Safety-oriented practices of adult bicycle riders in Brooklyn, New York USA: an interview study

Objective : The primary purpose of this investigation was to identify safety-oriented bicycling practices commonly used by adult riders in an urban setting (Brooklyn, New York), and to explore whether there are any differences between the safety-oriented practices of men and women riders. Methods : 24 adult riders (14 men, 10 women) in Brooklyn were interviewed concerning their perceptions of bicycling hazards and their safety-oriented practices. Interviews were recorded, transcribed, and analyzed through thematic analysis. Fisher’s Exact Test was employed to test for gender differences. Results : Participants identified a variety of hazards, mainly due to motor vehicles but also pedestrians and roadway conditions. The analysis distilled twenty-one bicycling practices to summarize prevalent views of the participants about safe riding practices. Related items were grouped under broader categories, generating seven safety-oriented bicycling strategies. Few differences based on gender were found in the analysis; however, women in this study were more likely than men to say that they felt disrespected by other road users. Conclusion : Seven strategies may be important for safe urban bicycling: minimizing exposure to other road users (especially motor vehicles) while riding, being vigilant and anticipating what others might do, riding in a predictable fashion, making one’s presence known to other road users, making sure it is safe before proceeding, obeying traffic rules, and riding at a safe speed. Future studies could develop these concepts further and test whether they are associated with involvement in traffic crashes.

Apparently, however, only a handful of recent studies have given attention to positive bicycling behaviors [6,8,[12][13][14], and only three of them analyzed their association with crash risk [8,12,14]. One study [12] utilized an eight-item index to represent "bicycle safety behaviours", including six desirable practices: helmet use, using bike lights in darkness, signaling when turning, walking the bike across the street at designated crossings, making eye contact with drivers, and assuming that drivers don't see you when you pass them.
Higher scores on this safety behaviors index were associated with lower crash involvement in the previous three years. Another study [8] tested [11][12][13] year old riders on some "higherorder cycling skills", comprising hazard perception, blind spot recognition, gap acceptance, and knowledge of priority (right-of-way) protocols. No association was found between these skills and either bicycling experience or crash history.
The third study, focusing on U.S. bicycle riders, analyzed published reports on traffic accidents and their causes to postulate six on-road "safety-oriented bicycling practices": speed, vigilance, position in traffic, yielding the right-of-way, obeying traffic laws, and checking before changing speed or direction [14]. Principal component analysis led to identification of three general safety-oriented behaviors: vigilance, cautiousness, and predictability. The study did not find evidence that vigilance or cautiousness reduced odds of crashes, but the results suggested a possible protective effect related to predictability.
It is possible that men and women have different views about what it means to ride safely. Previous studies have found women to be more cautious than men [15,16], and women may have greater concerns than men about bicycling in traffic, motorist aggression, inclement weather, cycling skills, and the capability to deal with bicycle maintenance issues on the road [17]. Women have self-reported higher levels of "positive cycling behaviors" and perception of cycling risk, but they also reported higher levels of psychological distress [6].
Although men have self-reported higher levels of risky cycling behaviors than women, men have also demonstrated more knowledge of traffic norms [6], and they scored higher on a measure of cycling competence [16]. Based on observations of bicyclists, men were deemed less likely than women to experience conflicts with motor vehicles at intersections [18].
Results of an observation study [19] suggested that women would rather ride on offroad paths than on roads, whether or not the roads had bicycle lanes. Although both men and women may prefer to bicycle off-road rather than on roads, men have shown greater likelihood than women to ride on roads [17]. Older women may favor bicycle paths more than older men do; among younger riders, men may be likelier to prefer fast, direct routes to their destinations, whereas women may prefer bicycle paths and crossings with traffic signals [15].
The primary purpose of this investigation was to identify safety-oriented bicycling practices commonly used by adult riders in an urban locale, so that future research can test whether these practices are protective against crashes. This study also explored whether men and women have different approaches to riding safely. The information obtained in this study addressed both of these study objectives. Sadly, the need for studies like this one has become much more urgent. The number of bicyclists killed in Brooklyn traffic crashes jumped from two in 2018 to eighteen in 2019, a nine-fold increase [22].

