�ݺ�ߣ

Australasian College of Road Safety Conference
“A Safe System: Making it Happen!” Melbourne 1-2 September 2011

Speeding Behaviour in School Zones

Ellison, A.B.1, Greaves, S.P.1 & Daniels, R.1
1
Institute of Transport and Logistics Studies, University of Sydney

Abstract
Introduction: Speeding in school zones is a significant and sensitive safety issue. School
zones have high levels of pedestrian activity that are particularly vulnerable in the event of a
crash. Yet, other than from targeted police enforcement there is little information on day-to-
day speeding in school zones.
Methods: Using GPS, spatial, demographic and psychological data we analyse the speeding
behaviour in school zones of 147 motorists in Sydney over a period of several weeks. The
focus is on both the duration and magnitude of speeding and whether this is related to
particular driver, trip, and roadway characteristics, with a particular emphasis on the most
prolific speeders.
Results: Overall, 23 percent of the distance travelled in school zones is above the speed limit,
higher than for any speed limit other than motorways. Additionally, all drivers exceed the
speed limit and there are a small but noteworthy number of drivers who consistently exceed
the speed limit for as much as half the distance travelled.
Conclusions: Despite efforts at reducing speeding in school zones, these results show that
speeding remains very common. The consequences of speeding in school zones make it
particularly important that additional resources are put into reducing speeding. The results
suggest that targeting the road environment such as transitions to school zones may be more
effective than psychological/personality-based campaigns in changing drivers’ speeding
behaviour.

Keywords
Speeding, School Zones, GPS, Driver Profiling, Psychological Factors

Introduction
Despite the well-documented dangers associated with speeding behaviour, excessive speed
continues to play a major role in serious road crashes. For instance, recent evidence suggests
that speeding was a ‘major factor’ in one-third of fatal road crashes in Australia in 2006
(Bureau of Infrastructure Transport and Regional Economics, 2011). More worryingly, the
prevalence of speeding as a major factor has increased from around one-fifth of fatal crashes
in 1990 to the current levels (BITRE, 2011), suggesting that despite strategies and campaigns,
the message simply is not getting through to motorists. The consequences of speeding are
particularly acute in areas with high levels of vulnerable road users (Pedan et al., 2004) of
which school zones are a prominent example (Roper et al., 2006). Although there have been
an increasing number of studies on the effectiveness of school zones (Fitzpatrick et al., 2009;
Tay, 2009; Kattan et al., 2011), there has been little research into the impact on speeding of
drivers’ demographics, personality and trip characteristics in this context. Previous research
has been limited by the reliance on police enforcement records and occasional data collection
using radar/laser speed guns and traffic/vehicle counting devices. These methods potentially
understate the extent of speeding behaviour in school zones.

With this in mind, this paper presents the results of an analysis of speeding behaviour in
school zones in the Sydney metropolitan area. The data used in the analysis represents several
weeks of driving collected from 147 motorists using a Global Positioning System (GPS)

1


device as part of a broader study of driving behaviour (Greaves et al., 2010). Crucially,
motorists were unaware that speeding was specifically being monitored and recorded because
of the (obvious) potential for this in itself to impact behaviour and analysis. GPS data
provides information into drivers’ day-to-day speeding behaviour across time. The paper is
structured as follows. First, the literature on speeding in school zones is reviewed, followed
by an overview of the methodology. The most important results are presented and the paper
concludes with a discussion of the results and policy considerations.

