Full text of DOI:10.1016/S0001-4575(99)00084-6

Accident Analysis and Prevention 32 (2000) 575–581
www.elsevier.com/locate/aap
Influence of drivers’ comprehension of posted signs on their safety
related characteristics
Hashim Al-Madani *
Department of Ci6il and Architectural Engineering, Uni6ersity of Bahrain, PO Box 32038, Bahrain
Received 22 September 1997; received in revised form 19 May 1999; accepted 16 June 1999
Abstract
The paper studies the relationship between drivers’ understanding of posted signs in three of the Gulf Cooperation Council
(
GCC) states, Bahrain, Qatar and the United Arab Emirates (UAE), and some of their safety related characteristics. These
characteristics are driving experience, accident involvement, experience per accident, citations received in the last 3 years on speed
limit violations, and seat belt usage. A total of 28 posted signs were investigated. These were categorized as warning and
regulatory. To achieve the above goals a questionnaire, specially prepared to collect the necessary data, was distributed to over
6
000 drivers in the three states. Over 2820 (47%) responded back. Comprehension of posted signs for drivers with high years of
driving experience proved to be significantly better than those with lesser experience. However, the results revealed no significant
influence on their accident involvements, even when the effect of age is incorporated; experience per accident ratios, or speed
citations. Further, the seat belt usage is also found to increase with understanding of posted signs. © 2000 Elsevier Science Ltd.
All rights reserved.
Keywords: Warning signs; Regulatory signs; Understanding; Comprehension; Accident; Seat belt; Speed; Citations; Experience
1
. Introduction
specific times, where hazards are not self-evident. Ac-
cording to US Department of Transportation (1989),
improvement in traffic signs at intersections lead to
34% reduction in fatal accidents and 93% in injury
related accidents in the United States. In fact the effec-
tiveness, as per benefit/cost ratio, appears to be the
highest compared with traffic channalization facilities,
sight distance, markings, illumination and traffic signal
improvement. The improvement in the posted signs
include the location and the illumination.
Although traffic signs are very uniform all over the
world, many countries have issued their own sign man-
uals. Traffic signs, however, are most effective when
they satisfy the following fundamentals: fulfill a need,
command attention, convey a clear and simple mean-
ing, command respect of the road users and give ade-
quate time for proper response (Institute of
Transportation Engineers, 1992). The subject of this
investigation is related to the third fundamental
parameter, namely conveying a clear meaning to the
users.
Posted road traffic signs are the most commonly used
traffic control devices. Their main function is to provide
the road users, specially the drivers, with the important
navigational messages. They serve other needs, as well,
as will follow. However, questions such as: does better
understanding of such signs mean less accident involve-
ment?; is there any relationship between drivers’ under-
standing of posted signs and their speed limit citations?;
does understanding of such signs increase with driving
experience and how is the drivers seat belt usage related
to their knowledge of posted signs?, are investigated in
this study.
Many accidents occur because the driver is suddenly
confronted by the unexpected. The driver should, there-
fore, be warned by means of traffic signs, as much as
possible, for any abnormal driving situations ahead
(
Pignataro, 1973). These signs convey messages in terms
of words or symbols. Signs are, therefore, essential
where special regulations apply at specific places or at
One of the earliest studies on posted signs was that of
Riegelneier (1942) on rehabilitation of signs. Roadway
*
Tel.: +973-681234; fax: +973-684844.
0001-4575/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0001-4575(99)00084-6

576
H. Al-Madani / Accident Analysis and Pre6ention 32 (2000) 575–581
posted signs, in fact, were evaluated by a wide range of
techniques. In the 1940s and early 1950s many of the
researches in this field were technical related studies,
such as those related to impact assessment, cost estima-
tion, accident evaluation and illumination effectiveness.