Administration of study activities
This investigation took the form of an interview study. Individuals who self-identified as adult (age 18 or older) experienced bicycle riders in Brooklyn and were conversant in English were eligible to participate in this study. Participants were recruited mainly at bicycle shops (three participants were recruited at other public locations). Recruiting took place in 13 different Brooklyn neighborhoods, to seek a diverse and representative sample of Brooklyn bicycle riders. This researcher approached individuals on location and invited them to participate. Each potential participant was offered an invitation letter, an information sheet explaining the study and their rights as a participant, and a copy of the interview questions.
No payments or other inducements to participate were offered. Individuals who agreed to participate were scheduled for a telephone interview to take place at a later time. A few individuals declined to schedule specific interview times but promised to call; none of these individuals were ultimately interviewed.
To ensure anonymity of participants, no identifying information was solicited. To avoid the need to obtain participants' phone numbers, this researcher gave each participant his phone number so that the participant could initiate the interview call from a private and confidential location at the scheduled time. This researcher's phone had no caller ID feature to reveal or record the caller's phone number.
The recruitment objective was to enlist approximately equal numbers of men and women to participate. The total number of participants was dependent on achieving information saturation even if divergent trends occurred between male and female participants in the information which they provided. The number of interviews required to achieve saturation could not be predicted with confidence in advance. There is a wide range of views among qualitative researchers; the answer seems to depend on several factors [23]. One investigation specifically designed to address this question [24] found that saturation was achieved by the twelfth interview.
Recruiting and interviewing commenced in early July 2018 and continued until mid-November 2018. Recruiting took place mainly on Saturdays or Sundays, when more riders were expected to visit bicycle shops. woman, and one Hispanic man. None of the African-American men who were scheduled for interviews followed up to be interviewed.

Analysis of data
Thematic analysis was utilized to search for patterns and themes in the interview narratives [25,26]. In the initial iteration, in vivo coding was employed to retain as much of the participants' own phrasing as possible [27]. A large number of codes emerged from this first stage of the process; codes were then grouped under more general categories to generate a smaller number of broader themes. All transcripts were then re-coded based on these broader themes.
It was this researcher's view that safety practices which were reported by a larger number of riders would be more likely to gain acceptance and be utilized by the general population of urban bicyclists. Consequently, themes which were retained generally reflected the comments of several participants (often one-third or more). Importantly, the themes represented the most prevalent views among the respondents. The process of coding the interview transcripts was facilitated by the use of the comments feature of  Twenty-four individuals participated in this study, including ten women (42%).

Demographic profile of the participants (Table 2)
Participants ranged in age from 25 to 75; the median age was 37. Four individuals (17%) self-identified as racial minorities: two African-Americans, one Hispanic, and one Filipino

Dangers/hazards to bicycle riders
Dangers/hazards posed by motor vehicles and pedestrians Table 3 lists dangers/hazards posed to bicycle riders by motor vehicles and pedestrians which were cited by participants. Entering or being in bike lanes 6 (4 men, 2 women) Jaywalking 6 (4 men, 2 women) More than half of the respondents expressed particular concern about three dangers posed by motor vehicles: Being doored by someone opening a parked car door (20 participants). Five participants reported that they had been doored themselves. Three others said it had happened to persons they knew.
Turning cars/vehicles (15 participants). Six respondents had themselves been involved in crashes with turning vehicles; one of these respondents had also witnessed such a crash, which he believed had been fatal to the bicyclist involved.
Double-parked vehicles (13 participants). Nine participants commented that doubleparked vehicles forced bicycle riders out into the flow of motorized traffic in order to pass them.
Women were more likely than men to comment explicitly about pedestrian unpredictability. No other gender differences were observed in these results.

Hazardous roadway conditions
Seven participants cited hazardous roadway conditions, especially potholes, as a safety factor (3 men, 4 women).

Dangerous bicycling practices
Several bicycling practices were cited as dangerous by participants in this study (Table 4). Most frequently cited were: wrong-way riding, running red lights without making sure it is safe, speeding, riding while distracted, and zigzagging/weaving through traffic.
Women were more likely than men to name zigzagging/weaving through traffic as a dangerous bicycling practice.

