FINAL REPORT
Racial Bias in Drivers’ Yielding Behavior at
Crosswalks: Understanding the Effect

NITC-RR-869

October 2017

NITC is a U.S. Department of Transportation
national university transportation center.

 RACIAL BIAS IN DRIVERS’ YIELDING BEHAVIOR AT
CROSSWALKS:
UNDERSTANDING THE EFFECT

Final Report
NITC-RR-869
by
PI: Kimberly Barsamian Kahn
Portland State University
Collaborators:
Jean McMahon, Portland State University
Tara Goddard, Portland State University
Arlie Adkins, University of Arizona

for
National Institute for Transportation and Communities (NITC)
P.O. Box 751
Portland, OR 97207

October 2017

 ACKNOWLEDGEMENTS
This project was funded by the National Institute for Transportation and Communities
(NITC). The authors would like to thank Jaboa Lake, J. Katherine Lee, Maria Villarreal,
and members of the Portland State University GRASP Lab for their assistance with this
project. A special thank you to all who served as pedestrians for these studies, and to
Oregon Walks and Unite Oregon for their help with the focus groups.

DISCLAIMER
The contents of this report reflect the views of the authors, who are solely responsible for
the facts and the accuracy of the material and information presented herein. This
document is disseminated under the sponsorship of the U.S. Department of
Transportation University Transportation Centers Program in the interest of information
exchange. The U.S. Government assumes no liability for the contents or use thereof. The
contents do not necessarily reflect the official views of the U.S. Government. This report
does not constitute a standard, specification, or regulation.

RECOMMENDED CITATION
Kahn, Kimberly, Jean McMahon, Tara Goddard and Arlie Adkins. 2017. Racial Bias in
Drivers’ Yielding Behavior at Crosswalks: Understanding the Effect. NITC-RR-869.
Portland, OR: Transportation Research and Education Center (TREC).

 Technical Report Documentation Page
1. Report No.
NITC-RR-869

2. Government Accession No.

3. Recipient’s Catalog No.

4. Title and Subtitle

5. Report Date
October 2017

Racial Bias in Drivers' Yielding Behavior at Crosswalks: Understanding the Effect
6. Performing Organization Code
7. Author(s)

8. Performing Organization Report No.
869

Kimberly Barsamian Kahn, Jean McMahon and Tara Goddard; Portland State University
Arlie Adkins; University of Arizona
9. Performing Organization Name and Address

10. Work Unit No. (TRAIS)
11. Contract or Grant No.

12. Sponsoring Agency Name and Address

13. Type of Report and Period Covered

National Institute for Transportation and Communities (NITC)
P.O. Box 751
Portland, Oregon 97207
15. Supplementary Notes

14. Sponsoring Agency Code

16. Abstract
This project explores social identity factors (race and gender) that influence drivers’ behavior in interactions with pedestrians at crosswalks.
One dangerous potential point of conflict for pedestrians within the transportation system is interactions with drivers at crosswalks (NHTSA,
2009). In 2010, there was one crash-related pedestrian death every two hours and an injury every eight minutes, and racial minorities are
disproportionately represented in these pedestrian fatalities (CDC, 2013). In light of this disparity, this project examines whether racial
discrimination occurs at crosswalks, which may lead to disparate crossing experiences and disproportionate safety outcomes. Racial minorities
experience racial discrimination across various domains in society. Consistent with this societal pattern, it is hypothesized that drivers will
exhibit racial bias when making decisions about whether or not to stop for pedestrians waiting to cross the street at a crosswalk, which may
reflect conscious or nonconscious biases. Our initial research on this topic revealed the predicted racial bias in drivers’ yielding behavior at
crosswalks: Black male pedestrians were passed by twice as many cars as, and waited 32% longer than, White male pedestrians (Goddard, Kahn
and Adkins, 2015). This study expands on these prior findings to examine the effect of additional pedestrian, driver, and crosswalk
characteristics on drivers’ yielding behavior with pedestrians.

17. Key Words
Social equity, safety

19. Security Classification (of this report)
Unclassified

18. Distribution Statement
No restrictions. Copies available from NITC:
www.nitc-utc.net
20. Security Classification (of this page)
Unclassified

21. No. of Pages
47

22. Price

 TABLE OF CONTENTS
ACKNOWLEDGEMENTS ................................................................................................ 1
DISCLAIMER .................................................................................................................... 1
RECOMMENDED CITATION ......................................................................................... 1
EXECUTIVE SUMMARY ................................................................................................ 1
1.0 INTRODUCTION ........................................................................................................ 3
2.0 FIELD EXPERIMENT ................................................................................................. 5
2.1 PURPOSE ................................................................................................................. 5
2.2 DESIGN .................................................................................................................... 5
2.3 LOCATION .............................................................................................................. 5
2.4 RECRUITMENT ...................................................................................................... 6
2.5 PROCEDURE ........................................................................................................... 7
2.5.1 Timing ................................................................................................................ 7
2.5.2 Trials .................................................................................................................. 7
2.5.3 Recorded Data .................................................................................................... 7
2.6 FINDINGS ................................................................................................................ 8
2.6.1 Phase 1 Results: Unmarked crosswalk .............................................................. 8
2.6.1.1 First car stop................................................................................................ 9
2.6.1.2 Number of cars passing without yielding ................................................. 10
2.6.1.3 Trial length ................................................................................................ 11
2.6.1.4 Driver characteristics ................................................................................ 12
2.6.1.5 Phase 1 Conclusions ................................................................................. 12
2.6.2 Phase 2 Results: Marked Crosswalk ................................................................ 12
2.6.2.1 First car stop.............................................................................................. 13
2.6.2.2 Number of cars passing without yielding ................................................. 14
2.6.2.3 Trial length ................................................................................................ 15
2.6.2.4 Driver characteristics ................................................................................ 16
2.6.2.5 Stop bar ..................................................................................................... 16
2.6.2.6 Phase 2 Conclusions ................................................................................. 17
2.7 FIELD EXPERIMENT CONCLUSIONS .............................................................. 19
3.0 FOCUS GROUPS ....................................................................................................... 21
3.1 PURPOSE ............................................................................................................... 21
3.2 RECRUITMENT .................................................................................................... 21

 3.3 PROCEDURE ......................................................................................................... 22
3.4 FINDINGS .............................................................................................................. 23
3.5 FOCUS GROUP CONCLUSIONS ........................................................................ 26
4.0 OVERALL RECOMMENDATIONS ........................................................................ 27
5.0 LIMITATIONS ........................................................................................................... 28
6.0 FUTURE DIRECTIONS FOR RESEARCH.............................................................. 29
7.0 REFERENCES ........................................................................................................... 31
8.0 APPENDICES ............................................................................................................. A
APPENDIX A: Primary trial coding form ..................................................................... A
APPENDIX B: Timing coding sheet .............................................................................. B
APPENDIX C: First Stop, Number Passed, Stop Bar coding sheet ............................... C
APPENDIX D: Driver characteristics coding sheet ...................................................... D
APPENDIX E: Screening Form for Focus Group .......................................................... E
APPENDIX F: Interview Guide for Focus Groups ........................................................ F

 LIST OF TABLES
Table 2.1: Means, standard deviations, and frequencies for results from Phase 1
(unmarked crosswalk) and Phase 2 (marked crosswalk). ............................................9

LIST OF FIGURES/PHOTOS
Figure 2.1: Marked crosswalk location used during Phase 2...............................................6
Figure 2.2: First car stop (unmarked crosswalk) .................................................................9
Figure 2.3: Cars passing without yiedling (unmarked crosswalk) .....................................10
Figure 2.4: Trial length (unmarked crosswalk)..................................................................10
Figure 2.5: First car stop (marked crosswalk) ...................................................................13
Figure 2.6: Cars passing without yiedling (marked crosswalk) .........................................14
Figure 2.7: Trial length (marked crosswalk)......................................................................15
Figure 2.8: Stop bar (marked crosswalk) ...........................................................................17
Figure 2.9: Oregonian Crossing campaign image..............................................................20

