60 0.0032   Passenger vehicle (teen driver) 50 600 2.76 0.0254   Total for age group 100 1,200 5.07 0.034   Years between events     0.2 29.7 School total     12.43 0.08 Years between events for school     0.08 12.43 (c) Net Effect of New Policy Change in risk (%)     +57 +65 to the less safe passenger vehicle with a teen driver. Moreover, interaction among the modes at the school, though not modeled in this scenario, will increase the risk to school-age pedestrians and bicyclists, who will now have to deal with increased traffic density as a result of the greater availability of parking. To complete the analysis, it was necessary to make assumptions about how these shifts would occur; the results are shown in Table 5-6(b). Injuries per year increase 57 percent to more than 12 per year, and fatalities per year increase nearly 65 percent to 1 every 13 years. As expected, the results are highly sensitive to the assumptions made about travel mode shifts. For this school, any policy resulting in an increase in teens driving to school is not advocated. CONCLUSIONS The scenarios presented in this chapter illustrate how the risk measures from Chapter 3 can be used with local school information to complete an assessment of the risks associated with different school travel policies. In these examples, adding after-school bus service or changing the minimum school bus pickup distance can easily increase or decrease injury and fatality risks by 20 to 50 percent or more. Though these cases are hypothetical, they are realistic. The use of risk estimates based on national averages does limit the accuracy of results for specific applications, but these types of analyses nonetheless provide policy makers with important insights into the risks associated with the different modes. As noted in earlier chapters, local conditions can change the magnitude of and relationships among these risks. Schools that have in place many of the risk reduction options suggested by the safety checklists presented in Chapter 4 can have risk rates below those used in these scenarios. For example, a well-designed

OCR for page 145
The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment network of bicycle paths separated from motor vehicle traffic would likely reduce the risks of bicycling below those shown in Chapter 3 and used in this chapter. Similarly, improvements in sidewalks, speed limit enforcement, and deployment of crossing guards would likely reduce the risks associated with walking to school to levels below those used in the scenarios. Unfortunately, the data available to the committee were not sufficient to determine the extent of risk reduction associated with these and other risk mitigation options. Nevertheless, before adopting policies that shift the distribution of travel among modes, policy makers would be well advised to consider the trade-offs involved. Budgetary and other criteria must be considered in conjunction with the injury and fatality risks derived in this study and illustrated in the above scenarios. School transportation solutions to districtwide equity or school-choice problems can have safety implications. Increases in risk might be justified if the monetary savings were applied to other, more pressing problems; however, large increases in risk in exchange for modest savings could prove difficult to defend.