The influence of bike share systems on cycling perception and behavior.

(1)

The Influence of Bike

Share Systems on Cycling

Perception and Behavior

Master’s Thesis

Seth Lowe

June2014

(2)

Many cities around the world are implementing bike sharing systems in an attempt to provide more sustainable transportation options and supplement their existing bicycle culture. Initial results of these programs show promise in achieving desired changes. Bike share systems have proven to appeal to a more diverse set of individuals and have been an effective supplement to existing bicycle cultures. What is unclear is the varying extents to which these systems can influence individuals attitudes and perceptions of bicycling. One of these influential factors seems to be the different bicycles used from system to system. As cities invest in different bike sharing systems it will be important to know how the different bikes used are acting as a barrier or facilitator to bicycling. This study will examine the influence that different bikes in bike sharing systems are having on the perceptions and behavior of cycling and provide insight into the extent this influence can lead to a more diverse bicycle culture.

1.0 Introduction

1.1 Background

As cities around the world continue to grapple with the realities of global warming and increasing automobile congestion, many have begun to place emphasis on bicycling as a viable transportation alternative. The benefits of bicycling have been widely acknowledged for many reasons, among them are individual health benefits (de Hartog et al., 2010)along with a unique ability to address citywide transportation concerns such as traffic congestion and air pollution (European Commission, 1999). Those cities that have attempted this transition have invariably looked to countries like the Netherlands for methods on how to effectively influence the culture of cycling. The Netherlands is held up as an ideal because of their high cycling numbers and also how normal and respected cycling is throughout the country. Cities such as Amsterdam, Eindhoven, and Groningen have cycling rates of over 30% (fietsberaad, 2009).The majority of cyclists in these cities don’t wear helmets or any type of specific cycling clothing or apparel, which shows that cycling is a normal and widely accepted mode of daily transport (Vilamoor et al., 2008 and Goodman, 2014). In addition, most bicycles used in the Netherlands are designed for comfort and daily transportation practicalities. A popular model often used in the Netherlands, for example, is the “stadsfiets.” These bikes are identified by their upright riding position, angled back handlebars, and overall sturdy, utilitarian design (Hembrow, 2010). These features make it convenient for daily cycling needs but also helps convey the image of cycling as normal and safe.

One of the common problems that cities which are attempting to achieve a higher level of cycling currently face is that too many potential cyclists are discouraged from cycling because they perceive it as too risky or sporty (Alred, 2013). The cycling culture in these areas are, in many ways, the opposite of the Netherlands. In the US, for example, total cycling has almost doubled over the last ten years (Swanson et al, 2012) but the overwhelming majority of this growth has come from young, white, males (Pucher, 2011). Similar trends are occurring in other countries like the UK and

Australiathat also have low cycling levels. One of the contributing reasons for this seems to be the influential culture of cycling that has been inadvertently established through transportation planning

(5)

and policies that have favored the automobile over the bicycle (Epperson, 2012). Because policies have favored accommodating the car, cyclists and other street users have been negatively impacted and non-cyclists have been influenced by the images and behavior that these different types of cycling have created (Daley, 2011). Until 2012, for example, cyclists in the US were expected to ride with automobile traffic on city streets, usually with no bicycle infrastructure (Epperson, 2012). As a result, this type of cycling has appealed to small cycling subcultures that prefer this type of cycling behavior. Cyclists in the US today are most likely to be, not only young, white, and male, but also riding a racing style bike, wearing a helmet and usually some type of sporty clothing (Dill, 2012, PBOT, 2011).Research suggests that each of these factors can be, to varying extents, barriers to more widespread bicycling. A morning commute by bicycle in the US more closely resembles a competitive race than people traveling to work. Compare this cycling culture with the

Netherlandswhere most cycling trips are made in normal clothing e.g. jeans or work apparel, very few people wear helmets, and most trips are made on traditional European city bicycles and it is clear to see the difference along with the unique obstacles that these countries face. In cities where these small cycling subcultures have been allowed to define what cycling is, there is increasing evidence that they preventing a more equitable expansion of cycling simply through the way cycling is being perceived by other road users (Colville-Andersen, 2009, Daley, 2011).

Studies have shown that bike sharing systems have been effective in addressing many of these concerns and, by extension, potentially improving or normalizing the way bicycling is perceived (Goodman, 2014). Bike sharing systems in the US have attracted a higher percentage of female cyclists when compared to personal cycling (Szczepanski, 2013). This is significant because female cyclists are often indicative of a mature bicycling culture (Pucher, 2003). Research has also shown that bike sharing users are less likely to wear helmets (Fischer et al, 2012) and also more likely to wear normal, non-sporty, clothing when cycling (Goodman, 2014). These are all examples of how bike sharing programs appeal to a more diverse group of people and have potential to normalize cycling through changing established cycling stereotypes that have previously defined the culture of cycling in these areas. Because bike sharing systems have proven to influence the perceptions and behavior of both cycling apparel and the use of helmets, it might also be influential in helping create a more diverse and accepted bicycle culture in cities with low cycling rates (Goodman 2014). Daley, (2011) has shown that the image and behavior of different types of cyclists can also be influential on how other people perceive cycling.One of the distinguishing characteristics of these different types of cyclists is the type of bicycle used. This previous research has shownthe ability that bike sharing systems have to address the gender, helmet, and apparel issue. Because the type of bike seems to relatively un-researched yet still an important portion of the problem, this study will aim to build upon what is known by investigating the influence that the different types of bikes in various bike sharing systems are having on individuals’ perception and behavior of cycling.

1.2 Central Problem and Objectives

In order to establish an equitable and diverse bicycling culture in cities with low cycling numbers, the attitudes and perceptions of cycling need to change from seeing it as a dangerous and exclusive activity to something safe, that everyone can do. Bike sharing systems are many city’s latest attempts to make these changes and have shown initial promise in normalizing cycling through attracting riders outside of the otherwise dominant cycling subcultures. However, because these programs are still a relatively new phenomenon, much remains unknown about their long-term

(6)

viability or how they will influence cycling behavior and perception. The latest generation of bike sharing programs, for example, have many variations between systems, most notably regarding the different characteristics of the bikes being used. Without an in depth analysis on how different bikes within various bike sharing systems are influencing travelers’ perceptions and behavior of cycling, predictions cannot be made into whether or not the current approach in these areas is promising and will lead to a more diverse cycling culture or will continue to encourage cycling from a limited group of individuals.

To address this problem this study will explore the following research question:

How are the Characteristics of Bicycles in Different Bike Sharing Systems Influencing the Perception and Behavior of Cycling?

Sub RQs:

1- What are the various elements influencing individuals’ attitudes and perceptions towards bicyclingand how are these elements related?

2- What are individuals’ attitudes and perceptions of the different bicycles in bike sharing systems?