 EXECUTIVE SUMMARY
This project explores social identity factors (race and gender) that influence
drivers’ behavior in interactions with pedestrians at crosswalks. One dangerous potential
point of conflict for pedestrians within the transportation system is interactions with
drivers at crosswalks (NHTSA, 2009). In 2010, there was one crash-related pedestrian
death every two hours and an injury every eight minutes, and racial minorities are
disproportionately represented in these pedestrian fatalities (CDC, 2013). In light of this
disparity, this project examines whether racial discrimination occurs at crosswalks, which
may lead to disparate crossing experiences and disproportionate safety outcomes.
Racial minorities experience racial discrimination across various domains in
society. Consistent with this societal pattern, it is hypothesized that drivers will exhibit
racial bias when making decisions about whether or not to stop for pedestrians waiting to
cross the street at a crosswalk, which may reflect conscious or nonconscious biases. Our
initial research on this topic revealed the predicted racial bias in drivers’ yielding
behavior at crosswalks: Black male pedestrians were passed by twice as many cars as,
and waited 32% longer than, White male pedestrians (Goddard, Kahn and Adkins, 2015).
This study expands on these prior findings to examine the effect of additional
pedestrian, driver, and crosswalk characteristics on drivers’ yielding behavior with
pedestrians. Specifically, this study investigates the roles of 1) pedestrian race, 2)
pedestrian gender, 3) crosswalk design (unmarked intersection crosswalk vs. marked
crosswalk), and 4) drivers’ identity characteristics (male vs. female, White vs. minority)
on drivers’ yielding behavior with pedestrians. A controlled field experiment in which
Black and White male and female pedestrians crossed the street at two different types of
crosswalks (unmarked vs. marked) was conducted, while trained coders marked drivers’
yielding behavior. Results indicated that overall stopping rates were very low at the
unmarked crosswalk, and few differences emerged based on pedestrian race and gender.
When the crosswalk became marked, stopping rates greatly increased; however,
treatment was less equitable. Drivers were less likely to stop for Black and male
pedestrians, and when they did stop, they were more likely to stop closer to Black male
and Black female pedestrians. These effects occurred regardless of drivers’ race and
gender.
In order to better understand African American and Black people’s experiences as
pedestrians, three focus groups were conducted. The results supported the findings of the
field experiment. Overall, African American and Black focus group participants
perceived that drivers treated them differently based on their race by not stopping or
infringing on their space in crosswalks. These negative experiences lead to increased
stress and harms their walking trips.
Elucidating the roles of pedestrians’ race and gender, drivers’ race and gender,
and crosswalk design on drivers’ stopping behavior with pedestrians is an important step
toward developing policies that promote safe transportation experiences. Although
marking the crosswalk increased drivers’ stopping behavior for pedestrians, it also
increased the likelihood of discrimination based on pedestrians’ race and gender. To
reduce this disparity in treatment, it is recommended that marked crosswalks include
1

 additional markings and/or design to reduce the sense that yielding is discretionary and to
increase driver yielding compliance. Stricter enforcement of driver’s yielding at
crosswalks can also help to make drivers stop equally for pedestrians. Increasing overall
stopping rates will help promote more equitable stopping for all pedestrians.

2

 1.0 INTRODUCTION
Pedestrians are most vulnerable when they are crossing the street, as pedestrian
deaths often occur during street crossings (NHTS, 2003). Racial minorities are
overrepresented in pedestrian fatalities, with the fatality rates for Black and Hispanic men
(3.93 and 3.73 per 100,000 population) being twice the rate for White men (1.78), even
when controlling for alcohol use, higher traffic exposure in urban areas, and
socioeconomic status (CDC, 2013). One potential reason for this disparity may be in the
different treatment that racial minority pedestrians experience when trying to cross the
street. That is, drivers’ yielding behaviors may be influenced by a pedestrian’s race,
resulting in different experiences for racial minority pedestrians that may ultimately
affect their safety.
Prior research has demonstrated that drivers do not treat all pedestrians the same.
Drivers’ yielding behavior is influenced by the demographic characteristics of both the
pedestrian and the driver. For instance, drivers of high-status cars are less likely to yield
to pedestrians than those with low-status cars (Piff, Stancato, Côté, Mendoza-Denton and
Keltner, 2012), and are more likely to yield to people who are in their own age group
(Rosenbloom, Nemrodov and Ben Eliyahu, 2006) or disabled (Harrell, 1992.) While
yielding to pedestrians may be the law or the prevailing norm, these findings suggest that
drivers may perceive yielding as a courtesy or at their own discretion, and do so
differentially depending on the type of pedestrian they encounter.
In our initial investigation into this topic (Goddard, Kahn and Adkins, 2015), we
tested whether pedestrian race influenced drivers’ stopping behavior with Black and
White male pedestrians. In a controlled field experiment in which Black and White male
pedestrians waited to cross the street at a marked crosswalk in Portland, Oregon, Black
male pedestrians were passed by twice as many cars and experienced wait times that were
32% longer than those found for White male pedestrians. This behavior may reflect
drivers’ implicit racial attitudes, which are subtle, biased beliefs that individuals hold
beneath their conscious awareness. Pro-white, anti-Black implicit attitudes are held by a
large percentage of Americans and are a cause of discriminatory behaviors (Nosek et al.,
2007). Driving and stopping involve fast-paced decisions with numerous distractions, and
may be perceived as discretionary, which are conditions under which implicit attitudes
are often predictors of behavior (Fazio and Olson, 2003; Greenwald, Poehlman, Uhlmann
and Banaji, 2009). If Black pedestrians experience longer wait times at crosswalks, this
could encourage risky crossing behaviors, which could potentially contribute to the
disparate pedestrian fatality rates. A later study using a similar crossing paradigm with
Black and White women in Las Vegas found that drivers in high-income areas were less
likely to yield to the Black female pedestrian when she was in the roadway, again
highlighting the potential safety issues faced by Black pedestrians trying to crossing the
street (Coughenour et al., 2017).
Based on these findings, we extended this research by examining how pedestrian
race and gender interact to influence drivers’ yielding behavior. In addition to race, this
study also examined female pedestrians, as the inclusion of gender is crucial to
understanding the full range of racial discrimination and is often neglected in studies of
prejudice (Goff and Kahn, 2013). We also examined whether the presence of a marked
crosswalk (with zebra stripes, MUTCD-approved regulatory and advisory signs, and a
3

 stop bar) influenced drivers’ stopping behavior, and whether it does so differentially
depending on the identity characteristics of the pedestrian. Given that both curb
extensions (Johnson, 2005) and signs warning drivers to yield to pedestrians (Van Houten
and Malenfant, 1992) increase compliance, it is possible that their presence reduces bias
in stopping decisions. Alternatively, while signage may help in promoting more stopping,
their nonobligatory nature may leave enough discretion for drivers’ biases to affect their
decisions to stop. Finally, we test whether driver characteristics (gender and race) also
influence stopping decisions. Social psychological theories of prejudice suggest that
White males direct more prejudice and discrimination at racial minority males (Sidanius
and Pratto, 2001), which would imply that White male drivers may be the least likely to
stop for Black male pedestrians. Alternatively, we might find that these biased stopping
decisions occur across driver race and gender identities, suggesting an implicit bias and a
more pervasive problem.
Finally, to better understand the perspective of racial minority pedestrians, focus
groups were conducted to probe African Americans’ experiences as pedestrians at
crosswalks in Portland, Oregon. Focus groups allow racial minorities to describe their
personal experiences crossing the street and how they feel they are treated by drivers. The
focus groups can help reveal whether any objective discrepancies in yielding behavior are
perceived by racial minority pedestrians, and whether that had any effect on their walking
experience or their travel behavior.
To test these hypotheses, two studies were conducted: 1) A controlled field
experiment in which pedestrians’ race and gender, drivers’ race and gender, and
crosswalk design (marked vs. unmarked) were compared to examine drivers’ bias in
yielding behavior, and 2) a series of three focus groups with members of the African
American and Black community in Portland to probe their experiences as pedestrians and
any discrimination they report at crosswalks. We report the results from both studies
below.

4

 2.0 FIELD EXPERIMENT
2.1 PURPOSE
The purpose of this study was to experimentally examine the role of pedestrian race,
pedestrian gender, and crosswalk design on drivers’ yielding behavior toward pedestrians
attempting to cross the street. The field experiment systematically varied pedestrian race
and gender, and the presence or absence of a marked crosswalk, to see what effect these
variables had on drivers’ yielding behavior. Driver demographics (drivers’ race and
gender) were also examined to study their influence on stopping behavior.
2.2 DESIGN
We conducted a field experiment with a between-subjects factorial design, with
pedestrian race (Black vs. White) and gender (male vs. female) as the independent
variables, and drivers’ behaviors as the dependent variables. The experiment was run
twice at the same location: first, at an unmarked crosswalk (Phase 1), and then
approximately nine months later at the same location following the installation of a
crosswalk consisting of “zebra” stripes, crosswalk signage, and a car stop bar (Phase 2).
Trained coders recorded the outcome of each crossing, allowing for examination of
different treatment by pedestrian race and gender.
2.3 LOCATION
The crosswalk that was selected for both phases of the experiment was located at
Southeast 14th Avenue and Belmont Street in Portland, Oregon. This is a two-lane, oneway street with consistent vehicle and pedestrian traffic, near single-family homes and
small businesses. It is a thriving and well maintained business district and neighborhood
near downtown Portland, Oregon. This crosswalk was two blocks downstream from an
intersection with traffic lights and had a new fleet of cars passing approximately once
every three to four minutes.
During Phase 1 of the experiment, the crosswalk at this intersection was unmarked. In
Oregon, all intersections, even if they are unmarked, are legal crosswalks that require
drivers to yield. On April 9, 2016, the City of Portland marked the crosswalk at this
location. The changes included the installation of “zebra” stripes on the pavement,
signage directing drivers to stop for pedestrians ahead, and a stop bar marked on the
pavement approximately 10 feet before the intersection, indicating where cars should stop
(seen in Figure 2.1 below). Phase 2 of data collection occurred more than two months
after the installation of the new crosswalk, in order to allow drivers to grow familiar with
the design change.