Literature review
School zones operate differently in each jurisdiction. In the state of New South Wales (where
this study took place), a speed limit of 40 km/h is applied in school zones on weekdays from
08:00 to 09:30 and from 14:30 to 16:00 during the four school terms of ten weeks that operate
over the calendar year (i.e., forty weeks total/annum). Various types of road markings and
signage are used to indicate to drivers where school zones begin and end (Roads and Traffic
Authority NSW, 2010). In Alberta, Canada, school zones maintain a 30 km/h speed limit from
08:00 to 16:30 on school days and some use speed displays (Kattan et al., 2011). In the United
States, speed limits vary by state. However, most states use 15 miles/hr (24 km/h) or 20
miles/hr (32 km/h). California, Kentucky, Michigan and Oklahoma apply a minimum speed
limit of 25 miles/hr (40 km/h). A number of other states do not have uniform school zone
speed limits (Fitzpatrick et al., 2009). In New Zealand there is no national school zone speed
limit, however 40 km/h variable speed limit zones (with signage) are increasingly common
(Mackie, 2010). In many European countries a speed limit of 30 km/h applies in residential
areas at all times (Garrard, 2009).

The need for drivers to abide by speed limits is particularly important in school zones given
the high levels of child pedestrian activity and the consequences of a collision. Recent
evidence suggests the likelihood of a fatality in the event of a pedestrian-vehicle collision is
more than double with an impact speed of 50 km/h compared to 40 km/h (Pedan et al., 2004).
At a speed of 30 km/h, the likelihood of a pedestrian fatality is 10 percent, significantly lower
than 40 percent at a speed of 40 km/h (Pedan et al., 2004). Evidence from Sydney and South
West New South Wales (covering one quarter of NSW school zones) shows the
implementation of school zones resulted in a 45 percent reduction in pedestrian casualties (all
ages, 46 percent for children) in school zones (Graham & Sparkes, 2010) but few crashes
have resulted in child fatalities (Hanan et al., 2011) and even fewer (5% of reported crashes,
1% of reported school zone crashes involving 5-16 year old pedestrians) were due to speeding
(Graham & Sparkes, 2010). Nonetheless, the danger and perception of danger posed by
speeding motorists is a significant barrier to more active transport (walking and cycling)
which has declined markedly over the last twenty years (Garrard, 2009). It is possible that the
reductions in school zone crashes and fatalities reflect this modal shift from walking and
cycling to cars rather than an improvement in behaviour or adherence to school zone speed
limits. An analysis of school zone crashes using exposure may have different results than
those found by Graham & Sparkes (2010). Furthermore, some studies have shown children to
be more likely to be involved in vehicle collisions near school zones than further away, most
likely the result of increased exposure (Abdel-Aty et al., 2007; Warsh et al., 2009).

There have been a number of studies into speeding in school zones. In terms of compliance
rates, using traffic counters Roper et al. (2006) determined that depending on the type of
signage in use, around half the vehicles exceeded the speed limit in New South Wales school
zones. As many as 45 percent of drivers also exceed the speed limit by 10 km/h or more in
some school zones and up to 10 percent by at least 20 km/h. These results are consistent with

2


a study conducted in Texas which found that up to 17 percent of vehicles were driven at
speeds of at least 25 miles/hr (40 km/h) in 20 miles/hr (32 km/h) school zones. The same
study found that as many as 59 percent of vehicles were driven at speeds of at least 35
miles/hr (56 km/h) in 30 miles/hr (48 km/h) school zones indicating that the speed limit
appears to be a factor in compliance (Hawkins, 2007). Although these are surprisingly high
rates of speeding, there appears to be relatively small changes in compliance rates from a
study conducted in 1999 where 45 percent of drivers exceeded the speed limit by at least 5
miles/hr (8 km/h) (Saibel et al., 1999). To a lesser extent, the same is true in Australia where a
2002 study in Perth determined that 50 percent of vehicles exceeded the 40 km/h speed limit
by 10 km/h or more, but it also found that drivers do slow down in school zones (Radalj,
2002).

More recently, a study conducted using laser guns in Calgary (Alberta, Canada) found 53
percent of vehicles sped in school zones and 11 percent sped by 10 km/h or more (Tay, 2009).
A follow up study had similar results, finding 54 percent of vehicles exceeded the speed limit
and 10 percent of vehicles exceeded the speed limit by 10 km/h or more. The mean speed
(31.4 km/h) was slightly above the school zone speed limit but lower than the default speed
limit of 50 km/h. These results varied considerably depending on the characteristics of the
school zones. For example, zones with fencing had a compliance rate of 51.3 percent
compared to 64.8 percent for zones without fencing (Kattan et al., 2011).