Safety-oriented practices of participants
The study participants responded to several questions about their safety-oriented riding behaviors, in general and also in particular situations. Twenty-one themes emerged from analyzing their responses, comprising various practices which participants advocated or utilized themselves while riding.
• Use bike lanes when they are available.
• Take streets with less traffic.
• In general, avoid taking the lane.
• Take the lane when the situation calls for it.
• Go onto the sidewalk if necessary.
• Maintain space between you and parked cars.
• Actively monitor the situation around you.
• Anticipate what will happen or might happen.
• Be very watchful riding near parked cars; look for indications that someone is inside.
• Avoid distraction from listening to an audio device.
• Avoid using your phone and/or texting while riding.
• Signal your intentions.
• Ride in a consistent or otherwise predictable fashion.
• Communicate your presence audibly.
• Use bike lights.
• Look before you go.
• Don't assume you have the right-of-way.
• Yield right-of-way to pedestrians.
• Adhere to the right-of-way rules.
• Don't ride too fast.
The results are presented in Table 5, including a sampling of individual participants' comments to illustrate how the themes were derived. So I think in certain cases, especially when speeds are slow and everyone's more or less traveling at the same speed, it can be much safer just to make your presence really known, because I think sometimes that's the more risky thing is when people don't realize that you're there, they're kind of (unintelligible) bikers, and you feel like cars will, like, you knowthey'll still pull out from being parked or they'll, you know, try to make a turn into you, not seeing that you're coming up in the bike lane.    Do not be distracted, do not answer your phone or look down to your phone….. And, yeah, just try not to be distracted. I've witnessed a person text-messaging while riding and flipped his bike over because he wasn't aware that there was a lady crossing with a young child in front of him slammed on the brakes, flipped over. (Male)   You still have to have an utmost respect for people that are crossing in that crosswalk. And I think that's a real concern that I see, it's an unfortunate -and I think that's why cyclists really earn a little bit of their -the ire of, you know, the public. I think -I think -you know, the interaction with cars is one thing, but to interact with people that are on the streets walking or, you know, doing other forms of exercise like running -yeah, I think that's a big prob-to me, that's -that's my biggest issue with cyclists and how they interact with pedestrians -not so much how they interact with cars, but how they interact with pedestrians. (Male) Adhere to the right-of-way rules.

Other safety-related considerations
Giving and receiving respect Participants were asked to respond to the concept respect (Table 6). A majority of participants advocated being respectful to all other road users. However, eight participants, including seven women, asserted that bicycle riders or bike lanes don't receive the respect that they deserve; this gender difference was statistically highly significant. I mean, that's a big one for me, I guess. Just, you know respect everyone that's out on the road, everyone has somewhere to go just like yourself and, you know, if you, I feel like as a cyclist, if you are gracious towards cars, that, you know, they'll look out for you or other cyclists in the future, and, instead of, you know -I've seen cyclists, like, smack car doors and things like that, and that really irritates me, because it's -that's not helping anyone, that's just, you know, creating a culture of drivers that hate cyclists. (Male) The main thing is that you've got to have just a level of respect for everybody that's out on the road -that's first and foremost to me I mean -I -I don't -I don't feel it's necessary to impose your will or your agenda on anyone else. I think you could take care of your own business -I think you can take, you know -do what you need to do as a cyclist and not put anybody in, like, you know, a dangerous situation, or even an inconvenient -or a situation where they're just not happy with you. Yeah, respect for -respect for others really is my biggest thing. (Male)

Bicycle riders and bicycle lanes don't get enough respect.
Respondents = 8 (1 man, 7 women; p = .002) Sample Responses: Respect -you know, what I said about that, like: that people don't care. And yeah, I think that I get annoyed sometimes by that. I'm just, like -You can see that I am here, and, like, why are you pretending that I am not there? You know, it's just that -I think that is actually the thing that annoys me the most because sometimes they can put me and them in, like, dangerous situations just because they pretend that I'm not there. (Female) This is a good one, because I feel like there is not much respect for our bike -you know, for riders and bike culture in New York -more and more, but -cars -there needs to be more respect, especially for the infrastructure and bike lanes so we can bike safer. (Female) Helmet use Eleven participants, including eight women (p = .011), advocated helmet use as a safety practice for bicycle riders. While this practice is not directly related to crash avoidance, it is widely recognized to prevent serious head injuries.