3- How are the attitudes and perceptions of differentbike sharing systems influencing personal bicycling behavior?

1.3 Outline of Thesis

The remainder of this thesis will be structured to address each of the above sub-research questions in order. Chapter 2 will explain the overall methodological and research design decisions that were made to address each research question. Chapter 3 will focus on answering the first sub research question through literature and theoretical analysis concluding with the logical formulation of a hypothesis, which will be tested throughout the remainder of the study. Chapter 4 will address the 2nd_{research question after first providing additional details on specific methodological choices used} to answer this question. Chapter 5 will address the 3rd_{sub-research question and Chapter 6 will} conclude by addressing the overall research question.

Attitudes and Perceptions of Bicycling Bicycling Behavior Characteristics of Bike Share Systems

(7)

2.0 Research Methods and Design

2.1 Research Design

To address these research questions this study has followed a multiple case study research design. The introduction mentioned the rising use and implementation of bike sharing systems and how they can be very influential but also that there are many differences from case to case, particularly regarding the different bikes being used. A multiple case study approach has been selected for this study because it will provide the best opportunity to replicate the research approach and identify differences across each case. Yin (2003) stresses the importance of carefully selecting cases because comparisons will be made between each case. A multiple case study design is ideal for this particular study because the differences between cities and bike sharing systems will show how individuals perceive their respective programs and how personal cycling behavior is being influenced.

2.2 Case Study Selection

Because bike sharing systems are becoming increasingly popular around the world there were many options to choose from. The criteria for selecting between these different systems was centered around identifying a representative sample of the commonly used bikes in popular bike sharing systems with the intent of selecting cases with different bicycle styles. Four different systems in four different cities were selected that met this criteria. They also happened to come from among current industry leaders in bike sharing: JCDecaux, Clear Channel, Bicincitta, and Alta Bike Share. Each of these industry leaders use a very similar third generation system which will be detailed below but each use different bicycles with different models and features. The next criteria narrowed the options to systems that have been in operation long enough to assess successes and failures over time, as well as access needed data. Finally, prior to the case study selection an in depth literature review was conducted in order to properly account for all potential variables. The four case studies below were selected In order gain the best insight into the areas which are influencing bicycling behavior.

Bike sharing systems “Bicing” in Barcelona, “Velib” in Paris, “Roma” in Rome and “Melbourne” in Melbourne have been selected as cases. Roma is operated by Bicincitta, who operate bike share systems throughout Italy and Spain. Velib is operated by JCDeucaux, who operate many bike share systems throughout France, but also have systems throughout Europe. Bicing is operated by Clear Channel who also have systems throughout the world, including Stockholm and Mexico City. Melbourne’s bike share system is operated by Alta Planning from Portland, Oregon. They have systems throughout North America including New York City along with a few cities throughout Australia. The goal was to get a broad representation of different bikes to see how attitudes and perceptions vary from case to case. Each bike has different types of frames, storage racks,

handlebars and a general look and feel that might appeal to different types of cyclists and promote different types of bicycling behavior. Each bike will be compared the traditional European city bike or “stadsfiets,” mentioned in the introduction. This has been selected because these bikes are specifically designed for city and utilitarian cycling and are the most commonly used bicycles in cities such as Amsterdam and Copenhagen that are known for their mature bicycling culture. The

distinguishing features of these bikes that are relevant to this study are angled handle bars so users ride more upright, a sturdy women’s frame, and storage or transportation racks, which include front

(8)

baskets and rear wheel storage racks. Each case has different variations on all of these features which will be detailed below. It is important to understand how these differences are being perceived by users and also the extent to which they are influencing a change in personal cycling growth and behavior.

2.2.1 Barcelona’s Bicing System

Barcelona’s bike share program, “Bicing,” started in 2007 and is generally regarded as one of the more successful bike sharing programs in the world. Bicing is operated and maintained by Barcelona City Council and a partnership with Clear Channel Communication, a major advertising agency. A yearly membership is required for usage of the system. This is different than many of the others which allow daily usage rates and, as a result, have a high amount of tourists. To rent a bike users scan an RFID card at the service station for identification. A bike is then unlocked for the user. The bikes are connected to docking stations via a locking mechanism on the front rack. When users are finished they simply hang the bike back on the rack in an available space and the system recognizes the return (Antoniades et al., 2009). The system receives funding through sign up fees as well as on street parking fees from vehicles.

Each Bicing bike is used about 10 times per day on average which equates to approximately 60,000 daily trips (Barcelona City Council Mobility Report, 2010). The city’s 400 stations are strategically placed with an average of 300 meters between

them to optimize coverage. As of 2012 there were 150,000 registered users. The system has been very successful in terms of usage but not necessarily profit. In 2012 the annual

membership fee went from 45 Euro to 97 Euro. This was an attempt by the city to not only help the system become more profitable but also encourage Bicing users to continue cycling, albeit with their own bikes (Munoz, 2012). Many city residents and Bicing users have protested the increase citing a fear of theft if they were to purchase and ride their own bicycle.

Bicing bikes are designed to limit theft and be a lightweight, agile, commuter type bike. One of the distinguishing characteristics of the aggressive bicycle culture previously mentioned, which is impacting other’s attitudes and perceptions of cycling is the type of bike generally used: usually a racing bike with a straight, or zero degree handlebar. The bicycles in the Bicing system, while not racing bikes, do have a zero degree handlebar causing users to ride much more hunched over in comparison to an upright bike with a forty five degree handlebar. The look and appearance of Bicing bikes resemble lightweight folding bikes and have a smaller front wheel than the rear wheel. Bicing bikes weigh 16.8kg which is significantly less than the other models which will be compared, the Velib bike weighs 25.2kg, for example. Bicycles in the Bicing system provide a small, open front rack where users can transport a small bag. This is different than some of the other systems that provide a larger, more secure front basket (Antoniades et al., 2009).

(9)

2.2.2 Paris’ Velib Bike Share System

Velib was introduced in 2007 and has been regarded as successful despite initial problems with theft and vandalism. One unique aspect of this system is that it covers the entire city of Paris and has become an integral part of the public transportation network. The city of Paris already has an efficient public transportation network in place

both locally and regionally for commuting purposes. When Velib was introduced in 2007 it provided another option to fill in the last mile distances that would normally be covered on foot or tram. The density of the bike sharing network makes it another efficient option. Users can sign up for a day, month, or year subscription with a credit/debit card which can be linked with a metro pass allowing users to quickly scan in and out. The system is financed and maintained by the

JCDecaux advertisement corporation who won the bid over Clear Channel. In return they have been given advertisement space in

the form of billboards and other street furniture from the city of Paris (NYC Dept of City Planning, 2010). Usage of the system seems to be high and distributed among many different types of groups. Thirty-Three percent of Velib users come from Parisian suburbs showing that it has been an effective transportation option for daily commuters (IBID). There has also been very high usage from one day passes (2.5 million in the first six months) showing that many short trips are being taken. The introduction of Velib in 2007 has coincided with an overall increase in bicycling in Paris. In 2001 the city began a steady increase in changing both the amount and quality of available bicycle and pedestrian facilities. Since 2009 Velib users represent about 1/3rd_{of all bicycle traffic (paris.fr,} 2009-2013).