Many studies have focused on the impact of signage, road markings and other speed
management techniques on compliance with speed limits in school zones. In terms of vehicle
speeds in school zones, it appears that the normal speed limit of the road appears to be a more
important factor in vehicle speeds than the speed limit of the school zone (McCoy &
Heimann, 1990). This may reflect the characteristics of the roads themselves rather than the
speed limit itself and may explain why many studies have found large variations in speeding
between school zones in the same geographic area (Roper et al., 2006; Hawkins, 2007).
Kattan et al. (2011) found the presence of children to be a factor in (lower) vehicle speeds as
well as road characteristics such as the use of speed display signs, fencing, fewer lanes and
longer school zones. These results are consistent with those of Tay (2009) but in the case of
school zone length somewhat contradicts the results of a study conducted in Texas. In that
study, drivers reached their minimum speed within the first half of the school zone but speeds
increased steadily as the distance from the start of the school zone increased (Fitzpatrick et al.,
2009). The use of electronic signs was also found to be effective in New Zealand, however
compliance was greater when they were combined with permanent signage and the presence
of children (Osmers, 2001). Roper et al. (2006) found the use of flashing lights on school
zone speed limit signs to have a statistically significant impact on reduced vehicle speeds in
school zones.

Overall, previous research has found compliance rates with school zone speed limits to vary
depending on the characteristics of the school zone but in all cases compliance rates can be
considered inadequate, below 50 percent of vehicles in some school zones. However, none of
these studies were able to determine the characteristics of the driver such as age, sex, trip
purpose or personality profile and therefore the influence of these characteristics on their
speeding behaviour.

Methods
Over a period of several weeks, 147 drivers in Sydney were tracked using GPS devices as part
of a broader study on driver behaviour (Greaves et al., 2010). Driver speeds and speed limits

3


were logged but participants were not made aware of this at the time due to the possibility of
this affecting behaviour. Note that data were excluded from the first week of monitoring to
mitigate any potential ‘halo’ effects resulting from the installation of the GPS device leaving
five weeks of data for this analysis. The GPS data were processed against a GIS-based
(Geographic Information System) speed limit and road network database to identify speeding.
Additional data sources included a demographic survey and a psychological survey conducted
on each driver at the time of recruitment (Greaves & Ellison, 2011) supplemented by
additional sources of spatial information such as weather and the location of school zones.
Data sources are summarised in Table 1.

Table 1: Summary of data sources
GPS data Second-by-second data including latitude, longitude, vehicle speed, speed limit, date
and time.
Demographics Collected from participants during recruitment, includes age, gender, vehicle
information, household size and other common attributes.
Psychological 50 question, 5 section survey incorporating personality, risk perception and driver
survey behaviour scales (Greaves & Ellison, 2011).
Trip details Using a website updated daily, drivers indicated the purpose, driver and the number of
passengers for each trip (Greaves et al., 2010). Origin, destination, time and date
information were retrieved from the GPS data.
Weather Hourly rainfall data for Sydney collected by the Australian Bureau of Meteorology
from 15 observation stations was matched to the GPS data based on location.
School zones Locations of school zones were used to identify school zone driving.

The second-by-second GPS data were aggregated into road segments such that a new segment
began at the start of each trip, school zone or when the speed limit changed (for segments that
were not school zones). Since the intention here was to assess the compliance of drivers to the
school zone speed limit as a function of distance and school zones, the aggregated variables
did not result in much loss of detail.