In the early 1960s researchers started to give more
weight to human related studies, particularly the mo-
torists behavior. Hakkinen (1965), for example, directly
assessed the influence of road signs on drivers’ behavior
by measuring their sign recall with such measures as
speed reduction. Other investigators measured the visi-
bility of the signs and drivers’ recognition at different
approaching speeds, taking into consideration different
social and psychological backgrounds of the drivers
This may be indicative of some sort of overconfidence
in them. Richard and Heathington (1988) made a sur-
vey of motorists’ comprehension of rail road grade
crossing traffic control devices (signs and signals). Their
results indicated that very young drivers (under 19
years) and elderly drivers (over 54 years) face
difficulties in understanding and recognizing such
devices. Significant differences between novice and ex-
perienced drivers were also observed. Drivers who dis-
regard, even plausible, speed limits face more serious
accidents and more traffic violations than those who
observe the limits. Typical observers of speed limits are
females aged between 40 and 60 years. Typical non-ob-
servers are males, between 25 and 40 years (Schmidt,
1982). Factors associated with compliance and non-
compliance with traffic regulations, as safe driving
practice and observation of regulatory signs, was inves-
tigated by Hofner (1982). Their results contradicted, at
least age wise, with Otani et al. (1992). Hofner found
compliers to be conscious (safety wise), either under 30
or over 60 years old, drove medium power cars, had
less driving experience and fewer traffic fines than non-
compliers. Typical non-compliers who tended to take
risks, were between 45 and 55 years, were executives or
self-employed and drove high power cars. Both groups,
however, provided similar ratings for different traffic
violations.
(
Johansson and Rumar, 1966; Johansson and Back-
lund, 1970). Others (Summala and Naatanen, 1974; Ells
and Dewar, 1979; Drory and Shiner, 1982), tested the
effect of illumination and brightness of the signs on
both users response and drivers’ reaction time. How-
ever, studies of drivers’ characteristics, particularly the
safety related ones, with respect to their understanding
of the informational values of the signs and how well
do such signs guide drivers are not, yet, extensively
covered.
Jabbar and Naqvi (1992) found that drivers commit
significant errors in detecting symbolic signs compared
with alphanumeric ones. Therefore, to incorporate
users’ comprehension, modifications to certain signs
might be necessary. Alphanumeric signs are better when
compared with symbolic ones, so are warning signs
compared with regulatory signs (Dewar et al., 1976)
because the drivers regard them to be more serious.
However, this was not true when ‘head room hazard’
sign was considered. Galer (1980) found that 21% of
the long truck drivers did not understand such signs.
Laboratory tests showed that drivers perceive the word
2. Aims
This study examines the influence of drivers’ under-
standing of posted signs on some of their safety related
characteristics in three of the Gulf Cooperation Council
(GCC) states; Bahrain, Qatar and UAE. These charac-
teristics include experience, accident involvement be-
cause of drivers’ own fault, experience per accident
ratio, speed citations received by the drivers in their
latest 3 years of driving and seat belt usage. More
details on experience per accident ratio is discussed
latter. To achieve the above goals the following hy-
potheses are tested:
‘danger’ to indicate the highest level of alert compared
with ‘caution’ and ‘warning’. When such words were
posted with different backgrounds, the greatest impact
was found for ‘danger’ with a red background
(
Chapanis, 1994).
Fisher (1992) measured the informational value of
road traffic signs by questioning drivers at road blocks.
He found drivers’ memory for road signs to be typically
poor. Further, he strongly recommended not to assess
the effectiveness of road signs in terms of drivers recall.
In fact, they should be assessed in terms of their ability
to sensitize the driver to hazards, regardless of those
who ignore such signs. However, Book and Bergstrom
1. If traffic sign knowledge is important for safety then
drivers with good understanding of posted signs
should be less involved in accidents compared with
those with less understanding.
2. Experienced drivers are expected to know the posted
signs better than novice ones; for obvious reasons as
higher exposure rate. Findings by Richard and
Heathington (1988), as has been mentioned earlier,
support this statement. Consequently, one may
fairly hypothesise that as drivers’ experience in-
crease so do their understanding of posted signs.
3. Drivers with good understanding of posted signs are
less involved in speed violations compared with
those with less understanding.
(
1993) tested the correlation between amount of reduc-
tion in frequency of sign occurrence and the complete
elimination of sign. He found a strong correlation
between the two, but for the higher experience group of
drivers only.
Otani et al. (1992) found that drivers over 60 years
old indicate higher risk in ignoring the warning signs.

H. Al-Madani / Accident Analysis and Pre6ention 32 (2000) 575–581
577
4
. If comprehension of signs is important for drivers’
safety then drivers with good understanding of
traffic signs should more frequently use their seat
belts compared with those with less understanding.