5

 Figure 2.1: Marked crosswalk location used during Phase 2
(Source: Google street view)

2.4 RECRUITMENT
For each phase of the experiment, 12 people (three Black men, three White men, three
Black women, and three White women) were recruited to serve as pedestrians in the
study (for a total of 24 pedestrians). These were residents of Portland who served as the
pedestrians crossing the street during the field experiment while the research team
recorded how drivers treated them at crosswalks.
Study pedestrians were recruited on the Portland State University campus, on Craigslist,
and through various local alumni networks. They were all young adults between 18 and
30 years old and received $10 an hour for their participation (for as many hours
completed up to 25 hours total). The average age of the pedestrians in Phase 1 was 24.6
years (Black men: ages 25, 23, and 26; Black women: ages 19, 23, and 25; White women:
ages 25, 28, and 19; White men: ages 28, 30, and 24) and for Phase 2 was 23.0 years
(Black men: ages 25, 25, and 18; Black woman: age 28; White women: ages 24, 22, and
19; White men: ages 27, 23, and 19). Despite small variations in actual ages, all
pedestrians were perceived to be of similar ages across each race and gender group and
drivers would view them as similarly aged when quickly encountering them at
crosswalks. All study pedestrians were “prototypical” in appearance, meaning they were
easily identified as their correct race and gender, and had an average build. They had no
visible tattoos or facial piercings. During data collection, the appearance of each study
pedestrian was standardized by having pedestrians wear identical plain gray T-shirts that
were provided for them and blue denim pants, with their hair down and no visible
jewelry. The standardized neutral appearance of the pedestrians attempts to control for
confounding factors, so that any differences in treatment can be explained by pedestrian
race and gender. Pedestrians were trained in the crossing procedure so that they all
6

 followed the same steps during the field trials (see below in procedure for details on
crossing procedure).
2.5 PROCEDURE
2.5.1 Timing
Phase 1 took place over the span of two weeks in September 2015. Phase 2 took place
over two weeks in June 2016. All trials took place during afternoon sessions between 2
p.m. and 5 p.m. to avoid traffic congestion that would restrict free-flow conditions.
2.5.2 Trials
Each trial consisted of one pedestrian attempting to cross the road at the designated
crossing spot. A member of the research team would signal the beginning of each
crossing trial, which occurred when the first car after a traffic light change crossed a
marked starting point approximately 200 feet upstream of the study crosswalk. Once
signaled, the study pedestrian (one of the 12 pedestrians who were part of the study for
each phase) would approach the edge of the crosswalk. The pedestrian would stand at the
edge of the curb (not in the street), position their body toward oncoming traffic, and look
at the oncoming cars to indicate their intent to cross the street. The trial ended and timing
ceased whenever one of the following happened:
1) A car clearly yielded and stopped, allowing the pedestrian to safely cross.
2) The entire platoon of cars passed the pedestrian without stopping (and the next
platoon was stopped at the upstream stoplight).
Once the trial ended, the pedestrian crossed the street and passed out of sight before
returning to the crosswalk for their next trial. The next pedestrian was cued once the prior
pedestrian had crossed the street and was no longer visible. Pedestrians were given a
randomized order to cross the street, such that pedestrian race and gender were
randomized across trials (e.g., first a White male would cross, then a White female, then a
Black male, etc.). Pedestrians waited for their turn to cross in a vacant lot two blocks
away from the crosswalk, not visible to the drivers on Southeast Belmont Street.
Trials in which nonstudy pedestrians or bicyclists joined the study pedestrian at the
crosswalk were continued but then excluded from final analyses, leaving only trials in
which the designated study pedestrians were alone at the unmarked crosswalk. Trials
were conducted over the span of two weeks during off-peak hours in the afternoon to
avoid overly congested traffic. Data collection was suspended on days with inclement
weather and resumed when conditions returned to mild and clear.
2.5.3 Recorded Data
Five members of the research team coded the drivers’ behaviors at each trial. Members of
the research team (positioned inconspicuously at a nearby restaurant and bus stop with
7

 clear sightlines to the intersection) recorded the following variables using coding sheets
(See Appendices A-D):
1) First car stop: whether the first car to reach the crosswalk stopped for the pedestrian,
recorded as yes or no.
2) Cars passed: if a car did stop, the number of cars that passed the pedestrian before a
car stopped, recorded as a count of cars.
3) Trial length: the amount of time for the pedestrian to cross (in seconds). Timing began
when the pedestrian was signaled to approach the crosswalk curb and ended when a car
clearly yielded or stopped, or the entire fleet of cars had passed.
4) Driver stop characteristics: if a car did stop for the pedestrian, the race (coded as:
white, nonwhite, unknown) and gender (coded as: male, female, unknown) of the driver
who stopped to let the pedestrian pass.
5) Stop bar (Phase 2 only): whether the car that yielded to the pedestrian stopped with its
front bumper over or behind the designated stop bar at the marked crosswalk (coded as
before or after). A car that stopped before the stop bar is obeying the signage and leaving
the pedestrian a wide berth to cross, whereas a car that stopped after the stop bar was
stopping closer to the pedestrian and infringing on the crossing space. Because the stop
bar was positioned further back from the intersection, it is possible that a car could stop
after the stop bar, but still allow enough space for the pedestrian to cross. It is therefore a
measure of how close or far the car stopped, but not necessarily that a car was inside or
outside the marked crosswalk.
2.6 FINDINGS
2.6.1 Phase 1 Results: Unmarked crosswalk
A total of 330 trials was conducted at the unmarked crosswalk during Phase 1. Eleven
trials in which the study pedestrian was joined by a nonstudy pedestrian or cyclist were
eliminated. A total of 319 trials remained (90 White female, 67 Black female, 91 White
male, 71 Black male). The number of trials completed by each individual pedestrian
ranged from 17-35. For each of the following tests, the conventional cutoff of p < .05 was
used as an indicator of statistical significance.
In 60 out of 319 trials (18.8%), one car from the passing platoon stopped for the
pedestrian; thus, in the vast majority of trials, no driver yielded to the pedestrian who was
waiting to cross. The low overall stopping rate may limit the effect of pedestrians’ race
and gender on drivers’ behavior by establishing a floor effect. See Table 2.1 for a
summary of the results from Phase 1.

8

 Table 2.1: Means, standard deviations, and frequencies for results from Phase 1
(unmarked crosswalk) and Phase 2 (marked crosswalk).
White
Men

White
Women

Black
Men

Black
Women

Omnibus
test

6/91

0/90

1/71

2/67

Number of
cars

M=7.11
SD=7.77

M=7.07
SD=4.29

M=8.40
SD=7.40

M=5.82
SD=6.16

F(1, 55) = .51, p=.48

Trial
length

M=26.38
SD=10.03

M=28.79
SD=8.02

M=28.01
SD=10.14

M=28.12
SD=9.04

F(1, 314) = 1.19, p=.28

Phase 2
First car

59/105

76/122

50/108

23/45

Number of
cars

M=1.50
SD=2.24

M=1.73
SD=3.11

M=2.05
SD=3.01

M=1.19
SD=1.43

F(1, 373) = 2.99, p=.09

Trial
length

M=7.78
SD=6.39

M=8.79
SD=8.44

M=8.55
SD=7.49

M=8.32
SD=6.56

F(1, 394) = .56, p=.46

55/105

67/122

31/108

15/46

Phase 1
First car

Stop bar

χ2(3, N=319) = 7.86, p=.05

χ2(3, N=380) = 6.25, p=.10

χ2(3, N=381) = 21.56, p<.001

2.6.1.1 First car stop
Out of the 319 trials, the first car in the platoon to reach the unmarked crosswalk stopped
for the pedestrian in only 9 trials (2.8%), an extremely low first-car stopping rate. The
overall chi-square test determined that there was a difference in the frequency with which
the first car stopped for a pedestrian group, χ2(3, N=319) = 7.86, p=.05, see Figure 2.2;
however, none of the simple effects was significantly different from any other. Despite
being non-significant, the pattern [White women (0% of trials), Black women (3% of
trials), White men (7% of trials), and Black men, (1% of trials)] suggests that White men
may be favored in having the first car stop, although the overall low first stop rates
should be taken into consideration.

9

 First car stop (unmarked crosswalk)
100
90
80

Percent

70
60
50

No

40

Yes

30
20
10
0
White Men

White Women
Black Men
Pedestrian Group

Black Women

Figure 2.2: Percentage of trials in which the first car stopped for the pedestrian during Phase 1 (unmarked
crosswalk) out of 319 total trials.