For the purposes of this analysis only data from primary drivers, defined as the driver who
completed the demographic and psychological surveys, was included. Of the 147 drivers, 16
had no driving recorded in school zones and therefore were excluded. A further 12 drivers had
very little driving recorded in school zones (less than 1 km) and were therefore also excluded.
After excluding drivers with little or no driving in school zones, 119 drivers remained for
further analysis.

In total more than 1,300 km was driven in school zones by the 119 selected drivers
representing one percent of the total kilometres driven over the monitoring period. On
average, 11.4 km or 1.2 percent of distance driven was in school zones with a standard
deviation of 1.5 percent. As was expected, drivers with a greater proportion of school zone
driving tended to have a greater proportion of school zone distance during education trips
whilst work trips exhibited the opposite trend.

Results
Overall drivers exceed the speed limit in school zones for 23 percent (i.e., 300 kilometres in
total) of the distance driven in school zones which is higher than the average (20 percent) and
any other speed limit other than 100 and 110 km/h roads. 50 km/h roads, many of which are
residential, have only marginally less speeding as shown in Figure 1. More concerning given
the more severe consequences of speeding in school zones is that higher magnitude speeding
(>=10 km/h) appears to be more prevalent than on roads of any other speed limit.

4


Figure 1: Percentage of total distance speeding by road speed limit
% of Distance Speeding 35%
30%
25%
20%
15%
10+ km/h over
10%
1-9 km/h over
5%
0%
40 50 60 70 80 90 100 110 All
School Speed
Zone Speed Limit (km/h) Limits

Further analyses were conducted to determine differences in speeding behaviour between
personal, trip, vehicle and road characteristics. Analysis of variance (ANOVA) tests were
performed to assess the statistical significance of any observed differences. These results are
discussed below and summarised in Table 2 at the end of this section.

Personal characteristics
Age and gender are commonly used variables for classifying drivers. In total drivers aged 18
to 30 comprised 24 percent of the final sample of which 31 percent were male and 69 percent
female. Drivers 31 to 45 and 46 to 65 years of age accounted for 44 percent and 32 percent
respectively divided fairly evenly between males and females.

In school zones, results show that although younger drivers speed slightly more (24 percent)
than drivers aged 31-45 (23 percent) and 46-65 (20 percent) these differences are not
statistically significant. Interestingly, it appears that drivers on provisional licenses speed
considerably less frequently in school zones (12 percent) than drivers holding full licenses (24
percent) suggesting that the more severe penalties applied to provisional drivers may be
effective. This relationship is statistically significant.

As part of the study, drivers completed a five section, fifty question psychological survey.
The sections of relevance here are personality and risk perception scales. The personality
scales (aggression, excitement seeking and altruism) are measured on a ten point scale from
“Not at all” to “Very much” and have been found to be correlated with self-reported speeding
behaviour (Ulleberg & Rundmo, 2003). Two risk perception scales are also included – worry
and concern, and likelihood of an accident, for the driver and others of the same age,
measured on a five point and ten point scale respectively (Rundmo & Iversen, 2004; Machin
& Sankey, 2008). These scales have been adjusted to a common 100 point scale to allow for
better comparisons between the different personality scales.

Drivers’ answers to the personality survey were converted into five normalised personality
scales representing aggression, altruism, excitement, worry and concern and the perceived
likelihood of an accident. For each driver, these scales were plotted against the percentage of
distance speeding in school zones and logarithmic trend lines were calculated (see Figure 2).
While the trends are generally in line with a priori expectations (i.e., increasing altruism and
worry/concern are associated with less speeding, while increasing aggression and excitement
are associated with more speeding), there is significant heterogeneity across the participants.

5


This suggests that psychological factors play a role in school zone speeding but this is
influenced by other factors such as the opportunity to speed and other characteristics of the
road environment. The impact on speeding in school zones of perceptions of the dangers of
speeding also appears mixed. However, this may be a reflection of the lack of any questions
specifically addressing school zones.