Derivation of safety-oriented bicycling strategies
In the urban landscape of Brooklyn, New York, bicycle riders, drivers, and pedestrians co-exist within very tight spaces, many are in a hurry, and some may not be watching out for others. A bicycle rider is often challenged to ride safely among the plethora of hazards described by the participants. Twenty-two specific riding practices meant to reduce crash risk appeared to be prevalent among the participants in this study. Based on these practices, it is possible to postulate a manageable number of safety-oriented bicycling strategies for adult bicycle riders to employ in Brooklyn or other urban settings.
Minimize exposure to other road users, especially motor vehicles, while riding.
Twelve individuals favored using bike lanes when they are available, and seven individuals said that they tried to avoid streets with a lot of traffic. Similar bicyclist route preferences have been reported by other recent studies [29][30][31], and evidence suggests that crash risk may be reduced by using bike lanes [32] or avoiding challenging traffic situations [33][34][35].
These two practices suggest a more general strategy: reduce exposure to other road users, especially motor vehicles. The same objective may be achieved in a number of different ways. Five respondents reported that they sometimes found it expedient to go onto the sidewalk to escape difficult traffic situations (however, riding on the sidewalk is illegal in New York City). Another method was described by a female participant: It must be acknowledged that reducing exposure to other road users is not always straightforward. Perhaps the most difficult issue for a bicycle rider in Brooklyn is where to position oneself in the roadway while riding. Fourteen participants expressed a preference for riding toward the side of the road rather than in the traffic lane (seven individuals worried specifically about drivers becoming angry or impatient due to their presence in the traffic lane). On the other hand, eight respondents advocated taking the lane when the situation calls for it. Because parked cars are present on most Brooklyn streets, riding toward the side of the road may increase crash risk [36]. The possibility of being doored is a frequent danger [37] which was recognized by many respondents. Eleven participants said they try to maintain space between themselves and parked cars when they ride. As one male participant noted, though, avoiding the door zone and the flow of traffic concurrently can be challenging: Anticipating what would or might happen was also cited as a safety practice by fourteen participants. Twelve individuals said they were especially watchful near parked cars (some described clues they looked for to indicate whether someone might be inside). Eleven individuals recognized that listening to music while riding would impair situational awareness, and five respondents cited risks of cell phone use and/or texting while riding. (Electronic device use while riding has been shown to increase crash odds among teen and young adult bicycle riders [33].) Although a previous study [14] did not find an association between vigilance and reduced traffic accident involvement, further research is warranted.
Ride in a predictable fashion. Respondents also seemed to recognize value in a corollary objective: enabling other road users to predict the bicycle rider's actions in traffic.
Predictability as a strategy emanated from two specific categories of participant commentary: using hand signals to indicate intended moves such as turns (9 participants) and riding in a consistent or otherwise predictable fashion (7 participants). Prior research has suggested that predictability may be associated with reduced traffic accident involvement [14]. Research has also shown that, in the absence of signaling by a bicyclist, someone following behind him/her may have difficulty predicting that the bicyclist is about to turn, as well as difficulty predicting the direction of the turn [38].
Make presence known to other road users. Another corollary strategy involves making one's presence known to other road users. Ten respondents reported using a bell or voice to alert others nearby, and ten respondents also reported using bike lights. Making one's presence known can involve a variety of measures, both passive and active. The safety benefits of passive visibility aids such as bike lights and brightly-colored or reflective apparel have been demonstrated by one study [39] although not by others [40][41][42]. No studies were found concerning active measures such as sounding a bell or yelling.
Make sure it is safe before proceeding. This strategy was suggested by participant commentary concerning looking before proceeding and right-of-way. Eleven individuals spoke explicitly about looking before going, especially at intersections. While five individuals advocated adherence to right-of-way rules, thirteen respondents stated that a rider should not assume that he/she has the right-of-way, and eight individuals maintained that bicycle riders should always yield to pedestrians.
Obey traffic rules. Ten respondents spoke in favor of obeying traffic lights. In addition, ten individuals spoke out against wrong-way riding, and six respondents reported that wrong-way riders had posed hazards to them while they were riding. Some studies have shown an association between higher risk of crashes and bicyclists disobeying traffic signals and signs [35,[43][44][45].
Ride at a safe speed. A majority of interviewees (fourteen) advocated not riding too fast. Previous studies have implicated speed as a factor in bicyclist injuries due to crashes [46] and as a factor in more serious bicycle-vehicle conflicts for male riders at urban intersections [18].