The bicycles used in Paris’ bike sharing system, Velib, are extremely sturdy and designed for the harsh abuse that comes with public bike share systems. Each bicycle weighs 22.5kg. Users generally sit more upright from the angled back, forty five degree handlebars. They also provide a large front basket where items can safely and securely be placed while cycling. The bikes are cleverly branded and easily recognizable as a bike sharing bike. All of the bike’s cables and chains are hidden and protected to prevent any potential snags or grease getting on user’s clothing. Velib bikes have an overall look and feel of a state of the art, 3rd generation bike sharing system (Antoniades et al., 2009).

2.2.3 Melbourne’s Bike Sharing System

Melbourne’s bike share system began in 2010 and has struggled to maintain steady usage compared to other successful bike sharing systems. The system is operated by Alta Bike share and allows for annual memberships or short term rentals. The service works similar to the other systems. Users reserve their bikes using a credit card. Short term users are given a temporary code which allows them to then unlock the bike. Annual members are given a key which allows them to quickly scan in and out at each station. Melbourne’s bike sharing system is financed via subsidies from the regional

(10)

government along with sign up and usage fees but has no private sponsor. The city views bike sharing as another mode of public transport similar to the bus or tram services (Phillips, 2010. This is different from some of the other successful systems which often have a private sponsor and

operator.

The bikes used in Melbourne’s system more closely resemble the Velib in that they are heavy and sturdy. An open front rack is provided to users similar to the Bicing system. Handlebars are also angled at forty five degrees. The bikes used in this

system are emerging as a popular model. The major difference with this system is that helmets are required for all trips. Because of the helmet requirement this system has not achieved the same amount of growth and popularity as the other systems. Bicycles are used between 0.3 and 0.8 times per day while the worldwide average is between 2 and 4 per day (Midgely, 2009). The Australian government has responded by

subsidizing cheap helmets, which can be purchased at convenience stores for $5 or are available at docking stations for free based upon

the honor system. Users are expected to ride with the helmet, then leave it for the next user (VIC.gov.au, 2014). Much has been written about the struggles and these results have been a major influence on US systems which have decided to not require helmets.

2.2.4 Rome’s Bicincitta Bike Sharing System

Bike sharing was introduced in Rome in early 2008. Bicincittaoperate bike share systems with partnership of city councils in approximately 21 cities throughout Italy and Spain. Some inspiration has been take from OV Fiets in the Netherlands as systems in many of these cities are meant to coordinate with other public transportation

systems both locally and regionally, giving users as many transportation options as possible (Antoniades et al., 2009). Each bike is anchored to an electric locking device similar to other 3rd_{generation systems. Users must} register online or at the pay station and can access the service with an issued identification card or number. One major difference with this system and the others previously detailed is that users are charged .50 for every half hour used while other systems provide free service for the first thirty minutes

(bikesharing.roma.it, 2013). Since the system was initially introduced in 2008 there have been different

versions of the program. Data on the success and failure of the system including average ridership was not accessible.

Image 2.2.3 SOURCE: Melbournebikeshare.com

(11)

As mentioned above, the bicycles used by Bicincitta, in cities such as Rome, closely resemble traditional European city bikes, or the “stadsfiets” previously referred to. Each steel frame is

designed from a women’s city bike, handlebars are angled, and users are provided with both a large front basket and a rear-wheel rack (Antoniades et al., 2009). The bikes used in this system have been the subject of debate because of their traditional look (IBID). Some feel that bikes used in a bike share system should be easily identified as a 21st_{century system with a look and feel that is unique} and separate from personal cyclists.

2.3 Research Methods

The first sub-research question asked to identify the various elements that influence individuals’

attitudes and perceptions towards bicycling and also how these elements are related. To answer this question a comprehensive literature review was done to map the latest theories and studies that relate to the topic. This included the transportation behavioral theory of Van Acker et. al (2010) which combines theories from social psychology and transport geography. These theories help provide a broad view of transportation behavior and were used as a theoretical foundation for the rest of the study. The decision to use this theory was made because it provides a big picture

understanding of the important variables that influence overall transportation behavior. Because the theory of Van Acker et. al (2010) doesn’t explicitly address bicycling behavior other theories have been included that provide more specific bicycling determinants and variables. The goal of this section of the research was to identify the needed variables which would then be carried over to the other portions of the study. The results of this question will be detailed in chapter 3.

The goal of the second sub-research question was to find out individuals’ attitudes and perceptions of different bicycles from different bike sharing systems. Answering this question was dependent upon the results of the previous sub-research question, which identified the necessary variables that influence bicycling behavior. To answer this question the variables identified from sub-research question number one were used in an online survey to individuals living in the four selected case study cities. A survey was selected as the most effective way of gathering this information because the questions asked needed to identify a mix of individuals’ current transportation habits as well as their attitudes and perceptions of different bicycles and bicycle characteristics. The survey design and administration is based upon the article “Survey Says? A Primer on Web Based Survey Design and Distribution” by Adam Oppenheimer(2011). The web based service “kwiksurvey” was used to distribute and, to some extent, analyze the survey. Respondents came from a variety of online sources including facebook, Twitter, and general online discussion boards and forums. The questions were designed to be general enough to remain applicable to each of the four case studies but specific enough to gain insight into the differences of each system. The desired outcome of this portion of the study is to see how opinions vary according the different features mentioned above. Because each system and city has different characteristics (density, population, handlebars, helmet requirements, bicycle storage, bicycle weight etc.) but the same general goal, different responses were expected from system to system. Because there is a language barrier, each survey was

translated into the appropriate language by a native speaker prior to distribution. The results of the collected survey data was analyzed primarily in Microsoft excel and within the survey distribution service which has some basic built in analysis tools. This was selected as the best analytical tool because of the need to calculate basic averages and results from a relatively wide group of variables.

(12)

The specific details of this survey including questions and the selection of respondents will be provided at the beginning of chapter 4.