In other words, safe drivers attend to signs, under-
stand them and wear seat belts.
Analysis of variance, using Scheffe’s technique, was
then performed in order to test the hypotheses. The
procedure evaluated the significance of the difference at
0.05 level, between the score of the drivers in the
various groups.
4
. Data collection
3. Methodology
The questionnaire was distributed to over 6000 driv-
The method adopted here involved questionnaires to
ers; 1400 drivers in Bahrain, 2200 drivers in Qatar and
2400 drivers in Abu Dhabi, to represent the UAE. Over
970 drivers in Bahrain (Al-Madani and Abdul-Ghani,
1995; Al-Madani et al., 1996), 1000 drivers in Qatar
and 850 drivers in the UAE, responded back. A re-
sponse rate of 47% was achieved. This represented over
0.05% of the total number of vehicles registered in each
state. This is equivalent to over one million question-
naires that would be required for the United States
(Economic Commission for Europe, 1994) as per this
response rate. Both the response and the response rate
are thought to be sufficient, as many studies (e.g. Dean,
1981; Steven, 1992) consider samples as high as 500 to
be representative for any large populations if properly
distributed according to a known statistical technique.
Moreover, Hofner (1982), for example, ended up with a
39% response rate when he investigated compliance and
non-compliance with regulatory signs. However,
Schmidt (1982) successfully ended up with 67%.
be filled by the drivers who were selected randomly
based on stratified random sampling technique (Sto-
pher and Meyburg, 1979; Steven, 1992; Ortuzar and
Willumsen, 1996). The stratification was based on occu-
pation. The proportion of the various occupations,
countrywise, was first obtained. Accordingly, the ques-
tionnaires were distributed to the various related insti-
tutions. The various occupations were classified into
twelve groups.
The questionnaire involved short-answer and multi-
ple-choice questions. The short-answer questions were
designed to identify the drivers’ traffic and safety re-
lated characteristics satisfying the above goals and the
multiple-choice questions evaluated drivers’ compre-
hension of posted traffic signs. The questionnaire in-
cluded 28 multiple-choice questions on different posted
signs, printed in colors, both in Arabic and English; 18
on regulatory signs, e.g. speed, weight and height limits;
parking, turning and overtaking prohibitions; and di-
rectional movements, and ten on warning signs, e.g.
staggered junction, road narrowing and splitting, turn-
ings and diversions. The suitability of the questionnaire
was approved by language and educational experts,
before being distributed. The understanding scores were
considered as percentages of signs answered correctly to
the total number of questions in each category. The
scores were statistically tested against drivers’ accident
involvement, years of driving experience, speed cita-
tions received in the past 3 years, frequency of seat belt
usage and experience per accident. The latter fairly
combines the effect of both the driving experience and
the driver’s accident involvement. Therefore, compar-
sion between the drivers can be made on a more
rational bassis. The prameter can be determined by
simply dividing driver’s years of experience over driv-
er’s accident involvement. However, the reciprical of
the parameter is first determined. This to avoid the
presence of infinite values resulting from the division
process of experience by accident involvement for the
drivers with no accidents. Influence of age on accident
and experience was also considered. The investigated
characteristics were subdivided into uniform groups, as
will be discussed later. The average scores representing
drivers’ understanding of both regulatory and warning
signs were then determined for the different groups.
All the necessary data, including the accident in-
volvements, are from drivers self-response to the ques-
tionnaires. The reduced data were analyzed using
statistical software called Statistical Package for Social
Sciences (SPSS, 1996).
5. Results and discussion
Scores representing drivers’ understanding of signs
were statistically tested based on the previously men-
tioned characteristics.
5.1. Posted signs by dri6ing experience
The results (Table 1) showed an increasing trend of
drivers comprehension of various posted signs with the
experience. Drivers in Bahrain and Qatar are compara-
ble in their understanding capabilities towards posted
signs, while those in UAE were slightly better. This
might be because of the fact that drivers in the UAE
are required to pass a written test on various traffic
signs before being awarded a driving license, while
drivers in Qatar and Bahrain are not.