2.6.1.2 Number of cars passing without yielding
A car stopped in the passing platoon on only 60 of the 319 trials. Of these 60 trials where
a car did stop, a 2 (pedestrian race: Black vs. White) X 2 (pedestrian gender: male vs.
female) ANOVA was used to test how many cars passed the pedestrian before one
yielded; see Figure 2.3. There were no significant main effects of gender [F(1, 55) = .54,
p=.47] and race [F(1, 55) = .00, p=.99], nor were interactions found [F(1, 55) = .51,
p=.48]. These results indicate that no significant differences were found among White
women (M=7.07, SD=4.29), Black women (M=5.82, SD=6.16), White men (M=7.11,
SD=7.77), and Black men, (M=8.40, SD=7.40) in the number of cars that passed them
before one stopped, if there was a car that stopped.

10

 Count

Cars Passing Without Yielding
(unmarked crosswalk)
10
9
8
7
6
5
4
3
2
1
0

White Men

White Women

Black Men

Pedestrian Group

Black Women

Figure 2.3: Average number of cars that passed without yielding to the pedestrian during Phase 1
(unmarked crosswalk) out of 60 total trials in which a car did stop for the pedestrian.

2.6.1.3 Trial length
A 2 (Pedestrian race: Black vs. white) X 2 (Pedestrian gender: male vs. female) ANOVA
was used to examine how long it took for the pedestrian to be able to cross the unmarked
crosswalk; see Figure 2.4. Pedestrians waited between 2.35 and 52.00 seconds to cross,
either because a car yielded or the full platoon of cars passed without stopping (signaling
the end of the trial). Results revealed no significant main effects of gender [F(1, 314) =
1.43, p=.23] or race [F(1, 314) = .20, p=.65], nor was there a significant interaction
between gender and race [F(1, 314) = 1.19, p=.28]. This indicates that there was no
significant difference in average wait time (in seconds) among White women (M=28.79,
SD=8.02), Black women (M=28.12, SD=9.04), White men (M=26.38, SD=10.03), and
Black men, (M=28.01, SD=10.14).

Time (seconds)

Trial Length (unmarked crosswalk)
35
30
25
20
15
10

5
0

White Men

White Women

Black Men

Pedestrian Group

Black Women

Figure 2.4: Length of trial (in seconds) during Phase 1 (unmarked crosswalk) out of 319 total trials.

11

 2.6.1.4 Driver characteristics
It was not logistically possible to code the race and gender of all drivers that passed
pedestrians. We have this information only about the driver that eventually stopped.
Because pedestrians were randomly ordered, drivers should be equally likely to encounter
pedestrians for all groups. If drivers that stop for pedestrians are disproportionately found
for one group of pedestrians compared to others, it suggests that they may be more likely
to stop for this particular group. These results are suggestive, and we recommend they be
interpreted as such.
Of the 60 drivers who stopped, 32 were coded as males, 19 were coded females, and 9
were coded as unknown for their gender (coder could not determine). The observed
coded race of the drivers that stopped was 48 White, 3 nonwhite, and 9 unknown. For the
60 trials in which a car yielded to the pedestrian at the unmarked crosswalk, we examined
whether the drivers’ race and gender was related to the pedestrian group for which they
stopped. That is, did the drivers’ race and gender make it more likely that they stopped
for a particular pedestrian group? Neither driver gender, χ2(2, N=52) = 1.41, p=.70, nor
driver race, χ2(2, N=52) = 1.40, p=. 71, was related to the frequency with which different
race and gender pedestrian groups were yielded to. Given the low number of racial
minority drivers, and the low number of drivers who stopped overall, these results should
be interpreted with caution.
2.6.1.5 Phase 1 Conclusions
The overall non-significant findings from Phase 1 at the unmarked crosswalk indicate
that drivers did not differentiate among Black and White men and women at the
unmarked crosswalk. That is, each gender and race group had similar experiences
attempting to cross at the unmarked crossing. However, this should be interpreted in light
of the overall low stopping rates of trials in which any car yielded in a platoon of multiple
cars (~18%), such that any differences may have been suppressed by a floor effect.
Drivers were unlikely to stop for any pedestrian at the unmarked intersection. While
Oregon law says that any intersection is a crosswalk, the vast majority of drivers did not
yield to the waiting pedestrian at the unmarked intersection, despite their waiting on the
curb with indications of waiting to cross. Further, Oregon law requires drivers to stop if a
pedestrian is in the roadway. Because of safety concerns, the study pedestrians did not
step into the roadway to force compliance, which further allowed drivers to pass the
waiting pedestrian.
2.6.2 Phase 2 Results: Marked Crosswalk
Phase 2 of the field experiment included 12 new study pedestrians (3 from each race and
gender group, ages 18-30, average age 23.0 years) to cross the street at the same
intersection that had received a newly marked crosswalk. As before, the pedestrians were
all young adults and were perceived to be similar in age across groups. They were again
of similar build and clearly identified as members of their race and gender. However, due
12

 to participant attrition, only one of the Black female pedestrians completed trials (n=55).
While data from the Black female pedestrian are included in these analyses (see
Coughenour et al., 2017, for similar field experiments using only one pedestrian per
group), we suggest caution in its interpretation due to the low sample size from one
individual for the Black female group. Because these are different pedestrians in Phase 1
and Phase 2, analyses focus on group differences within each crosswalk type separately,
rather than comparing across the marked and unmarked locations for any given
pedestrian group.
A total of 485 trials were collected in Phase 2 at the marked crosswalk. Pedestrians
crossed in the marked zebra crosswalk. As in Phase 1, trials in which the study pedestrian
was joined by another nonstudy pedestrian or bicyclist at the crosswalk (N=76) were
eliminated. The increase in the number of trials excluded during Phase 2 may be due to
the greater number of pedestrians attempting to cross this intersection once it became a
marked crosswalk. This left a total of 409 trials collected in Phase 2 (129 White female,
110 White male, 115 Black male, 55 Black female). The number of trials completed by
each study pedestrian ranged from 23 to 55.
With the addition of crosswalk markings, a car yielded to a pedestrian in 396 out of 409
trials (96.8%). This indicates that the installed marked crosswalk was very effective in
prompting drivers to stop for pedestrians. Given the higher stopping rates, it is possible
that a ceiling effect might mask changes in drivers’ stopping behavior based on
pedestrian race and gender. See Table 2.1, above, for a complete representation of the
results from Phase 2.
2.6.2.1 First car stop
The first car stopped for the pedestrian in 208 out of 380 trials (55%), demonstrating a
sizeable increase in yielding behavior at the same location once the crosswalk was
marked. A chi-square was used to determine if these initial stopping behaviors were
equitably applied to each type of pedestrian. The test was marginally significant, χ2(3,
N=380) = 6.25, p=.10; see Figure 2.5. Significant simple effects emerged when
pedestrians were categorized by race (collapsing across gender), the White pedestrians
more likely to have the first car stop for them (59% of trials) than the Black pedestrians
(49%), χ2(1, N=378) = 3.87, p=.05. When pedestrians were categorized by gender
(collapsing across race), the female pedestrians were more likely to have the first car stop
for them (61%) than the male pedestrians (51%), χ2(1, N=378) = 4.06, p=.04. No other
significant differences emerged.
Results are consistent if the Black female pedestrian is removed. The chi-square test was
significant, χ2(2, N=354) = 5.93, p=.05, indicating that stopping rates differed by
pedestrian race and gender. Comparing across the three pedestrian groups, simple effects
revealed significant differences between the White female and Black male pedestrians,
such that White women were more likely to have the first car stop for them (62% of
trials) than Black men, who had a first car stop for them in 46% of their trials. When
pedestrians were categorized by gender (collapsing across race), the female pedestrians
13

 were more likely to have the first car stop for them (62%) than the male pedestrians
(47%), χ2(1, N=354) = 6.54, p=.01. When pedestrians were categorized by race
(collapsing across gender), the White pedestrians more likely to have the first car stop for
them (57% of trials) than the Black pedestrians (44%), χ2(1, N=354) = 5.27, p=.02.

Percent

First car stop (marked crosswalk)
100
90
80
70
60
50
40
30
20
10
0

No
Yes

White Men

White Women
Black Men
Pedestrian Group

Black women

Figure 2.5: Percentage of trials in which the first car stopped for the pedestrian during Phase 2 (marked
crosswalk) out of 409 total trials.

2.6.2.2 Number of cars passing without yielding
A car eventually yielded in 396 out of 409 trials (96.8%) at the marked crosswalk. Of the
396 trials when one did yield, a 2 (pedestrian race: Black vs. white) X 2 (pedestrian
gender: male vs. female) ANOVA was used to test how many cars passed the pedestrian
before one yielded; see Figure 2.6. Analyses revealed no significant main effects of
gender [F(1, 373) = .99, p=.32] or race [F(1, 373) = .00, p=.98]. However, a marginally
significant interaction between gender and race was found [F(1, 373) = 2.99, p=.09], with
Black men (M=2.05, SD=3.10) being passed by nearly twice as many cars as the Black
woman (M=1.19, SD=1.43), t(128.59)= -2.30, p=.02. There were no differences in the
number of cars to pass White men (M=1.50, SD=2.24) and White women (M=1.73,
SD=3.11), t(230)= .65, p=.52.
When a one-way ANOVA (White men, White women, Black men) was used to test this
outcome without the Black female pedestrian, there were no significant differences
between pedestrian groups at the omnibus level, F(2, 330) = 1.02, p=.36.