The final sub-research question asked: How are the attitudes and perceptions of different bike sharing systems influencing personal bicycling behavior? This question was designed to use the results from the second sub-research question to see if any of the results were having a significant impact on overall bicycling behavior. To answer this question semi structured interviews were conducted with owners and operators of bicycle sales/repair shops in each case study city as well as bicycle planning experts and advocates throughout Western Europe, the United Kingdom, and the United States. Bicycle sales and repair shop owners and operators were selected because they have a unique and perhaps overlooked perspective regarding the bicycle culture in a city along with changing trends and attitudes. Their opinions are valuable in better understanding the influence that bike sharing is having on overall bicycling behavior. Bicycle planning experts and advocates were selected because they also have a unique perspective on bicycling in a city and would be able to add context and insight into the research. All interviews were conducted from a series of initial online questions to respondents with follow up questions and discussions after the initial response. The results of these interviews were analyzed with basic coding principles. Most of the responses were a paragraph or less so it was not necessary to use any specified coding software like Atlas TI. More details of this survey including specific questions and responses as well as the collection of respondents will be detailed at the beginning of chapter 5.

2.4 Conclusion

To address each of the above research questions this study has followed a multiple case study research design. This was selected because of the need to compare results and differences across multiple areas. The case studies that will be used are: “Bicing” in Barcelona, “Velib” in Paris, “Melbourne” in Melbourne, and “Bicincitta” in Rome. These were selected after identifying the criteria and narrowing the results to these four. The first sub-research question will be answered from comprehensive literature review, which is detailed in chapter 3. The second sub-research question will be answered by using the variables identified from the results of the first sub-research question in an online survey to measure individual’s attitudes and perceptions. The details and results of this survey are provided in chapter 4. The final sub-research question will be answered from semi structured interviews with bicycle sales/repair shop owners and operators as well as bicycle planning experts and advocates. The details and results of this will be provided in chapter 5.

3-Identifying Elements That Influence Individuals’ Attitudes and Perceptions

Towards Bicycling

The theoretical structure of this study started with the transportation behavior theory of Van Acker et. al (2010)and has been adapted to this thesis. This was chosen as the starting point and

foundation because of its relationship to the topic and overall approach of this paper. Both are focused on assessing issues that influence transportation behavior. Section 3.1 will detail the study of Van Acker et. al (2010) regarding how attitudes and perceptions combined with spatial and social circumstances can influence transportation behavior. An important aspect of their theory is

understanding that transportation decisions are influenced by a variety of factors, both internal (social psychology) and external (transport geography) to an individual, which all need to be

(13)

takeninto account. Using this as a theoretical foundation, section 3.2 will provide an overview of emerging trends in bicycling behavior around the world and can be seen as a link between section 3.1 and the rest of the theories that follow to show that the theory of Van Acker et al. (2010) also has application for bicycling. Section 3.3 details variables and indicators that are specific to bicycling behavior. Because Van Acker et al. (2010) don’t specifically address bicycling behavior, many variables will be taken from this portion of the theoretical framework and used in the data collection. Section 3.4 details the latest , or “state of the art” on bicycling marketing and imagery theories. This section is important to include because it provides context into why exactly decisions regarding which bicycles to use or design are important and influential. Section 3.5 provides the state of the art in bike sharing studies and theories to show what has been done and eventually where this study will depart from. This chapter, then, is intended to map the latest studies and literature that relate to the overall research topic and also identify all relevant variables that will be used throughout the empirical portion of the data collection. Some of these theories and concepts have already been mentioned or implied in the introduction but will be detailed for clarification. This will lead into section 3.6 which will provide a conclusion of the theoretical framework and lead into the logical formulation of a problem statement and hypothesis. The updated conceptual framework in section 3.7 will build upon the one above but more explicitly identify which variables will be used going forward and their relationship to the overall research questions.

3.1 Transportation behavior theories:

3.1.1 Reasoned Influences:

Many studies have been conducted to gain insight into the relationship between perception, attitude, and behavior. Ajzan and Fishbein’s (1975, 1980) theory of reasoned action attempts to predict behavior based upon an individual’s behavioral intention, which is assessed through

individuals’ attitudes towards a stimulus combined with their subjective norms (IBID). Attitudes are established through the collective formation of multiple perceptions, which are based upon

exposure and interpretation to a particular stimulus (issue, object, person etc.) (Gold, 1980; Van Acker et al., 2000). From here these various, related perceptions surrounding this same stimulus lead to the formation of an overall attitude or behavioral intent towards the object (Van Aker, 2010 Pg.228). This overall behavioral intent also depends upon individuals’ subjective norms which measure the context and social acceptability of the intended behavior. For example, a person may form a favorable attitude towards cycling based upon a television advertisement about the health benefits of cycling. This individual’s attitude may become more favorable towards cycling after seeing other people cycling. This person’s favorable attitude may then be reinforced by peers that feel his/her decision to cycle would be a good idea. This process leads to this individual’s overall behavioral intent.

One of the pivotal, reoccurring factors in these behavioral theories is an individual’s intention to perform a specified behavior. IcekAjzan’s (1991, pg.181)theory of planned behavior follows a general rule that the stronger an individual’s intentions towards a behavior, the more likely they are to engage in this behavior. However, a person’s behavioral intentions also depend to some extent on behavioral control through the individual’s access to the resources needed to perform the intended behavior. An individual, for example, may have every intention (positive attitude and social support) to begin cycling each day but lack the funds to purchase a new bicycle. The relationship between

(14)

intent and actual behavioral control has been extensively studied so,Ajzan’s theory of planned behavior focuses on individual’s perception of behavioral control and its impact on intentions. In the theory of planned behavior an individual may have a favorable attitude towards a stimulus and be supported socially but still feel that they can or cannot act on this intention for a certain reason. A person with more confidence in their abilities to perform the intended behavior will have a better perceived behavioral control and will be more likely to act on their intentions. Reasoned behavior deals directly with theories from social psychology where attitudes and perceptions are formed from different stimuli and eventually lead to an individual’s behavioral intent. Reasoned behavior is an important component for this study for two reasons: First, a cyclist or bike sharing user makes an initial decision to cycle or use the bike sharing service based upon his/her individual formation of a perception from a variety of different stimuli, one of them being the features of the bike being used e.g. the handlebars, storage racks etc., and how it fits withinthis individual’s social circumstances. This cyclist, now having made the initial decision to cycle, is also now an advertisement of sorts to other, non-cyclist or non-bike share users, who also form perceptions of cycling based upon their exposure to these different features and individual circumstances. If this person also has a favorable perception of the stimulus, they will be more likely to begin bicycling. As mentioned in the

introduction, there are many different types of cyclists and features of different bikes that influence individuals’ overall attitude of cycling. Going forward these reasoned influences will be a pivotal factor in understanding how the perceptions of different bicycle features are influencing overall bicycle behavior.