Regardless of the above findings, the hypothesis that
the understanding of posted signs increases with the
experience is rejected, because the score did not increase

578
H. Al-Madani / Accident Analysis and Pre6ention 32 (2000) 575–581
significantly in all the corresponding groups. Neverthe-
less, the hyposesis is still valid when only the two
extreme cases are compared, i.e. those with at least 2
decades of experience with those with at most 5 years of
experience, at 0.0001 level of significance.
between the two groups in any of the two categories of
the signs.
5.2. Posted signs by accident in6ol6ement
Results of drivers’ understanding of posted signs
Because experience and age are correlated highly in
motorized countries, age and experience were also com-
pared in this study, for all the three states. The results
showed that age and years of experience (since licenc-
ing) are highly correlated. The correlation, at 0.01
significance level, varried between 0.56 and 0.72 for the
three states.
An attempt was also made to separate age and
experience effects on comprehension of signs. This was
performed by dividing the drivers in the highest experi-
ence group, those with at least 20 years of experience,
into two groups as follows: those who obtained
their driving licences within 15 years after the
minimum permissible driving age (18 years) and those
who obtained them afterwards. The drivers in the for-
mer group scored 64.5 and 61.1% in the warning and
the regulatory signs, respectively. Those in the latter
group scored 68.9 and 62.4%, respectively.
However, the t-test indicated no significant differences
versus accident involvement of their own faults (Table
) revealed the existence of no significant difference in
2
the drivers’ scores, representing understanding of
posted signs, between any two accident involvement
groups. This means that drivers’ better understanding
of posted regulatory and warning signs has no signifi-
cant influence in reducing their accident involvement
because of their faults. In other words, drivers with
high accident involvement are, statistically, as good as
those with no accident involvement in cognition traffic
signs. Moreover, similar results were observed when
drivers understanding of such signs was tested versus
their total accident involvement, i.e. of their faults and
otherwise, in the three states.
Regardless of the insignificant difference, the scores
from the regulatory signs showed a descending ten-
dency with the accident involvement. Moreover, drivers
involved in at least three accidents scored the lowest
comprehension of both warning and regulatory signs.
Table 1
Drivers’ understanding of posted signs by driving experience
State
Group
1
2
3
4
F-test
Significance level
Groups differed significantly
Years of experience
0–5
93
6–10
123
11–20
20+
UAE
Sample size
334
202
–
–
–
Warning signs (%)
Regulatory signs (%)
Sample size
Warning signs (%)
Regulatory signs (%)
Sample size
57.00
56.80
185
55.30
56.00
61.90
59.40
215
58.10
58.30
66.10
61.80
395
58.00
58.40
309
60.20
64.00
138
61.8
58.6
114
9.302
4.992
–
3.375
1.595
–
0.000
0.002
–
0.018
0.189
–
1&3,4; 2&4
1&4
–
1&4
None
–
Qatar
Bahrain
218
281
Warning signs (%)
Regulatory signs (%)
56.16
58.3
54.47
56.3
58.37
58.6
66.43
59.5
7.2
1.4
0.001
0.230
4&1, 2, 3
None
Table 2
Drivers’ understanding of posted signs by accidents involvement because of drivers’ fault
State
Group
1
2
3
4
F-test
Significance level Groups differed
significantly
Life time accident
involvements
None
One
Two
Three or more
UAE
Sample size
553
216
63
16
–
–
–
Warning signs (%)
Regulatory signs (%)
Sample size
65.17
61.36
569
64.95
60.65
273
65.24 61.87
85.82 57.33
0.126
0.897
–
0.930
0.465
–
None
None
–
Qatar
75
22
Warning signs (%)
Regulatory signs (%)
Bahrain Sample size
Warning signs (%)
Regulatory signs (%)
57.77
57.34
528
58.37
57.9
59.19
58.27
272
56.55
58.5
60.53 58.37
57.56 55.58
0.821
0.918
–
1.4
1.2
0.512
0.453
–
0.226
0.295
None
None
–
None
None
87
57.40 57.08
56.1 59.92
23

H. Al-Madani / Accident Analysis and Pre6ention 32 (2000) 575–581
579
Table 3
Drivers’ understanding of posted signs by speed tickets
State
Group
1
2
3
4
F-test
Significance level Groups differed significantly
Speed citation received (3 years)
0
1–2
250
3–6
84
\6
16
UAE
Sample size
438
–
–
–
Warning signs (%)
Regulatory signs (%)
Sample size
66.76
62.98
673
65.68
59.60
144
60.00 60.63
56.08 45.83 10.478
3.274
0.020
0.000
–
None
1&4, 2&4
–
Qatar
18
5
–
Warning signs (%)
Regulatory signs (%)
Bahrain Sample size
Warning signs (%)
Regulatory signs (%)
59.49
58.27
798
57.72
58.02
57.50
57.48
94
56.70
58.45
61.67 62.00
56.79 56.67
0.687
0.203
–
0.151
0.109
0.559
0.894
–
0.929
0.955
None
None
–
None
None
6
1
55.00 50.00
59.26 50.00
Yet, no significant difference was observed with any
accident group. Nevertheless, drivers in this group may
be considered for safety rehabilitation programs. How-
ever, this does not mean that authorities should ignore
posted signs’ improvements or drivers’ education pro-
grams, because the main aim of the posted signs is to
instruct the drivers. In fact, Ofusu et al. (1988) encour-
aged such programs to improve drivers’ behaviour be-
cause the rapid motorization has not been paralleled by
proper development in drivers’ education. In brief,
comprehension of posted signs has no influence on their
accident involvement.