Figure 2: Psychological scales by percentage of school zone speeding
Normalised Psychological Scales (0 – 100)

Percentage of Distance Speeding in School Zones

The percentage of distance speeding by each driver (Figure 3) included in the study shows
there is considerable heterogeneity in driver behaviour both in terms of overall speeding and
magnitude. All participants exceeded the school zone speed limit (by 10 km/h or more) at
some point over the five weeks with 15% of the sample exceeding the speed limit for more
than 40% of the distance. Arguably of more concern is the magnitude of the speeding as this
suggests negligent or ignorant behaviour – Figure 3 suggests that many of the most frequent
speeders drive for a considerable proportion of the driving distance at speeds in excess of 10
km/h above the speed limit (i.e. more than 50 km/h in a 40 km/h zone).

6


Figure 3: Percentage of distance speeding in school zones by driver
70%
% of Distance Speeding

60%
50%
40%
30%
20%
10%
0%
Driver (n = 119)
1-9 km/h over 10+ km/h over

Trip characteristics
Intuitively the characteristics of the trip such as trip purpose and number of passengers may
influence drivers’ speeding behaviour in school zones. As expected, a gradual downward
trend in speeding behaviour in school zones exists in terms of overall speeding in school
zones by number of passengers. As the number of passengers increases, overall speeding is
reduced from 25% (drive alone and one passenger), to 19% (two passengers), and 18% (3+
passengers) but this is not the case for speeding by 10 km/h or more. Neither trend is
statistically significant.

In terms of trip purpose (Figure 4), education trips have the lowest levels of overall speeding
in school zones (19%) and work trips have the highest (28%). This may reflect that drivers are
likely to be more aware of school zones during education trips when they are more likely to
be carrying children. However, as with speeding by number of passengers, the same trend
does not hold for speeding by 10 km/h or more. The differences between work-related,
education and other trip purposes are statistically significant for speeding 1 km/h or more but
not for 10 km/h or more. There is little temporal difference in speeding behaviour by day of
the week.

Figure 4: Percentage of distance speeding in school zones by trip purpose
30%

25%
20%
15%
10% 10+ km/h over
5%
1-9 km/h over
0%

Vehicle characteristics
In terms of the vehicle characteristics, the type of vehicle does not appear to be a factor in
speeding in school zones which contradicts some of the broader literature on speeding (Liu,
2007). At an aggregate level, the results suggest that drivers of newer vehicles (manufactured
since 2005) speed more frequently – for 26 percent of the distance – compared to 23 percent
(manufactured from 2000 to 2004) and 16 percent (manufactured before 2000) but this
relationship is not statistically significant.

7


Road characteristics
Previous research (Fitzpatrick et al., 2009; Tay, 2009; Kattan et al., 2011) has found road
characteristics to be a significant factor in drivers’ compliance with school zone speed limits.
In this study, the speed limit of the previous road segment was used as a (crude) proxy for the
type of road characteristics as detailed characteristics of school zones (e.g., presence of
flashing lights, dragon’s teeth road markings, speed humps, road narrowing, etc.) were not
readily available in an easily accessible digital format. It should be noted that the speed limit
used here is the speed limit in effect in the previous road segment for each driver. It may not
be the same as the speed limit of the school zone segment outside of a school zone’s operating
hours. Figure 5 shows that in general, the higher the speed limit of the previous segment, the
greater the distance driven above the speed limit. In contrast to other trip characteristics, this
effect is stronger for higher magnitudes of speeding. Rain does not appear to be a significant
factor in school zone speeding.

Figure 5: Percentage of distance speeding in school zones by speed limit before school zone
35%

30%
25%
20%
15%
10+ km/h over
10%
5% 1-9 km/h over
0%
50 60 70 80 90 100 All
Speed Limit Before School Zone (km/h)

While it appears that the road characteristics have an impact on speeding behaviour and that
overall speeding in school zones constitutes a significant proportion of school zone driving,
Figure 6 shows that, in general, drivers do slow down for school zones, including from higher
speeds. With the exception of school zone segments following roads with 70 and 100 km/h
roads, drivers maintain an average maximum speed below the 40 km/h school zone speed
limit.