3.1.2 Unreasoned Influences:

The major critique of these behavioral theories is that one has to assume an individual’s behavior comes from rational choice while there may be many other influential factors. Ronis et al’s

(1989)Theory of Repeated Behavior shows that once something becomes a habit “decision making is no longer based upon attitudes and other well-reasoned influences” (Van Acker et al, 2010. Pg.229). Initial behavior is still the result of reasoned influences, however, if this behavior is repeated enough the newly formed habit becomes potentially more influential in the subsequent behavioral decisions. Many individuals have daily or weekly needs such as grocery shopping or commuting to work that require transportation decisions. The decision on how to initially meet these needs depends on the aforementioned formation of an attitude through reasoned influences but after the behavior is repeated the newly formed attitude plays a less significant role and individuals’ seem to plan more around the established routine. Van Acker et al. (2010) explain that “These repeated activities (or habits) establish an initial skeleton schedule, in which well-reasoned decisions related to pre-planned activities as well as impulsive decisions related to events-of-the-day are fit in” (pg. 230). An individual’s overall transportation habits and routines are based upon the ongoing interaction between these reasoned and unreasoned elements. Taken together they form a significant portion of an individual’s transportation decisions but there are also other variables that are influential. 3.1.3 Spatial and Social Influences:

The spatial circumstances of an individual also play an important role in understanding the overall transportation context and are one of the elements that influence reasoned and unreasoned behavior. A single person living in a urban setting might be more likely to walk or cycle to work than a married person with children living in the suburbs. Because of the importance of spatial context,

(15)

the theories of transport geography are also important to understand. A person’s daily transportation decisions can be influenced by both

spatial circumstances and timeconstraints

(Hagerstrand, 1970). Individuals are bound to those activities that are within their spatial-temporal reach and transportation behavior can be predicted and planned based upon these circumstances

(IBID).Hagerstrand’s three dimensional models, displayed in figure 3.1.3 A, show a person’s

movement through space and time. The horizontal plane represents an individual’s spatial

circumstances and the vertical plane their time

constraints. The angled line shows a person’s movement throughout the day ultimately showing that individuals are limited by both time and space. A person travelling in a car from home to school may reach their destination faster than a pedestrian, depending on circumstances. In an assessment of transportation behavior, spatial components are important variables that need to be taken into account.

Van Acker, Van Wee, and Witlox (2010) combine the above theories from social psychology and transportation geography to produce a comprehensive conceptual model and improved overall understanding of transportation behavior (figure 3.1.3 B).An individual’s travel behavior can be taken from the combined theoretical structure which show the importance of mapping the external spatial and social components as well as identifying internal influences and perceptions. This study isn’t meant to follow or test the theory of Van Acker et al, but use it as an initial starting point to build around and identify key variables in order to better understand transportation perception and behavior as it relates specifically to bicycling. In broad terms they have shown the importance of understanding an individual’s attitudes and perceptions toward a stimulus through reasoned and unreasoned influences while also taking both spatial and social components into account. The theories below are meant to outline the studies and literature relating to the topic but will also build upon the factors identified by Van Acker et al. by more explicitly identifying which variables are necessary in order to properly address the research question.

(16)

Figure 3.1.3 B SOURCE: Van Acker et al., 2010

3.2 Bicycling Trends, Facts and Figures

Section 3.1 outlined the various elements that influence overall transportation behavior. Bicycling is becoming an increasingly popular transportation option around the world. Western European countries have generally been leaders in utilizing the bicycle as a viable transportation mode although historically it has been the most popular in countries like the Netherlands, Denmark and, Germany. A 1999 report from the European Union showed that 73% of Europeans feel that bicycles should receive preferential treatment compared to cars and that environmental concerns should be priority issues (EU, 1999). The report also detailed the significant benefits that cycling brings to a city including reductions in noise and congestion, improved air quality and individual health, along with financial and socioeconomic benefits for municipalities (IBID). The majority of Europeans also feel that the effects of vehicle traffic in town have become difficult to deal with and have significantly contributed to the poor air quality in many cities. Many agree that a good solution would be to restrict vehicle access in city centers and promote increased bicycle and pedestrian infrastructure (IBID). These pro-cycling perceptions have correlated with impressive increases in overall cycling rates in cities and countries around the European Union. Similar trends are occurring around the world albeit on a smaller scale in most cases. Bicycling in the US, for example, has almost doubled in the last ten years but still has an overall bicycle mode share of less than 2% (Swanson et al.,

2012).Similar trends are occurring throughout the UK and Australia. The shifting perceptions are a sign of positive growth but most cities and countries have a lot of work to do before bicycling becomes as common as it is in the Netherlands, Denmark, or Germany. These countries also provide examples and established techniques that cities can turn to for inspiration.

3.3 Identifying Determinants of Bicycling Behavior

Because bicycling is an increasingly demanded transportation option it is important to understand the multiple ways in which it is being influenced. Rietveld and Daniel(2004) provide a list of factors

(17)

that indicate a person’s likelihood to cycle. Many of the variables used in their study agree with some of the more general indicators previously outlined by Van Acker et al. (2010). Both studies, for example explain the need to understand a city’s spatial and social circumstances. The difference being that Rietveld and Daniel (2004) looked specifically at bicycling determinants while Van Acker et al. (2010) more broadly studied overall transportation behavior. Both studies show that

transportation behavior depends on a multitude of factors and can vary from case to case.Reasoned and unreasoned influences have already been mentioned as variables and will be used in the data collection going forward. The two remaining variables outlined by Van Acker et al., (2010) which are and individual’s spatial and socialinfluences, will be further detailed by Rietveld and Daniel (2004). Figure 3.3 below details the additional variables that need to be taken into account to understand bicycling behavior. Some variables have been added which are not mentioned by either study because they relate specifically to the issue of bike sharing. Each sub-section below will detail the specific variables.

Figure 3.3 Source Rietvelt and Daniel, 2004 3.3.1 Spatial Environment

3.3.1.1 Bicycle Infrastructure: One of the most effective ways to increase cycling is by improving bicycling infrastructure. A European Union study in 1999 found that access to cycling infrastructure was the best way to encourage bicycle sales and usage (EU Cycling Commission, 1999). Major cycling cities like Amsterdam and Copenhagen have shown their commitment to cycling by constant bicycle infrastructure maintenance and expansion, as well as established policies that give preference to the bicycle. This has been one of the many important aspects in creating a mature bicycle network. Several studies have been conducted that show a connection between cycling levels and both demographic and geographic variables like topography, weather, average income, and public policy (Pucher, 2003, Reitveld, 2004). Studies have also shown significant correlations between the amount of available bicycle infrastructure and overall bicycling levels (Dill, 2003). Cities that invest in bicycle infrastructure generally see increase in bicycle usage. This is particularly noticeable among female

(18)

cyclists who are more likely to cycle out of their way to access some form of bicycle infrastructure or traffic calmed street (Dill,2009). Studies by the US Department of Transportation from the 1970s and 1980s showed that bicycle lanes and paths have a significant positive impact on cyclists’ perception of safety (US DOT, 2010). Implementing bike lanes and paths has always been a source of contention in countries like the UK and US where the automobile is the dominant mode of transportation. Recent trends and increased financing in the both countries have led to growth in available bicycle infrastructure. Federal policies in the US, for example, have set aside funding specifically reserved for bicycle and pedestrian projects for local governments (Swanson et al, 2012). This increased

commitment and involvement from government leaders is an important aspect in pursuing bicycle transportation growth.