When lifetime accident involvements are considered,
the age and licence age (experience), are to be given due
consideration because experienced drivers are compared
with less experienced ones in each accident category.
Although accidents caused by drivers’ own fault are
partially because of random variance, the argument is
still true because they increase with time. Therefore to
incorporate the influence of age on understanding of
signs in each accident category, analysis of variance was
carried out at 0.05 level of significance. The test was
performed for the following age groups: (i) less than 24
years; (ii) 25–34 years, (iii) 35–44 years; (iv) 45–54
years, and (v) over 54 years. When drivers with either
one or no accident involvement are considered, the
results indicated that drivers in age groups between 35
and 54 years understand warning and regulatory signs
significantly better than those in younger ages. When
those with only two accidents are considered, drivers
between the age of 35 and 44 understand the sign
significantly better than those between 25 and 34 years.
Finally, when those with three or more accidents are
considered no significant difference between any two
age groups was observed.
comprehension of sign for those with higher accident
involvement. Furthermore, drivers’ comprehension of
posted signs are compared as per their accident involve-
ment considering each age category separately. For
example drivers’ understanding of traffic signs, for
those in age group between 16 and 24 years, is com-
pared for those with 0,1,2 and 3 or more accidents. This
is to identify level of understanding of safe drivers
compared with less safe and risky ones. In order to
enlarge the sample size in the various subcategories, the
data for all the three states were considered as one
group. The results indicated no significant difference
between any two accident groups in any age category.
In other words, risky drivers are as good as less risky
ones in understanding the traffic signs.
5.3. Posted signs by experience per accident ratio
In order to consider the combined effect of both
drivers’ accident involvement and their experience on
their understanding of posted signs, ‘experience per
accident ratio’ is used. This ratio properly takes care of
the cumulative effect of the accidents with years of
experience. It is not fair, for example, to compare the
understanding level of a driver who was involved in a
single accident in 20 years of driving to that with 5
years. Drivers’ comprehension of posted signs were
tested for the drivers with the following years of experi-
ence per accident involvement: under 3, 3–5, 5–7,
7
–10, 10–15 and over 15. The results, considering
drivers total accident involvements and accidents be-
cause of their own faults, indicated no significant differ-
ences between any two groups. However, drivers in the
groups with high experience per accident ratios were
generally better than those in low groups.
As a result of the above findings one can conclude
that middle age drivers, i.e. those between 35 and 54
years, with up to two accident involvements understand
the posted signs significantly better than younger ones.