14

 Cars Passing Without Yielding
(marked crosswalk)

Frequency

3

2.5

2

1.5

1

0.5

0

White Men

White Women

Black Men

Pedestrian Group

Black Women

Figure 2.6: Average number of cars that passed without yielding to the pedestrian during Phase 2 (marked
crosswalk) out of 396 total trials.

2.6.2.3 Trial length
A 2 (pedestrian race: Black vs. White) X 2 (pedestrian gender: male vs. female) ANOVA
was used to examine how long it took for a pedestrian to cross at the marked crosswalk;
see Figure 2.7. Pedestrians waited between 1 and 40 seconds to cross at the marked
crosswalk, either because a car yielded or the full platoon of cars passed without stopping
(signaling the end of the trial). Analyses revealed no significant main effects of gender
[F(1, 394) = .21, p=.65] or race [F(1, 394) = .03, p=.86], nor was there a significant
interaction between gender and race [F(1, 394) = .56, p=.46]. There was no difference in
average wait time (in seconds) among White women (M=8.79, SD=8.44), White men
(M=7.74, SD=6.39), Black men, (M=8.55, SD=7.49), and the Black woman (M=8.32,
SD=6.56). This pattern is consistent when the Black female pedestrian is excluded, F(2,
343) = .34, p=.72.

15

 Time (seconds)

Trial Length (marked crosswalk)
10

8
6
4
2
0

White Men

White Women

Black Men

Pedestrian Group

Black Women

Figure 2.7: Average length of trial (in seconds) during Phase 2 (marked crosswalk) out of 409 total trials in
which a car did stop for the pedestrian.

2.6.2.4 Driver characteristics
Of the 396 drivers who stopped for a pedestrian during a trial in Phase 2, 227 were coded
as males, 124 were coded as females, and 45 were coded as unknown for their gender,
while for race, 250 were coded as White, 34 were coded nonwhite, and 112 were coded
as unknown for their race. For these 396 trials, we examined whether the drivers’ race
and gender were related to the pedestrian group for which they stopped. Neither driver
gender, χ2(2, N=309) = .09, p=.96, nor driver race, χ2(2, N=253) = .44, p=.81, was related
to the frequency with which different race and gender pedestrian groups were yielded.
The inclusion of the Black female pedestrian did not affect the driver gender [χ2(3,
N=351) = 1.27, p=.74] or driver race [χ2(3, N=284) = .62, p=.89] outcomes. Because of
the limited diversity in the race of the driver, these results should be interpreted with
caution.
2.6.2.5 Stop bar
The stop bar is meant to provide ample space for pedestrians to cross safely. Because it
was more than 10 feet away from the intersection, it does not mean that a car stopping
past the stop bar was inside the crosswalk. It does indicate that they were stopping closer
to the pedestrian than those that stopped behind it and following the sign. When a car did
yield and its position could be clearly observed and coded (in 381 trials; the position of
the stopped car could not be determined in 14), a chi square test examined whether
drivers were obeying the signage by stopping behind the bar on the street when yielding
to pedestrians. Significant differences in the position of the stopped car emerged; χ2(3,
N=381) = 21.56, p<.001, see Figure 2.8. When the pedestrian was a Black male, the cars
were more likely to stop after the bar (71% of trials) than before the bar. This indicates
that in the majority of trials involving Black men, cars were stopping closer to the
16

 pedestrian. Similarly, the Black female pedestrian was more likely to have a car stop after
the bar (in 67% of trials) than before it, consistent with the treatment of the Black male
pedestrians. However, when the pedestrians were White, the cars were more likely to stop
behind the bar (52% for men, 55% for women) rather than after it, meaning the cars were
farther from the pedestrians. In other words, the stop bar’s intended function of creating
more space between cars and pedestrians worked most often when pedestrians were
White (over half the time), but not when they were Black men (less than a third of the
time). When the data for the Black female pedestrian were excluded, the same pattern of
results remained, χ2(2, N=335) = 18.64, p<.001.

Percent

Stop bar (marked crosswalk)
100
90
80
70
60
50
40
30
20
10
0

After bar
Before bar

White Men White Women Black Men Black Women
Pedestrian Group
Figure 2.8: Percentage of trials in which cars stopped before or after the stop bar during Phase 2 (at the
marked crosswalk) out of 381 total trials in which a car did stop for the pedestrian and its position was
recorded.

2.6.2.6 Phase 2 Conclusions
In Phase 2, the presence of a marked crosswalk led to increased stopping for pedestrians,
but did not always lead to equitable treatment across groups. The marked crosswalk
succeeded in persuading a car in each platoon of traffic to eventually stop for nearly all
pedestrians (in 96.8% of the trials, a car stopped). While there was no significant
difference in the time that pedestrians waited to cross the street, race and gender
differences did emerge for whether the first car stopped, the number of cars that passed
before one yielded, and the positioning of the car that stopped relative to the stop bar.
White and female pedestrians were more likely to have the first car stop for them, while
Black and male pedestrians were more likely to not have the first car stop for them. Black
men were also more likely to have more cars pass them before one stopped. Further,
when they did stop, drivers were more likely to stop with their vehicle behind the stop bar
when the pedestrian was White, but after the bar when the pedestrian was Black,
demonstrating an intrusion into the crossing space for Black pedestrians and a possible
safety risk. These patterns occurred regardless of the race and gender of drivers who
stopped, suggesting a prevalent bias.
17

 18

2.7 FIELD EXPERIMENT CONCLUSIONS
Two field experiments tested whether pedestrians’ race and gender affected drivers’
decision to yield: one when the crossing intersection was unmarked (Phase 1) and one
when the crossing intersection was marked (Phase 2). When the crossing intersection was
unmarked, drivers’ overall compliance in yielding for pedestrians was very low toward
all pedestrians waiting to cross the street. On only 18% of all trials did a car in the
platoon eventually yield to the pedestrian, allowing them to cross the street. Once the
same intersection received markings and additional signs, drivers’ yielding increased
dramatically, to 96.8%, for each platoon of traffic (representing one trial). The increase in
yielding compliance after the installation of crosswalk marking and signs is consistent
with previous research (Zegeer et al., 2005) and professional practice experience. The
magnitude of the change, in which yielding rates increased from 18% to 96.8% of trials,
is significant. This confirms that marking and signing intersections results in higher rates
of driver yielding. However, even at the marked crosswalk, the first car to reach the
pedestrian stopped only 55% of the time, indicating that there was still ample perceived
discretion in drivers’ stopping behavior.
The benefit of increased yielding in the marked crosswalk, however, may not be
equitably distributed. While no pedestrian race and gender differences were found to
affect drivers’ stopping behavior at the unmarked intersection, primarily due to the low
overall stopping rates, differences did emerge at the marked crosswalk. Drivers were less
likely to stop for Black and male pedestrians than for White and female pedestrians.
Furthermore, how drivers yield may be different depending on the pedestrians’ race, such
that they when they do decide to stop, they do so closer to Black pedestrians. This
infringement on the crosswalk space for Black pedestrians can carry with it with potential
safety implications. It is possible that this infringement can also affect crash rates,
playing a part in the disproportionate pedestrian fatality rates for racial minorities.
Further, the intrusion into a pedestrian’s space may negatively affect the pedestrian’s
experience walking, and result in changes in people’s choice to walk or how they cross
the street. We follow up on this theme in the subsequent focus groups to further probe
this issue.
This biased stopping pattern based on pedestrian race is consistent with contemporary
forms of racial bias that are more subtle and nuanced in nature (Dovidio, 2001). Although
it was not possible to directly measure drivers’ psychological attitudes in this study, these
results are in line with other outcomes of implicit bias, which occurs beneath conscious
awareness and is pervasive across society and groups (Greenwald and Banaji, 1995;
Greenwald et al. 2009). Implicit biases are more likely to influence behavior under
conditions in which quick decisions are made, there are distractions and attention might
be diverted (Fazio and Olson, 2003), which are characteristic of the driving task. Drivers
may be unaware that they hold these implicit associations based on race and also be
unaware that they can influence their decisions to stop for pedestrians. Because they are
unconscious, they are also more difficult and require different routes to reduce than
traditional explicit forms of racism or prejudice. Further, people from all groups (e.g.,
19

 race, gender, age) can hold implicit associations due to their exposure to cultural group
stereotypes in society (Greenwald et al., 2009), which is consistent with the study finding
that stopping decisions were not affected by the drivers’ race or gender (although caution
is noted on this finding, given the low number of racial minority drivers in the sample).
Further research is needed to directly test the causal relationship between drivers’
psychological attitudes and their stopping decisions, and pinpoint the role that implicit
and explicit racial beliefs play in this process.
In Oregon, every intersection is a crosswalk, with or without additional markings. The
Oregonian Crossing campaign seeks to educate Oregonians on this issue through media
campaigns; see Figure 2.9 as an example.