Figure 6: Mean and maximum speeds 1 in school zones by speed limit of previous segment
50
School Zone Speed

45
40
(km/h)

35
30
25
20
50 60 70 80 90 100
Speed Limit of Previous Segment (km/h)
Mean* Maximum* Speed Limit
* Differences in mean and maximum speeds by previous speed limit have been tested and found to be significant
at the p < 0.01 level.

1
Maximum and mean speeds were calculated for each school zone segment recorded. These are the means of the
maximum and mean speeds for all observed school zone segments.

8


Tests of statistical significance suggest that other than licence type (caution interpretation due
to a very small number of provisional licence holders in the sample) and the purpose of the
trip, none of the other variables were significant when taken over the sample (Table 2).
However, Figure 3 shows there are considerable differences between drivers, which suggests
we have not captured the crucial behavioural dimensions or (more likely) roadway
characteristics known to reduce speeding in school zones (Fitzpatrick et al., 2009; Kattan et
al., 2011).

Table 2: Factors, sample and significance testing of differences in speeding behaviour
Variable(s) Possible Values Percentage Distance Speeding
(% of sample in brackets) 1+ km/h 10+ km/h
Sig. Sig.
Driver Characteristics
Gender Male (43%), Female (57%) .936 .834
Age 18-30 (24%), 31-45 (44%), 46-65 (32%) .294 .336
Age x Gender Male 18-30 (8%), 31-45 (20%), 46-65 (15%) .679 .944
Female 18-30 (17%), 31-45 (24%), 46-65 (17%)
Licence Type Provisional (6%), Full (94%) .011* .129
Vehicle Characteristics
Year of Vehicle <= 1999 (20%), 2000 to 2004 (43%), 2005+ (37%) .810 .994
Manufacture
Vehicle Type Sedan (52%), Hatchback (27%), Other (21%) .993 .998
Trip Characteristics
Number of Passengers 0 (43%), 1 (26%), 2 (22%), 3+ (9%) .133 .931
Trip Purpose (6 Returning Home (20%), Work-Related (29%), .059 .044*
categories) Education (16%), Recreation (5%),
Shopping (13%), Other (17%)
Trip Purpose (3 cat.) Work-related (29%), Education (16%), Other (55%) .020* .398
Road characteristics
Previous Speed Limit 50 (22%), 60 (37%), 70 (14%), 80 (11%), 90 .099 .186
(km/h) (10%), 100 (6%)
Rain Yes (3%), No (97%) .148 .695
* Differences in percentage of distance speeding tested for significance and found to be significant
at the p < 0.05 level

Discussion
The results of this study are broadly consistent with that of previous studies on speeding in
school zones which show that reduced speed limits in school zones are effective at reducing
vehicle speeds. Overall findings are that over 20 percent of the distance driven in school
zones is at speeds above the 40 km/h speed limit and 8 percent is driven at speeds 10 km/h or
more above the speed limit. This is somewhat lower, but comparable given the different
measurements, to the results of Roper et al. (2006) which found that approximately half of all
vehicles exceeded the speed limit and those of Kattan et al. (2011) which found that 10
percent of vehicles exceeded the speed limit by 10 km/h or more. However, a significant
proportion of the distance driven in school zones remains above the speed limit, in many
cases in excess of 10 km/h above the speed limit.

As previous studies did not collect information about drivers or trips, it is not possible to
directly compare the results of the driver and trip analyses with other studies of school zone
behaviour. There is strong evidence from studies of speeding in a broader context (Ogle,
2005; Fleiter et al., 2006; Machin & Sankey, 2008) which illustrates the importance of
studying how personal and trip factors influence drivers’ behaviour. The results of this
research are consistent with that of the broader speeding literature suggesting that speeding in

9


school zones has similar characteristics to speeding behaviour in other contexts but more
study is needed, likely at a more micro level.