Age has no significant influence in improving drivers’
5.4. Posted signs by speed citations
Drivers’ scores, representing understanding level,
were tested based on number of speed tickets received

580
H. Al-Madani / Accident Analysis and Pre6ention 32 (2000) 575–581
by the drivers in their latest 3 years of driving. The
results (Table 3), generally, revealed no significant dif-
ferences between the various speed ticket groups. In
other words, all the groups are statistically similar. The
scores in the various posted signs, excluding the warning
signs in Qatar, decreased with the increase in number of
speed tickets received. The results, therefore, reject the
hypothesis stated earlier. Therefore, drivers with good
understanding of signs are not necessarily less involved
in speed citations compared with those with less under-
standing. It is, however, worth mentioning that drivers
with over six speed violations scored low in understand-
ing the regulatory signs, one of which is the speed limit
sign. While only 9% of the drivers in Bahrain do not
know the speed limit signs, in UAE and Qatar the
figures are 19 and 23%, respectively. This might be
because of the higher average years of education per
capita as well as higher literacy level in Bahrain com-
pared with UAE and Qatar. However, further investiga-
tion is still necessary to check both the adequacy of such
5.5. Posted signs by seat belt utilization
Drivers’ understanding of posted signs were tested for
the following seat belt utilization categories: those who
always fasten them, those who sometimes fasten them
and those who occasionally fasten them. Sometimes
putting on seat belt was defined to be less than frequent
(almost always) and greater than occasional (very little).
The results (Table 4), revealed significant differences in
understanding of the posted signs between the various
seat belt utilization groups. It was indicated that as
drivers’ awareness of posted signs increases so does their
seat belt usage. This, therefore, confirms the hypothesis
stated earlier.
6. Conclusions and recommendations
Drivers’ knowledge of posted signs showed significant
improvement with the seat belt usage. However, drivers
better comprehension of signs does not mean less acci-
dent involvement, of their fault. This is true even when
age is incorporated when accidents are considered. Driv-
ing experience, since licensing, showed better under-
standing among only highly experienced drivers
compared with novice ones. Moreover, those with very
good experience per accident ratios did not prove to be
better than those with bad ratios in understanding the
signs. Similarly, those with no speed citations or with
low number of citations were not significantly better
than those with high number of citations.
signs
for
the
problematic
group
of
the
drivers, those with high number of tickets and low
comprehension of signs, and the behavioral aspects of
such drivers. This will be of interest to both the design-
ers and the manufacturers of the coming generation of
signs.
Another point to be mentioned here is that
drivers with high number of speed violations in
Qatar showed quite high understanding of posted
signs, at least towards the warning signs. This may
indicate
a serious irresponsible driving behavior
when compared with the drivers in Bahrain and UAE
where similar rules are applied, but with tougher en-
forcements. Nevertheless, UAE and Bahrain, are com-
parable in police enforcement, yet drivers in UAE
clearly receive higher number of speed citations than
those in Bahrain. This might be attributed to more
affluence in the UAE where the average per
capita income is almost twice that in Bahrain (Al-
Madani and Al-Sada, 1998). Therefore, in addition to
citations, enforcement of more punitive measures are
desirable.
As a result of the above findings, one can not agree
with the commonly used assumption that knowledge of
posted signs certainly improves safety. Therefore safety
campaigners are recommended to be more conservative
in spending on programs related to drivers’ understand-
ing of posted signs in order to improve the safety. Yet,
agencies that have responsibility for drivers’ licensing
and for developing the handbooks should be adequately
supported; because traffic signs fulfill other driving
navigational needs. These agencies should certainly in-
clude some programs to improve the driving behaviours.
Table 4
Drivers’ understanding of posted signs by seat belt usage
State* Group
Seat belt usage
1
2
3
F-test
Significance level
Groups differed significantly
Always
306
Sometimes
314
Occasionally
172
UAE
Sample size
–
–
–
Warning signs (%)
Regulatory signs (%)
Sample size
73.59
65.70
161
62.48
59.38
359
57.33
57.36
430
53.508
20.723
–
0.000
0.000
–
All
1&2,1&3
–
Qatar
Warning signs (%)
Regulatory signs (%)
66.46
63.53
59.53
58.82
54.14
54.88
33.943
26.714
0.000
0.000
All
All
*
Bahrain is not included because the seat belt part was not included when the study first started in Bahrain.

H. Al-Madani / Accident Analysis and Pre6ention 32 (2000) 575–581
581
Hofner, K.J., 1982. Causes of traffic violations. Arbeiten-aus-dam-
verkehrspsychologischen-Institute 19 (6), 47–58.
References
Institute of Transportation Engineers, 1992. Traffic Engineering
Handbook. In: Pline, J.L. (Ed.), Prentice Hall.