3.3.1.2 Physical Features: Rietveld and Daniel (2004) break this category into two variables: topography, and the city’s meteorological conditions. The impact of climate and topography has been debated and studied. Countries such as the Netherlands, Denmark, and Germany have high cycling rates and a relatively flat topography but also have varying, sometimes high, amounts of precipitation. Crouse (2010) suggests that both factors might be more perceived than actual barriers by citing examples such as Barcelona which is hilly and hot but still has a relatively high percentage of cycling. Crouse concludes that “Perhaps, instead of seeing topography and climate as a barrier it should be seen as a stimulator of bicycle use, that is, the more favorable the topographic and climatic conditions the more chance there is to have people riding a bicycle (pg.20). These variables will be addressed in each case study in order to take all necessary factors into consideration. 3.3.2 Social Environment (Individual and Sociocultural Factors)

Rietveld and Daniel (2004) also note that cycling depends on individual factors such as age, gender, and an individual’s activity patterns. Understanding activity patterns and habits is similar to the unreasoned influences outlined by Van Acker et al. (2010) and can be identified in the context of this study by understanding which transportation modes individuals use and which they would prefer to use. The sociocultural element deals with how cycling is accepted by the city and can be identified, again, by public commitment through policies but also a city’s overall cycling culture. Examples from Australia that will be detailed below show that one type of cycling (racing, leisure) is acceptable while cycling for daily utilitarian purposes is seen as dangerous and risk taking. The way cycling is socially accepted in Australia has an impact on how individuals perceive cycling. There are many other costs associated with transportation that should be taken into account. Individuals need to consider a wide variety of factors before they decide to cycle. Costs of purchasing a new cycle and lock as well as the potential cost of parking need to be factored in. Rietveld and Daniel (2004) also mention the risk of theft or injury. There is also generalized costs of other transportation modes, these include many factors such as a fuel and parking. These variables will not be specifically addressed but are indirectly answered through the analysis of each city’s commitment to cycling as well as the individual factors listed above.

3.3.2.1 Bicycle Plan: A key indicator in the overall amount of bicycle infrastructure in a city can be assumed by the commitment to cycling shown by city leaders. John Pucher (2008) mentions that the high level of cycling in the Netherlands, Germany, and Denmark can be largely attributed to a variety of initiatives that public officials have done to influence bicycling behavior. City and State

(19)

enforcement as well as overall guidelines and examples. A key determinant in bicycle use is whether or not a city has a comprehensive bicycle plan (OECD, 2004).One example of this commitment can be found in the Netherlands’ integration of public transportation and cycling. Abundant bicycle parking is provided at regional and local transportation hubs. Public bikes (OV Fiets) are also available at regional stations allowing users to take a train from city to city and utilize a public bike for local transportation. This model has been praised in the bicycle planning community and even held up as an ideal example of the way bike share programs should be run (Colville-Andersen, 2013). 3.3.2.2 Helmet Policy: Another way bicycling can be influenced by public policy is through helmet laws. This is not an issue mentioned by Rietveld and Daniel (2004) but will be an important variable to take into account. Studies have shown that helmets can be effective in reducing head and brain injury rates in the event of an accident (Thompson et. al, 1989).This has led to mandatory bicycle helmet laws in a several cities including many in Australia, Canada, and the US. Studies have also shown that both pedestrians and drivers face a greater risk of serious head injury when compared to cyclists (Javouhey et al, 2006 and Klauber, 1978). D.L Robinson (1996) suggests that a mandatory helmet law for motor vehicle users similar to the ones for bicyclists would prevent seventeen times more cases of serious head injury or death. A study from the University of Bath showed that helmeted cyclists are more likely to be struck by passing automobiles than cyclists not wearing helmets. Among the reasons the authors suggest for these findings are that helmeted cyclists may be perceived as safer or better protected in the event of a collision, and also that they are potentially perceived as better, more skilled cyclists (Walker, 2006). While helmets may help in the event of an accident they may also put individuals at greater risk of being in an accident to begin with. These studies align with the opinions of bicycling experts and advocates interviewed for this thesis who overwhelmingly felt that mandatory helmet laws were ultimately a major deterrent to mainstream cycling and overall cycling safety. Mandatory helmet laws are slowly fading away but still exist in places such as Australia and Vancouver, British Columbia. The helmet debate was significant when cities like New York City and Washington D.C. began implementing their bike share systems without a helmet requirement. Results from each of these systems have been positive and the argument has more or less faded away.

Mandatory helmet laws also have an impact on the image and perception of cycling. The first year that mandatory helmet laws were introduced in Australia, helmet usage rose from 31% to 76% (Robinson, 1996). However, multiple follow up studies concluded that the greatest impact

mandatory helmet laws had in Australia was decreasing the amount of overall cycling, not increasing the amount of helmeted cyclists (Robinson, 1996). There are multiple theories for why helmets have been such a barrier for cycling, most of them have the perception of helmets or helmeted cyclists as a central concern. One of the common reasons people give for not cycling is because they perceive it as too risky. A study of adults in the UK found that 86% felt cycling was the most risky mode of transportation (Thornton, 2010 and Goodman, 2014). Alred (2012) and Steinbach et al., (2011) have shown that potential cyclists might be also be deterred by the perception that cycling is exclusively for “sporty” type people, and identity they may not want for themselves. The presence of existing cyclists wearing safety clothing and sporty cycling apparel has shown to discourage non cyclists to take up cycling because of the perceived effort involved in always remembering a helmet, changing at work, etc. (Green et al., 2012). Because there is a perceived need to cycle with a helmet and specific cycling apparel, people riding a bicycle in a city have been unintentionally branded or type cast as a “cyclist” to an extent that is not true with any other transportation mode (Goodman, 2014,

(20)

Alred, 2012). Compare this with riding a bicycle in a city like Amsterdam or Copenhagen where a “cyclist” could be anyone and doesn’t carry the same connotation and typecast.