Furthermore, the strong link between speeds in school zones and the speed limit of segments
preceding school zones indicates that there is a need to reduce vehicle speeds in some way
before the start of the school zone. The introduction of transition or buffer areas (Fitzpatrick
et al., 2009) with reduced speed limits and speed management methods such as speed humps
are possibilities but may be problematic due to the temporal nature of school zones. Kattan et
al. (2011) and Roper (2006) identified a number of other road characteristics – including
fencing and flashing lights – that reduces drivers’ speeds without impeding vehicle flow
during times when the school zone is not operating and these may be other potential solutions.
Issues with inadequate signage and road markings which may not provide sufficient advance
warning of school zones for approaching motorists are also likely to influence speeding.

The influence of speed display signs found to be significant by a number of studies (Osmers,
2001; Kattan et al., 2011) may be indicative that drivers either are unaware they are speeding
or are influenced by perceived enforcement. Incorporating the presence of signage, flashing
lights, speed cameras – increasingly common near school zones in New South Wales (Roads
and Traffic Authority NSW, 2009) – and other road characteristics into future research would
be beneficial in determining the extent of this influence.

Conclusion
The main contribution of this paper has been to ascertain the extent of speeding (both duration
and by magnitude) in school zones within the context of normal/regular driving routines and
known driver characteristics rather than the ‘spot check’ approaches that provide most current
information. A noteworthy finding is the use of average figures on speeding duration,
typically reported, masks the fact that a small, but significant number of drivers are regularly
violating school zone speed limits by magnitudes that indicate a negligent and/or ignorant
approach to driving. It is these drivers that particular attention must be paid to through either
more draconian enforcement or physical countermeasures within the road environment to
reduce the opportunity to speed. This behaviour has implications both on injuries and
fatalities in school zones and – by virtue of perceived danger – on increased dependence on
car travel by children and adults. This trend has social and health implications which go
beyond the small number of fatalities that specifically occur in designated school zones.

The results of this research lend credence to the need to focus on the design of school zones to
require drivers to slow down. Current research is focused on refining the analysis by
incorporating physical measures characterising school zones using GIS tools including
Google Earth. This is a painstaking process, but clearly crucial to explaining the differences
observed here, which were not explainable by driver socio-demographic characteristics alone.

References

Abdel-Aty, M., Chundi, S. S. & Lee, C. (2007). Geo-spatial and log-linear analysis of
pedestrian and bicyclist crashes involving school-aged children. Journal of Safety
Research, 38(5), 571-579.

Bureau of Infrastructure Transport and Regional Economics. (2011). Fatal road crashes in
Australia in the 1990s and 2000s: crash types and major factors. Canberra, Australia.

10


Fitzpatrick, K., Brewer, M., Obeng-boampong, K., Park, E. S. & Trout, N. (2009). Speeds in
School Zones. Federal Highway Administration, Washington, D.C.

Fleiter, J., Watson, B., Lennon, A. & Lewis, I. M. (2006). Significant others, who are they? -
Examining normative influences on speeding, in Australasian Road Safety Research,
Policing and Education Conference. Gold Coast, Australia.

Garrard, J. (2009). Active transport : Children and young people. VicHealth, Melbourne,
Australia.

Graham, A. & Sparkes, P. (2010). Casualty reductions in NSW associated with the 40 km/h
school zone initiative, in Australasian Road Safety Research, Policing and Education
Conference. Canberra, Australia.

Greaves, S. P. & Ellison, A. B. (2011). Personality, risk aversion and speeding: An empirical
investigation. Accident Analysis and Prevention, 43(5), 1828-1836.

Greaves, S. P., Fifer, S., Ellison, R. B. & Germanos, G. (2010). Development of a GPS /
Web-based Prompted-Recall Solution for Longitudinal Travel Surveys. Transportation
Research Record, 2183, 69-77.