Consideration of gender differences in the results
Caution is required when drawing conclusions from the statistical tests of gender differences in the participants' commentaries. If fifteen individuals discussed a particular idea, this does not necessarily mean that the other nine individuals had a different view of it; we might presume that they deemed it less important, but even this may not be true. In the statistical analysis, the tests may have missed some gender differences: with a small number of participants, statistical significance could only be achieved by relatively large differences in the number of comments by women versus men. Conversely, more than 40 individual statistical tests were computed during the analysis. At the 90% confidence level, one of every ten statistical tests could be significant by pure chance.
Across the board, most dangers/hazards and dangerous bicyclist behaviors were equally likely to be identified by either men or women in this sample of participants. Only two issues appeared to be on women's radars more than on men's (p < .10): pedestrian unpredictability and bicyclist zigzagging or weaving through traffic. Since these two significant results arose among many statistical tests, both may be spurious outcomes. Any attempt to explain these apparent gender differences would be speculative.
Overall, men and women in this study also advocated and/or utilized the same safety-oriented bicycling practices. Among all the safety-related practices identified in the analysis, only three practices showed significant gender differences: women were more likely than men to (1) advocate using bike lights, (2) state or imply that looking before you go is an important safety tactic, and (3) advocate wearing a helmet. Although these results may be spurious, they could instead indicate that female bicyclists are more cautious than male bicyclists, as prior studies have suggested [15,16]. However, riding experience may be a confounding factor. In this sample of respondents, the men had generally ridden bicycles in Brooklyn longer than the women had: 10 of the 12 participants above the median for years riding in Brooklyn were men (p = .036). In addition, 10 of the 13 participants who reported riding at least 4-8 hours per week during the previous twelve months were men (p = .095).
The lack of respect reported by women who participated in this study supports an assertion that women have greater concerns about motorist aggression, as a previous study reported [17]. Motorist aggression, especially harassment, should be treated as a very serious public safety issue. One participant offered a disturbing account of being harassed while she rode: I wish people had more (respect). It also sucks being, like, a female bike rider and being, like, cat-called and followed while you're on your bike. Like, that's happened to me lots of times, and, like, I don't know, they just, like -there was one time, like, this car followed me, like, it kept on following me every intersection, and then he said, like, these nasty things to me, and I'm, like, f--k it, I'm just going to go into the intersection so this car won't follow me anymore. Yeah, I feel like it's a whole other level of, like, disrespect to be, like, put in such a position of vulnerability where, like, someone's, you know, saying these things to you and also in a car, and if you, like, say anything back, they could, like, run you over.
So, respect, I hope people have more of it.
Research into motorist harassment of riders based on gender or any other demographic characteristic should be given high priority.

Conclusions
The purpose of employing safety-oriented bicycling practices is to minimize the likelihood of a crash or collision with another road user. This interview study afforded an opportunity to learn about the safety practices of 24 bicycle riders in Brooklyn, New York.
Based on the interviews, seven strategies for safe riding were identified. Now it is important, through additional research, to develop and refine these concepts and to learn whether or not they offer protection against crashes.
Several limitations must be recognized. It was beyond the scope of this study to demonstrate that any particular safety-oriented strategy or practice is associated with reduced traffic crashes. Geographically, this study was limited to a single urban locale; findings from this study may not be transferable to non-urban settings, perhaps not even to other urban settings. The under-representation of minorities and the absence of African-American men, as well as the possible over-representation of riders with college and postgraduate educations, may have restricted our understanding of the bicycle riding experience in Brooklyn and the most appropriate strategies for safe riding. This researcher was male; a female researcher might have been more effective in both recruiting female participants and eliciting information from them in the interviews. Furthermore, there were moments during some interviews when additional follow-up questions could have clarified participant statements which later proved to be unclear or ambiguous. It is also impossible to rule out investigator bias. This researcher tried not to distort or overreach the data; nevertheless, this researcher's personal experiences as a bicycle rider undoubtedly influenced the construct of the interview questions as well as the conceptualization of safety-oriented bicycling practices and strategies during the analysis.