3.4 Cycling as an Influential Factor in Itself

The variables that will be used in the empirical data collection have been identified above. An important part of this study is understanding how transportation attitudes and perceptions can also be influenced through marketing and imagery. Daley et al. (2011) have shown that the public images of different types of cyclists has an impact on how others perceive cycling. Their study organized focus groups in Sydney, Australia and discussed cycling issues like social status, acceptability, and image. Respondents generally felt that cycling was a positive thing both for individuals and the environment but disliked the images and actions of certain cyclists. The major highlights of the study found that individuals in Sydney feel cycling should be primarily a recreational or sport activity and that people who cycle for utilitarian or commuting purposes are taking unnecessary risks (IBID). There are many different types of cycling e.g. touring, utility riding, commuting, racing, mountain biking etc., each style having its own individual features that include differences between the bikes themselves but also with the various behaviors from the users. Each style portraying an

impressionable image and unique identity to the outside observer (Horton et al., 2007). How these different types of cycling are viewed can promote or discourage cycling depending on the observer. The perception of risk taking cyclists on a racing bike, for example, can affect the observers’ feeling toward cycling one way or another. If the observer self identifies with this sub group of risk takers they may have a favorable opinion of the image while others may view the image and behavior as obnoxious or dangerous. This is significant because the majority (53%) of bike trips in Sydney are for social or recreational purposes and mandatory helmet laws cause the majority of cyclists in Australia to wear safety equipment while cycling (Daley, 2011). Respondents felt positive towards the images of recreation and racing types of cycling but negative towards commuting and courier cycling. These opinions are fitting with the type of cycling generally experienced in Australia. A similar study in Amsterdam or Copenhagen would likely conclude with very different, maybe even opposite results. This is important to understand going forward because it has been argued that bike sharing systems have been very influential and that the different features of the different bikes are more appealing to different types of people.

Thomas Krag and SidselBirkHjuler (2013) used images to study the influence of positive and negative marketing messages that relate to bicycling on individuals’ attitudes and perceptions. A survey was conducted to rank opinion statements regarding mobility issues using images to see how different transportation modes were perceived. One aspect of the study asked questions based upon how a given transportation mode impacts a person’s appearance. The results found that most respondents feel the image of cycling is preferred above all other transportation modes (IBID). Another result shows that the image presented had an impact on respondent’s perception of transportation. An image of a bicycle accident, for example, increased respondents’ opinion of both general cycling risk and personal risk of cycling. The major takeaway was that safety campaigns and images can have a negative impact on cycling but campaigns that promote cycling as safe and fun often have positive impacts on the perception of cycling. The extent to which marketing influences cycling behavior is difficult to measure but is one of the many influential factors (Crouse, 2010). Different types of marketing strategies can be effective at stimulating responses.

(21)

Mikael Colville-Andersen (2009)argues that successful cycling cities like Copenhagen or Amsterdam were able to immediately achieve “mainstream cycling” because there was never a significant influence from bicycling “subcultures.” While he doesn’t explicitly define either mainstream or subculture, the implied meaning is that smaller groups of cyclists e.g. urban warriors, mountain bikers, bike messengers, have had a negative impact on cycling in cities/countries (the USA or Australia for example) that weren't able to make the initial leap into a mainstream bicycling culture. In other words, because these fringe groups initially dominated cycling many potential cyclists were discouraged because of the extreme stereotype that was produced. He goes on to argue that cities like Paris and Barcelona are now positioning themselves to jump directly to mainstream cycling in a similar fashion that the Netherlands and Copenhagen have because they lack the subcultures that exist in the US and have implemented effective bike sharing programs. Colville-Andersen also began the “Cycle Chic” movement in 2007, which captures cyclists in everyday, often fashionable clothing in photos and blogs. The idea is to promote a shift back to the idea that cycling is stylish and accessible for all types of people.

3.5 Bike Sharing System’s Influence on Bicycling Behavior

The concept of bike sharing began in the 1960s in Amsterdam with the white bike movement. An anarchist sociopolitical organization called “Provo” initially worked with city council leaders to implement a free bike sharing scheme. When negotiations failed between the ideologically opposed groups, the Provo’s began to implement unsanctioned ideas on their own. Bikes were painted white and left unlocked at various locations in the city’s center for anyone to use. This idea was only one part of a larger ideology that included a similar car sharing idea as well as promoting the occupation of unused houses in Amsterdam. The white bike program quickly faded away due to vandalism and opposition from city leaders. At one point city leaders impounded many of the bicycles because it was deemed illegal to leave unlocked bikes on the streets. The Provo’s response was to provide locks on each bike and paint the combination to each lock on the frame of the bike(Voeten, 1990).

LuudSchimmelpennik was among the bicycle experts interviewed for this study. Luud was the originator of the white bike and white car ideas in the 1960s and continues to come up with innovative solutions to urban planning issues. He has pioneered the technology and approach of all successive bike share programs including the latest systems in use today.

The second generation of bike sharing came in the 1990s with programs in cities such as

Copenhagen where users had to deposit a coin to utilize the system. Upon return of the bike users would get his/her coin returned. This approach was different from the first generation in many ways. The program was now sanctioned by city leaders and organized by a local non-profit organization. The goals and ideology were based more upon finding innovative solutions for increasing congestion and providing citizens with different transportation options. The bikes used in this system were specifically designed for bike sharing use with large sturdy frames and thick, rubber tires as well as unique mechanisms that locked each bike to a stationary rack. Many features established in this phase ended up lasting into the next generation of bike sharing including the stationary racks and specially designed bicycles. Systems of these kind also eventually faded away due to theft and vandalism. One of the major critiques of these systems was that there was too much anonymity between the system and the users (DeMaio, 2009).

(22)

The latest generation of bike-share programs have built upon the mistakes of previous attempts and have also benefitted from current technology finally converging with progressive ideals. These systems use a variety of high tech innovations that have greatly increased the efficiency of bike sharing. Specially designed bikes are electronically locked to stationary racks throughout the city. Users can access a bike after signing up online and usually paying a nominal fee. Most systems provide users with a recognizable card or code allowing them to sign in and out as they use the service. These features have greatly cut down on theft and vandalism because all users and bikes are tracked and accounted for with GPS. There are many variations regarding actual implementation but most 3rd_{generation bike share systems follow this general structure. Berlin’s “call-a-bike,” for} example, has users call a phone number when they are ready to utilize the system. They are then given a code to unlock a specified bike. The process is repeated when the bike is returned at the users’ destination (visitberlin.de, 2013). Compared with the two previous generations there is now significantly more upfront and maintenance costs with these systems. Most are operated through some variation of a public/private partnership with the public providing most of the upfront costs, although this is not always the case. The cost of bikes in each system vary from $500/bike up to $3500 per bike as will be detailed below. The largest and most successful programs have been organized and maintained by advertisement companies partnering with city governments.

Advertisement companies like Clear Channel and JCDecaux operate systems in exchange for public advertisement space.