Figure 2.9: Educational material from Oregonian Crossing Campaign to promote awareness that every
intersection is a crosswalk.

However, it is clear from Phase 1 of our study that most drivers passing through
Southeast 14th Avenue on Belmont Street did not stop for pedestrians at the unmarked
intersection. Crosswalk markings and signs, however, are more universally understood as
requiring a driver to yield to a pedestrian trying to cross the street, as evidenced by Phase
2 results. Perhaps as a result, the number of pedestrians crossing at this location increased
after the crosswalk was signed and marked (as suggested by the increase in number of
times a nonstudy pedestrian or cyclist joined the waiting study pedestrian during Phase
2). Without controlling for confounding factors like changes in nearby land uses, this
suggests that pedestrians may be aware of drivers’ increased likelihood of stopping if the
crosswalk is marked. Future research could examine whether these were new walking
trips or route choices by intercepting pedestrians after the installation of a new marked
crossing and asking about their travel behaviors before and after the installation.

20

 3.0 FOCUS GROUPS
3.1 PURPOSE
The controlled field experiment provided quantitative evidence of racial and gender
disparities at crosswalks. To complement these findings, three focus groups were
conducted to better understand Black and African Americans’ lived experiences and
perspectives as pedestrians in Portland, focusing on their interactions with drivers and at
crosswalks. We wanted to see if their experiences matched the quantitative results and if
they perceived differential treatment as pedestrians.
Focus groups bring people together to discuss an issue of interest to researchers, making
use of group dynamics to encourage participants to share their perceptions and
experiences around the topic of study (Krueger, 2000). Although not as widely utilized as
surveys in existing transportation research, focus groups can invoke the social
interactions found between roadway users (Musselwhite et al., 2010) in an environment
that is both easy and safe to create and observe in a way that field observations or
experiments are not.
As a qualitative method, focus groups provide the opportunity to collect rich, contextspecific information about the topic, without overly constraining the boundaries based on
initial hypotheses. The qualitative approach of focus groups allows researchers to explore
not only the what of people’s opinions, but also the how and why (Morgan, 1997). In
addition, focus groups present an opportunity to go beyond attitudes and opinions to
include participant experiences and behaviors, albeit self-reported ones (Morgan, 1997,
pg. 20).

3.2 RECRUITMENT
Several recruitment tactics were used to locate participants of the Black and African
American community for these focus groups. Flyers were created and posted at the
Portland State University campus and in various locations around Portland. An email
describing the study was distributed on several community groups and listservs. The
focus groups were co-sponsored by Oregon Walks, a nonprofit advocacy organization,
dedicated to promoting walking for everyone, that aided by distributing the study
announcement to their members.
People who were interested in participating were invited to email the research team.
Upon doing so, they were asked to complete a short screening form (Appendix E). All
respondents who indicated that they were Black/African American, had lived in the
Portland metropolitan area for at least one year, and regularly walked to places in
Portland were invited to participate.
A total of 12 Black and African American people participated across the three focus
groups. Of these 12 participants, 10 self-identified as women, 1 as a man, and 1 as a
21

 trans-man. Their ages ranged from 20 to 46 years old, and averaged approximately 29
years old, with seven in their 20s, four in their 30s, and one in their 40s. They had each
lived in the Portland metropolitan area between 4 and 27 years, with an average around
16 years, and reported walking on average 12 hours per week. Most said they walked to
work or school, with exercise and errands being other common reasons.
3.3 PROCEDURE
Three focus groups were held in April 2017. Two sessions took place on the Portland
State University campus and the third at a community organization’s space in North
Portland. Once participants arrived for a focus group, they were greeted by a research
assistant who asked them to read and sign a consent form and make a nametag. They
were then seated around a large table, with the focus group moderator at the head. Snacks
were provided.
NW CART + Captioning LLC provided on-site, live transcription services. The
transcriber was seated in the corner of the room, in view of participants but out of the
way of the group members. The focus groups were also audio recorded. Participants’
names were changed to numbers for the transcription in order to provide anonymity.
The focus groups began with the moderator introducing herself and the general purpose
of the study. Participants were encouraged to speak candidly about their positive and
negative experiences as pedestrians in Portland, and were assured of their confidentiality.
The interview guide is provided in Appendix F. Focus groups are a purposive group
interaction around a topic of interest, therefore the research questions should drive the
selection of potentially “information-rich” questions (Krueger, 2000). The first question
about where people walk was intended to get participants talking and thinking generally
about walking in Portland. Because the focus group was advertised as being about
experiences of walking in Portland as a person of color, the next question was intended to
hear from participants about that topic – in whatever format and on whatever aspect of
walking they wished to discuss. This set up allows participants to be heard on any topic
before turning to the research-focused questions about issues of driver treatment and
potentially racially-biased yielding. This first research-focused question was open-ended,
in order to achieve the wide end of the interview “funnel” (Morgan, 1997). The next
question asked participants to focus first on positive experiences and then on negative
ones.
As the question funnel narrowed, participants were guided to more specific areas of
research interest, specifically about drivers’ yielding behavior and whether it was
perceived to be racially-biased. The moderator then shared information about previous
findings about drivers’ biased yielding in Portland (Goddard, Kahn, & Adkins, 2015),
and asked participants to respond to this information. Finally, participants were given a
chance to offer final thoughts or opinions more generally, in order to provide additional
qualitative data about which issues people feel most strongly.

22

 Each focus group took place over two hours. Once the focus group had completed,
participants received $30 for compensation.
For the purposes of this report, we focus on participants’ perceptions of the ways in race
impacted their experiences as pedestrians, in order to see if the results from the objective
field study matched their experiences. While a full qualitative analysis is currently
ongoing, we report initial findings most closely related to the field experiment here.
The research team analyzed the focus groups transcripts to explore the extent to which
interactions with drivers were perceived to be influenced by race or bias and looked for
additional recurring themes. Members of the research team read through the transcripts
multiple times and highlighted consistent themes, which were then discussed and agreed
upon as a group. Several additional themes were identified through this process. Each of
these is discussed in detail below.
3.4 FINDINGS
The participants of the three focus groups provided rich information about their
experiences walking around Portland, and specifically interacting with drivers when
trying to cross the street. The results echoed findings from the field experiment.
Black and African American participants from each of the three focus groups not only
reported that their interactions with drivers were perceived to be affected by race, but also
observed that this was made evident by the different treatment of White pedestrians
nearby or walking with the participant. For example, one participant noted that:
“I will come to an intersection on one side and waiting because there’s cars, and
then as soon as a white person steps on the other side, oh, car stops. I’m like, this
is magic, what’s going on? It was you. It wasn’t me.”
Another participant described a situation where walking with a White partner
resulted in more cars stopping:
I haven’t thought about, like, my identities as I’m crossing the street. I—no, I
do, because when I had a white partner they would stop a lot more often than
when I was by myself or with a black person. So, yes, race has something to
do with it. They don’t see us like human beings like they are. They think that
we’re second-class citizens, so when I’m crossing the street I do see a lot
more negative experiences and they don’t want to be inconvenienced or stop
when they are supposed to. That’s why we always get hit and they get upset.
Participants perceived a sense that drivers, particularly White drivers, were
dismissive of their value or right to be in the street as a pedestrian:
I walk everywhere. I take public transportation. I think that there’s this, you
know, understanding of that for drivers especially, like, white drivers, white
23

 male drivers and there’s this kind of subconscious, like, this person isn’t
important, so I can rush them through while they are trying to cross the street
or whatever. You know, I don’t have to be as, like, attentive or whatever. And
it’s probably not something that they actually acknowledge, but it happens all
the time.
Another added:
It’s normally white women who do this and they are like, I’m sorry. If you
were really sorry, you wouldn’t be in the crosswalk in my way.
While these statements were not made by every focus group participant, there did
seem to be widespread agreement, either verbally through additional stories or with
verbal and non-verbal group affirmation. This affirmation manifested itself clearly
when the moderator shared the results of our previous crosswalk study (Goddard et
al., 2015) with focus group participants later in the discussion. Participants agreed
that these results matched their experiences and they were not surprised by the
findings. One of the participants had previously heard of the study and shared the
following reaction:
When we saw the study, my mom was like, duh, why did they have to do that?
It kind of goes—we understand the point of surveys and stuff, but it’s, like,
one of the things it’s, like, you did the surveys, but you could have asked
black people and black people would have told you yes, and you could
believe them because that shit happens.
Several respondents discussed problems with drivers’ stopping in the crosswalk,
which echoed the field experiment findings (Section 2.6.2.5 above) about
infringement into the crossing space. One participant shared that this infringement
causes frustration, stating,
“I think for me, it’s because of how these things intersect with our identities,
mine’s and whoever it is doing it to me. I think the most irritating one is when
cars are, like, in the crosswalk and then like they will like roll down their
window and say, my God, I’m sorry. I’m so sorry. I’m so sorry. And I’m like
looking at them if you were sorry you wouldn’t be rolling down your window,
you would be hitting reverse and backing the fuck up. So, like, you are not
really sorry.”
Another participant described how a driver’s infringement into the crosswalk caused
them to show their frustration, saying
“ I know when cars are in the crosswalk, it pisses me off, like, when you say
you touch their car. I will get close and look at you. Sometimes they say
sorry, and sometimes they will give me a stank look. Like, you are not going