Jabbar, A.S., Naqvi, S.A., 1992. A study of road signs. Derasat-Naf-
seyah, Riyadh 2 (3), 79–87.
Johansson, G., Rumar, K., 1966. Drivers and road signs: a prelimi-
nary investigation of the capacity of car drivers to get information
from road signs. Ergonomics 9, 57–62.
Johansson, G., Backlund, F., 1970. Drivers and road signs. Ergonom-
ics 13, 749–759.
Ofusu, J.B., Abouammoh, A.M., Bener, A., 1988. A study of road
traffic accidents in Saudi Arabia. Accident Analysis And Preven-
tion 20, 95–101.
Ortuzar, J.B., Willumsen, L.G., 1996. Modelling Transport. John
Wiley, New York.
Al-Madani, H.M.N., Abdul-Ghani, A., 1995. Characteristics of driv-
ers understanding of posted signs. Proceedings of the Interna-
tional Forum on Road Research, Bangkok, Thailand. Swedish
National Road and Transport Research Institute, Sweden.
Al-Madani, H.M.N., Al-Janahi, A.R., Abdul-Ghani, A., 1996. Mo-
torists Conception of Posted Regulatory and Warning Signs, 29th
ISATA International Symposium on Automotive Technology and
Automation. Florence, Italy.
Al-Madani, H.M.H., Al-Sada, I., 1998. Personal and social character-
istics of the drivers towards posted signs. University of Bahrain
and Swedish Road and Transport Research Institute, Safety on
Roads: an International Conference (SORIC’ 98), Bahrain.
Book, A., Bergstrom, P., 1993. Road-sign utility as judged by drivers
with different amounts of driving experience. Scandinavian Jour-
nal of Psychology 34 (2), 183–192.
Otani, H., Leonard, S.D., Ashford, V.L., Bushore, M., 1992. Age
differences in perception of risk. Perceptual and Motor Skills 74
Chapanis, A., 1994. Hazards associated with three signal words and
four colours on warning signs. Ergonomics 37 (2), 265–275.
Dean, J., 1981. Basic Statistics for Social Research. Macmillan, New
York.
(
2), 587–594.
Pignataro, L., 1973. Traffic Engineering: Theory and Practice. Pren-
tice-Hall, Inc, New Jersey.
Richard, H., Heathington, K.W., 1988. Motorist understanding of
railroad-highway grade crossing traffic control devices and associ-
ated traffic laws. Transportation Research Record 1160, 52–59.
Riegelneier, 1942. Rehabilitating Signs. Proceedings of the ITE.
SPSS, 1996. 6.1 Syntex Reference Guide. SPSS Inc., USA.
Schmidt, L., 1982. Observance and transgression of local speed limits.
Arbeiten-aus-dam-verkehrspsychologischen-Institute 19 (6), 107–
116.
Steven, K., 1992. Sampling. John Wiley and Sons, New York.
Stopher, P.R., Meyburg, A.H., 1979. Survey Sampling and Multivari-
ate Analysis for Social Scientists and Engineers. Lexington Books.
Summala, H., Naatanen, R., 1974. Perception of highway traffic signs
and motivation. Journal of Safety Research 6, 150–154.
US Department of Transportation, 1989. Annual Report on Highway
Safety Improvement Programs, USA.
Dewar, R.E., Ells, J.G., Mundy, G., 1976. Reaction times as an index
of traffic sign reception. Human Factors 18 (4), 381–392.
Drory, A., Shiner, D., 1982. The effects of roadway environment and
fatigue on sign perception. Journal of Safety Research 13 (1),
2
5–32.
Economic Commission for Europe, 1994. Statistics of Road Traffic
Accidents in Europe. United Nations, XXXLX, Geneva.
Ells, J.G., Dewar, R.E., 1979. Rapid comprehension of verbal and
symbolic traffic sign message. Human Factors 21, 161–168.
Fisher, J., 1992. Testing the effects of road traffic signs’ informational
value on driver behavior. Human Factors 34 (2), 231–237.
Galer, M., 1980. An ergonomics approach to the problem of high
vehicles striking low bridges. Applied Ergonomics 11 (1), 43–46.
Hakkinen, S., 1965. Perception of Highway Traffic Signs. Report 1,
Helsinki, TALJA. Meddelar 4, 6–12.
.