Bike sharing programs have become the world’s fastest growing mode of transportation

(Straatemeier, 2013). Cities that have implemented bike sharing programs have generally seen the positive impacts these programs directly provide, such as individual health benefits and financial savings for users, as well as emission reductions and reduced congestion for cities. (Shaheen et al., 2010). They are designed to promote short trips from station to station, ideally within 30 minutes (Fishman et al., 2013).They have proven to be an efficient, flexible transportation option but more effective at converting users from other sustainable transportation modes like walking or public transport. Only a small percentage of car drivers have converted to bike sharing (Shaheen et al., 2011). Trips are taken for a wide variety of purposes but most are taken for recreation or commuting (LDA Consulting, 2012). At a recent transportation planning conference in Strasbourg a group of planning professionals discussed the role of bike sharing programs. They concluded that the Netherlands’ OV Fiets should be regarded as the ideal example of how bike sharing should be used,primarily because it has been used to supplement the bicycling culture that already exists (Colville-Andersen, 2013). Many current systems are using bike sharing systems for the opposite reason, to influence a different cycling culture. Among the initial questions that emerge regarding the role of these nascent bike sharing systems is whether or not they are an effective tool that is encouraging the transition towards more personal cycling and, if so, which of the system variations is the most influential.

As mentioned in the introduction, recent studies have also shown the indirect impacts bike sharing systems have on cycling behavior such as their ability to potentially normalize cycling. An

observational study in the US has shown that users of bike share systems are less likely to wear a helmet (Fishman et al., 2012). This is significant because, as previously mentioned, there is

increasing evidence that helmet usage can be a barrier to cycling and also convey an image to other road users that cycling is a dangerous activity (Krag et al., 2012). Bike share users are also more likely to be wearing normal clothing instead of sporty cycling apparel (Goodman et al., 2014).This plays a

(23)

similar role in helping cycling seem normal and acceptable to all, not just certain groups. Goodman et al. have also shown that, similar to helmet usage, cycling apparel can be a barrier to cycling because not many people want deal with the hassle of extra clothes, shoes etc., but it also conveys to other users that cycling is a sporty, exclusive activity. Also mentioned above is the higher

percentage of females using bike sharing systems when compared to personal cycling (Szczepanski, 2013). This is significant because women are underrepresented in cities with low cycling numbers and “the indicator species,” (Pucher et al., 2011) meaning that a high percentage of female cyclists in a city is a sign of a mature bicycling culture.

Bike sharing systems have proven to have direct benefits for both individuals and cities as well as indirectly influencing certain aspects of bicycling perception and behavior. These studies imply that bike sharing systems might be used as a tool to normalize the image and perception of cycling. The area that hasn’t been examined, which was identified in the introduction as part of the issue, is how different types of bikes are impacting cycling perception. The latest generation of bike sharing systems have many similarities but there are also noticeable differences between systems, particularly within the individual bikes themselves. Barcelona’s “bicing” bikes weigh 16.8 kg, have smaller (20 inch) front wheels than their larger rear wheels, straight (zero degree) handlebars, and U shaped front racks with attached bungee cords to transport items (Antoniades et al., 2009).The bicycles used in Paris’ “Velib” system weigh 22kg have a large front basket to transport items, a forty five degree handle bar which allows riders to cycle more upright when compared to Bicing bikes. These bikes are also very easily identified and branded as a bike sharing bike (Antoniades et al., 2009). Bicycles used in Rome more closely resemble traditional, personal bikes like the “stadsfiets” mentioned above. Each bike has forty five degree handlebars and is equipped with a large front basket and a rear-wheel rack. One of the critiques of this system is that the bikes look so traditional that they may not convey a 3rd_{generation bike share system (Antoniades et al., 2009). These are a} few examples of the many differences that exist between the bikes used in various bike sharing systems.

3.6 Conclusions of Theoretical Research

The above literature review hasanswered the first sub-research question , which asked: “What are

the various elements influencing individuals’ attitudes and perceptions towards bicycling and how are these elements related?” Individuals’ travel behavior is influenced from a wide variety of

variables. Current transportation habits and activities (unreasoned influences) play a significant role. Transportation behavior is also continuously influenced by exposure to transportation stimuli through reasoned influences. Both factors must be understood within a spatial and social context to gain a proper understanding of individuals transportation behavior. The spatial and social variables were further detailed using the bicycling determinants from Rietveld and Daniel (2004). These included spatial elements such as density, climate, and bicycle infrastructure as well as

social/sociocultural elements such as a city’s bicycle culture and commitment to cycling through a comprehensive bicycle plan and helmet policies.

In addition to detailing the variables that will be used in this thesis, this chapter has provided necessary context for understanding how the topic of bike sharing relates to theory. Bicycling has become an increasingly popular mode of transportation not just for recreation but for daily utilitarian purposes which has led many cities around the world to follow the example of mature

(24)

cycling cultures like the Netherlands, Denmark, and Germany. One of the barriers these cities face seems to be the image and behavior of cycling.The image and behavior of cycling can influence individuals’ perceptions of cycling positively or negatively based upon the kind of cycling they

identify with (Daley, 2011). Bike sharing programs have become many cities’ primary response to the growing demand for sustainable transportation modes. The latest generation of bike share systems have high upfront costs but have shown to be an effective way to get more individuals cycling. They have also shown efficiency in normalizingthe image of cycling by breaking down cycling stereotypes that exist in many cities that lack a normalized bicycle culture. Bike share systems that have

influenced the normalization of cycling through type of clothing (Goodman, 2014) and not requiring a helmet (Fischer et al., 2012). According to the above theory these are positive steps towards normalizing bicycling, however, there are many variations from system to system, particularly regarding the type of bike being used. If the ultimate goal of these cities with low cycling numbers is to move towards a cycling culture similar to the Netherlands, Denmark, or Germany and the bulk of their efforts is being put towards bike sharing systems, then it will be important to understand how the differences between types of bikes in these systems are influencing cycling perception and behavior. Following this logic this study will proceed with the following hypothesis:

Different Characteristics used in 3rd_{Generation bike sharing systems significantly affect the}

unreasoned elements of cycling behavior

3.7 Conceptual Framework

The updated conceptual framework builds upon the one in the introduction by including all the relevant variables mentioned throughout the theoretical framework. Reasoned influences are represented by the box on the left, in this case the reasoned influences are the different bike share systems. The red arrows represent the different characteristics of bike sharing systems. Taken together these are the elements which are influencing individuals’ attitudes and perceptions

towards bicycling which is then influencing bicycling behavior. Because there are many other factors that influence bicycling behavior, the box to the far right represents unreasoned influences such as habits and daily routines. The two boxes above and below bicycling behavior are the other two elements previously detailed that also influence bicycling behavior. This conceptual framework has been adapted from Van Acker et al. Also included are the variables from Rietveld and Daniel (2004). Mapping all of these variables provides an opportunity to see how each bike share system is

influencing individuals differently. If an individual’s spatial and social context is understood along with their daily transportation habits then their attitudes and perceptions of bike sharing systems can be measured by the varying characteristics of each system.

The influence of bike share systems on cycling perception and behavior.