24

 to give me a stank look. You are in the wrong. I agree. You feel like you have
to hurry. I’m a driver and a walker.”
Participants repeatedly noted the feeling that drivers valued their own time or life
more than pedestrians, even to the point of hitting someone in their hurry to get
somewhere. One participant shared a personal experience of being hit, saying
“Especially, like, when you are stopping in the crosswalk, that’s
inconvenience. You were trying to rush across the street. And when I got hit
by a car, they were rushing to go pick up somebody. And it was, like, oh, so
my life wasn’t valuable, and you just hit me. So, yeah, I’m inconvenienced or
disrespected.”
Black and African American participants reported the stress of these racially charged
interactions as similar to other experiences as a racial minority. One person shared
that
“I think it’s stressful. It’s just an added stress to your commute whenever—
even when it’s a leisurely commute. Stuff like that it’s, like, you just have to,
you know, I guess, just part of the black experience of being more aware and
cautious of everything that we do.”
Another indicated how these experiences impact them as pedestrians, causing them
to change their behavior:
“ I always think about it. I’m, like, I’m not going to run across the street
because I feel, like, they want me to get out of the way. I don’t want conflict. I
don’t like conflict. So I, like, submit and run across the street and get out of
their way. I shouldn’t have to do that. I have the right-of-way. I have the
light. It’s my turn to walk. You know the traffic rules.”
Participants did make a point to share that, despite their negative experiences, that
walking in Portland was sometimes positive, too, particularly relative to other places they
had been. One participant stated that
“Even though I had some negative experiences, I would say that, like, Portland
for me is one of the best cities to walk in. Like, I have lived and visited other
places and it’s, you know, despite some of these negative experiences, it’s still
one of the best. You know, even though I’m super-cautious as a woman walking,
at night especially, nothing has happened to me here. Like, or, like, I don’t
know. It’s just like it’s really safe. I used to live in Anchorage, Alaska, not safe
at all for pedestrians. I have family in L.A. and I visit them all the time, not safe
for pedestrians. And, like, Portland is not bad and I think that I, like, mentally
got stuck on some negative thoughts about things I have experienced as a
pedestrian here, but overall, it’s been great.”

25

 3.5 FOCUS GROUP CONCLUSIONS
Conducting focus groups complemented the field experiment by providing richer, more
nuanced details of the experiences of Black and African American people who walk in
Portland. The discussion was highly consistent across all three focus groups, and in line
with the findings of the systematic field experiment examining drivers’ behaviors in a
naturalistic setting. Even though Black and African American participants were first
asked about positive experiences, most of the discussion centered on negative
experiences, of which participants had many. The vast majority of the overall discussion
across all three focus groups was on negative experiences, indicating that these
experiences are salient and frequent. Participants reported a heightened sense of the
difficulties and dangers posed by negative interactions with drivers as pedestrians, which
they also often attributed to their race. Their race was often brought up on their own,
before specific questions about the field experiment. There was also a clear
understanding of how crosswalk markings and pedestrian signals are intended to affect
driver behavior by encouraging stopping, although several participants expressed
skepticism that marking crosswalks will keep drivers from racially-biased behaviors. This
concern is consistent with the findings from Phase 2 of the field experiment, in which
marking the crosswalk increased overall yielding behavior, but did so differentially for
White and Black pedestrians.
Portland’s overall investment in pedestrian infrastructure and pedestrian friendly
environment is noticed and appreciated by the focus group participants. The challenge,
however, is to make these pedestrian experiences equally accessible and safe to all.

26

 4.0 OVERALL RECOMMENDATIONS
As is evident from these study results, marking an intersection increases drivers’
yielding behavior. Therefore, at locations in which the City wants to immediately
increase driver yielding, it is clear that marking and signage can result in significant
change. However, it is also clear that these yielding behaviors are not equally applied to
all pedestrians. Although overall compliance increases when the crosswalk is marked,
drivers still perceive that stopping is discretionary and not required. It is this perceived
discretion that makes drivers feel able to choose to stop for some pedestrians but not for
others. Therefore, in order to reduce this disparate stopping behavior, it is recommended
that increased and additional signage can be employed to help reduce driver discretion
and make stopping register more as mandatory. This additional infrastructure change
might include installing flashing lights at crosswalk locations, additional signals
indicating a crosswalk ahead and sidewalk designs to make pedestrians more visually
prominent (e.g., curb extensions). Our findings also demonstrated that when a driver did
stop, they stopped closer to—past the stop bar for—Black pedestrians. Additional painted
markings on the ground might further highlight the existence of the stop bar, and flashing
or lighted signs can also relate the need to stop behind the bar. In addition to these
infrastructural changes, media awareness campaigns can focus on the need to stop for all
pedestrians. Finally, stricter enforcement of drivers’ yielding behaviors at crosswalks can
also help reduce perceived discretion and promote more equitable yielding. We further
recommend additional focus groups with disadvantaged community members to discuss
these issues and ways to reduce the disparities they encounter. In summary, changes
should focus on reducing drivers’ perceived discretion in stopping behavior at
crosswalks, which will lead to more stopping overall for all pedestrian groups.

27

 5.0 LIMITATIONS
Various limitations to the reported studies are worth noting. First is the fact that,
due to participant attrition, only one Black female completed the trials in Phase 2 of the
field experiment. The other race and gender groups had three pedestrians each, making
their estimates more reliable. As noted earlier, other field experiments have employed
only one pedestrian per group (Coughenour et al., 2017); however, we still suggest
caution in the interpretation. Second, the low number of racial minority drivers in the
study limits the ability to draw strong conclusions about the role of driver characteristics
on yielding behavior. Next, the studies were conducted in Portland, Oregon, and therefore
may be specific to this local context. If implicit biases are leading to these differences in
stopping decisions, the results should generalize beyond Portland, but additional research
in different contexts is needed to establish this point. Next, the focus on these studies has
been on the experiences of Black and African American pedestrians. Additional research
is needed to see if these patterns generalize to other racial minority groups in the United
States, such as Latinos and Asian Americans. We discuss additional future directions for
research to address these limitations and other extensions next.

28

 6.0 FUTURE DIRECTIONS FOR RESEARCH
Results from the field experiment and focus groups indicate that Black pedestrians face
discrimination at marked crosswalks, leading to potentially unsafe crossing experiences.
This research is consistent with prior studies (Coughenour et al., 2017; Goddard et al.,
2015) finding similar racial discrimination against pedestrians. In the systematic field
experiments, examining the effects of crosswalk signs and marking on driver’s yielding
behavior at the same location demonstrated that a) signs and marking significantly
increase driver compliance with crossing rules or expectations, and b) that increase in
compliance may not be equitably distributed. Specifically, drivers were less likely to
yield and more likely to infringe into the space of Black pedestrians. Focus groups of
Black and African American pedestrians largely echoed these same concerns, centering
on negative experiences as pedestrians that were often attributed to their race and
influenced the decisions to walk in the future.
These findings suggest several potential transportation- and psychology-related future
research directions, including:
1. Does racially biased yielding result in greater likelihood of drivers’ hitting,
injuring, or killing pedestrians of color?
2. Does infringement into pedestrian space have negative safety outcomes for
pedestrians of color?
3. Does infringement into pedestrian space or the overall racially biased interactions
between drivers and pedestrians significantly affect people’s choice of whether to
walk, or walking routes?
4. Can additional signs and markings remove the racial effects of driver yielding?
5. Are the qualitative experiences of pedestrians of color in Portland similar to the
experiences of pedestrians of color in other U.S. cities?
6. Can enforcement improve driver yielding without increasing negative effects on
pedestrians of color?
7. Are implicit or explicit biases influencing drivers’ decisions to stop for
pedestrians based on race?
8. Can reducing implicit biases improve drivers’ decisions to stop for all
pedestrians?
9. How do neighborhood location and city demographics influence drivers’ stopping
behavior based on pedestrian race and gender?
10. How do drivers treat other non-Black racial minorities (e.g., Latinos, Asians)?
Future research should address these questions, as well as others, to better understand and
remediate the effects of drivers’ disparate stopping behaviors based on pedestrian race
and gender. It is recommended that future research employs a variety of methodologies,
including naturalistic observation studies, field experiments to assess causality, survey
measures, and in-depth focus groups and interviews with pedestrians to gain deep insight
into their personal experiences. From this combination of methods and research studies,
more targeted interventions can be executed to improve equitable pedestrian experiences
and promote safe walking among city residents.
29