Hanan, S. A., King, M. J. & Lewis, I. M. (2011). Understanding speeding in school zones in
Malaysia and Australia using an extended Theory of Planned Behaviour : The potential
role of mindfulness. Journal of the Australasian College of Road Safety, 22(2), 56-62.

Hawkins, H. G. (2007). Rear-Facing School Speed Limit Beacons. Institute of Transportation
Engineers Journal, 77(6), 18-23.

Kattan, L., Tay, R. & Acharjee, S. (2011). Managing speed at school and playground zones.
Accident Analysis and Prevention, 43(5), 1887-1891.

Liu, B.-S. (2007). Association of intersection approach speed with driver characteristics,
vehicle type and traffic conditions comparing urban and suburban areas. Accident
Analysis and Prevention, 39(2), 216-223.

Machin, M. A. & Sankey, K. S. (2008). Relationships between young drivers΄ personality
characteristics, risk perceptions, and driving behaviour. Accident Analysis and
Prevention, 40(2), 541-547.

Mackie, H. (2010). Improving school travel systems. NZ Transport Agency, Wellington, New
Zealand.

McCoy, P. T. & Heimann, J. E. (1990). School Speed Limits and Speeds in School Zones.
Transportation Research Record, 1254, 1-7.

Ogle, J. H. (2005). Quantitative Assessment of Driver Speeding Behavior Using Instrumented
Vehicles. Georgia Institute of Technology, Atlanta, United States.

11


Osmers, W. F. (2001). The Effect on Vehicle Speeds of Electronically-Signed Part-Time
Speed Limits Outside Schools, in Australasian Road Safety Research, Policing and
Education Conference.

Pedan, M., Scurfield, R., Sleet, D., Mohan, D., Hyder, A. A., Jarawan, E., et al. (2004). World
Report on Road Traffic Injury Prevention. World Health Organisation, Geneva.

Radalj, T. (2002). Driver Speed Compliance within School Zones and Effects of “40” Painted
Speed Limit on Driver Speed Behaviours, in Australasian Road Safety Research,
Policing and Education Conference, 365-372. Sydney, Australia.

Roads and Traffic Authority NSW. (2009). Speed Camera Locations. Retrieved December 24,
2010, from
http://www.rta.nsw.gov.au/roadsafety/speedandspeedcameras/fixeddigitalspeedcameras/f
ixedspeedcameralocations/index.html

Roads and Traffic Authority NSW. (2010). School zone program. Retrieved August 15, 2010,
from
http://www.rta.nsw.gov.au/roadsafety/children/schoolroadsafety/schoolzoneprogram.htm
l

Roper, P., Thoresen, T., Tziotis, M. & Imberger, K. (2006). Evaluation of flashing lights in 40
km/h school speed zones. Arrb Consulting, Sydney, Australia.

Rundmo, T. & Iversen, H. (2004). Risk perception and driving behaviour among adolescents
in two Norwegian counties before and after a traffic safety campaign. Safety Science,
42(1), 1-21.

Saibel, C., Salzberg, P., Doane, R. & Moffat, J. (1999). Vehicle Speeds in School Zones.
Institute of Transportation Engineers Journal, 69(11), 38–43.

Tay, R. (2009). Speed Compliance in School and Playground Zones. Institute of
Transportation Engineers Journal, 79(3), 36-38.

Ulleberg, P. & Rundmo, T. (2003). Personality, attitudes and risk perception as predictors of
risky driving behaviour among young drivers. Safety Science, 41(5), 427-443.

Warsh, J., Rothman, L., Slater, M., Steverango, C. & Howard, A. (2009). Are school zones
effective? An examination of motor vehicle versus child pedestrian crashes near schools.
Injury Prevention, 15(4), 226-229.

12

�ݺ�ߣ

speeding behaviour in school zones

More Related Content

Similar to speeding behaviour in school zones (12)

speeding behaviour in school zones