 30

7.0 REFERENCES
Centers for Disease Control and Prevention. 2013. “Motor Vehicle Traffic-Related
Pedestrian Deaths—United States, 2001-2010.” Morbidity and Mortality Weekly Report
62, No. 15.
Coughenour, Courtney, Sheila Clark, Ashok Singh, Eudora Claw, James Abelar and
Joshua Huebner. 2017. “Examining racial bias as a potential factor in pedestrian
crashes.” Accident Analysis & Prevention 98: 96-100. doi:10.1016/j.aap.2016.09.031.
Dovidio, John F. 2001. “On the nature of contemporary prejudice: The third
wave.” Journal of Social Issues 57, No. 4: 829-849.
Fazio, Russell H., and Michael A. Olson. 2003. “Implicit measures in social cognition
research: Their meaning and use.” Annual Review of Psychology 54, No. 1: 297-327.
Goddard, Tara, Kimberly Kahn and Arlie Adkins. 2015. “Racial Bias in Driver Yielding
Behavior at Crosswalks.” Transportation Research Part F: Traffic Psychology and
Behaviour 33: 1-6. doi:10.15760/trec.130.
Goff, Phillip Atiba, and Kimberly Barsamian Kahn. 2013. “How psychological science
impedes intersectional thinking.” Du Bois Review: Social Science Research on Race 10,
No. 02: 365-384.
Greenwald, Anthony G., and Mahzarin R. Banaji. 1995. “Implicit social cognition:
attitudes, self-esteem, and stereotypes.” Psychological Review 102, No. 1: 4-27.
Greenwald, Anthony G., T. Andrew Poehlman, Eric Luis Uhlmann and Mahzarin R.
Banaji. 2009. “Understanding and using the Implicit Association Test: III. Meta-analysis
of predictive validity.” Journal of Personality and Social Psychology 97, No. 1: 17-41.
Harrell, W. Andrew. 1992. “Driver response to a disabled pedestrian using a dangerous
crosswalk.” Journal of Environmental Psychology 12, No. 4: 345-354.
Johnson, Randal S. 2005. Pedestrian safety impacts of curb extensions: a case study. No.
FHWA-OR-DF-06-01.
Krueger, R. A. (2000). Focus groups : a practical guide for applied research (3rd ed.).
Thousand Oaks, Calif: Sage Publications.
Morgan, D. L. (1997). Focus groups as qualitative research (2nd ed.). Thousand Oaks,
Calif: Sage Publications.
National Highway Traffic Safety Administration. 2009. “Traffic safety facts: 2007 data:
pedestrians.” Annals of Emergency Medicine 53, no. 6: 824.

31

 Nosek, Brian A., Frederick L. Smyth, Jeffrey J. Hansen, Thierry Devos, Nicole M.
Lindner, Kate A. Ranganath, Colin Tucker Smith et al. 2007. “Pervasiveness and
correlates of implicit attitudes and stereotypes.” European Review of Social
Psychology 18, no. 1: 36-88.
Piff, Paul K., Daniel M. Stancato, Stéphane Côté, Rodolfo Mendoza-Denton and Dacher
Keltner. 2012 “Higher social class predicts increased unethical behavior.” Proceedings of
the National Academy of Sciences 109, No. 11: 4086-4091.
Rosenbloom, Tova, Dan Nemrodov and Adar Ben Eliyahu. 2006. “Yielding behavior of
Israeli drivers: interaction of age and sex.” Perceptual and Motor Skills 103, No. 2: 387390.
Sidanius, Jim, and Felicia Pratto. 2001. Social dominance: An intergroup theory of social
hierarchy and oppression. Cambridge University Press.
Van Houten, Ron, and Louis Malenfant. 1992. “The influence of signs prompting
motorists to yield before marked crosswalks on motor vehicle-pedestrian conflicts at
crosswalks with flashing amber.” Accident Analysis & Prevention 24, No. 3: 217-225.
Zegeer, Charles V., J. Richard Stewart, Herman H. Huang, Peter A. Lagerwey, John
Feaganes and B. J. Campbell. 2005. Safety effects of marked versus unmarked crosswalks
at uncontrolled locations: Final report and recommended guidelines. No. FHWA-HRT04-100.

32

 8.0 APPENDICES
APPENDIX A: Primary trial coding form
Trial #

Ped #

Ped Gender and
Race
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W
M F B
W

Notes/comments

A

 APPENDIX B: Timing coding sheet
Trial #

Length of trial until car
stopped/yielded (sec)

Trial #

B

Length of trial until car
stopped/yielded (sec)

 APPENDIX C: First Stop, Number Passed, Stop Bar coding sheet
Trial #

First car
stop?
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N
Y N

# of cars passing
w/o stopping

Position of car that stopped
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar
Before stop bar

C

After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar
After stop bar

 APPENDIX D: Driver characteristics coding sheet
Trial # Driver
Gender
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F
M F

U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U

Driver White/
Nonwhite
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW
W NW

Driver Race
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn
Unknwn

D

Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck
Blck

Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc
Ltno/Hspnc

Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian
Asian

Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other
Other

 APPENDIX E: Screening Form for Focus Group

1) Name: ____________________________
2) Email: ____________________________
3) Age:
4) Ethnicity / Race:
5) Gender:
7) Number of years you have lived in Portland:
8) What part of the city (N, NE, NW, SE, SW) do you live in?
9) How much time do you spend walking per week on average? _______
10) Where/why do you usually walk? (example: to work, to the grocery store)

E

 APPENDIX F: Interview Guide for Focus Groups

Introduction: Good evening and welcome to our session. Thanks for taking the time to
join us tonight. My name is [X] and assisting me is [X]. We’re working with researchers
at Portland State University, along with Oregon Walks, who asked us to speak with
people of color about their experiences while walking in Portland. They want to know
about positive and negative experiences you have had, and what barriers you face when
getting around Portland on foot. We are having discussions like this with several groups
in Portland.
There are no wrong answers but rather differing points of view. Please feel free to share
your point of view even if it differs from what others have said. Keep in mind that we’re
just as interested in negative comments as positive comments, and at times the negative
comments are the most helpful. While we can’t control what people say when they leave
this room, we ask everyone to respect the confidentiality of this room and not to share
any personal details about people who have participated tonight. We also ask that you do
not share any stories that we would be required to report to authorities. We ask that your
turn off your phones or pagers. If you cannot and if you must respond to a call, please
step outside while you are on the phone, and rejoin us as quickly as you can.
[Motion to the tape-recorder.] We’re tape-recording the session because we don’t want to
miss any of your comments. People often say very helpful things in these discussions and
we can’t write fast enough to get them all down. We will be on a first-name basis tonight,
and we won’t use any names in our reports. You may be assured of complete
confidentiality. The reports will go back to the researchers to help them understand how
people feel about walking in Portland.
Let’s begin. We’ve placed name cards on the table in front of you to help us remember
each other’s names. Let’s find out some more about each other by going around the table.
Tell us your name and where you live.
1. Opening question [5 minutes]:
a. Where in town do you usually walk? Where do you walk to? Who do you walk
with?
2. I am interested in hearing about what experiences you have had while walking in your
neighborhood, or other places in Portland. First, I would like to hear about positive
experiences, then I’ll ask you to share any negative experiences. [15 minutes]
a. What are some positive experiences you have had while walking in Portland?
b. Have you had any negative experiences while walking in Portland?
3. Now, let’s talk specifically about your interactions interacting with car drivers while
you are walking. Have you ever had any negative experiences with drivers while
walking? [20 minutes]
a. Probes if not mentioned or little discussion:
F

 i. Experience crossing the street
ii. What effect does this have on you?
4. What role do you think your race plays in these experiences as a pedestrian interacting
with drivers? Or some other characteristics like age, gender, etc.? [20 minutes]
a. Probes if not mentioned or little discussion:
i. Do you feel this way in specific neighborhoods? How does this feeling
compare to different neighborhoods? How does Portland compare to other
cities? Why do you think this is so?
ii. Have you ever felt like drivers treat you differently when you are
walking because you are a person of color?
iii. What is your experience crossing the street, specifically?
iv. What effect does this have on you?
5. Researchers recently looked at crosswalks, and black pedestrians had to wait longer
and were passed by more drivers before someone would stop so they could cross the
street compared to White pedestrians. Drivers who did stop were also more likely to
infringe into the Black pedestrians’ space. [10 minutes]
a. How do you feel about this? Does this match your experience?
b. Do you have any ideas about what might be done to change people’s behavior?
Thank you so much for participating today. We have provided a large index card for each
of you. If there is anything else you would like to share that you did not get a chance to
share during this meeting, please feel free to fill out that and leave it for me. We are
happy to hear about anything you think is important regarding your experiences walking
in Portland. [5 minutes]

G

  



NATIONAL INSTITUTE for
TRANSPIORTATION and COMMUNITIES

Transportation Research and Education Center
Portland State University

1900 S.W. Fourth Ave., Suite 175

Portland, OR 97201