In this mixed-method study, we aimed to explore the motivations for walking and cycling of participants living in Cambridge, UK, who also reported unsupportive conditions for walking and cycling on their route to work in questionnaires. Although cycling in Cambridge is relatively normalised, the presence and quality of cycle infrastructure varies and has been described as “more infrastructural patchwork than a paradise” . In this context, we identified three main themes based on content analysis of qualitative interviews. First, it appeared that participants had gathered knowledge and experience which enabled them to remain resilient and cope with these unsupportive conditions on their route to work. Second, it was apparent that many people endured these conditions for reasons unrelated to the environment, such as childcare commitments or car parking arrangements at work. Finally, and perhaps most interestingly, it appeared that some of the more experienced active commuters may have responded in such a way as to represent broader public opinion; it might therefore have been others, such as family members or members of the general public, who perceived cycling or walking as too dangerous and whose behaviour was affected as a result.
Understanding the motivations for walking and cycling in unsupportive environments might be just as important as investigating the barriers to cycling. All participants had reasons for enduring these adverse environmental conditions and had done so in a variety of ways — whether by acquiring experience, knowledge or confidence, by making pragmatic choices to use more convenient or cheaper travel modes or to make longer but safer journeys, or as a result of weighing up the perceived benefits and costs of the options given their own circumstances. Just as insights gained from understanding successful ‘weight maintainers’ can make novel contributions to strategies to prevent weight gain and obesity , we suggest that tapping into these mechanisms of resilience of active travellers who overcome their unsupportive travel environments could inform policy and practice and contribute to improved theoretical frameworks of the interaction of individual and environmental characteristics in shaping behaviour change [6, 27].
We found that many of the reasons for walking or cycling in hostile environments were overcome by avoidance of certain sections of the route based on experience or knowledge. Joshi and Senior  conclude that existing cyclists who have experience of cycling in traffic report traffic danger as less of a threat than those who have no such experience. As such, perceptions of busyness may depend on personal circumstances, confidence or cycling ability. In this study, we also found that cyclists in particular sought to find spaces which were empty in order to physically avoid traffic and conflicts with other road users, including motor vehicles and other cyclists. It appears that building up knowledge and experience of sections of their route which were more pleasant, and using them, somehow counteracted the predominantly unsupportive and sometimes dangerous conditions which they faced on other parts of their route. In keeping with the predictions of some theories of health behaviour change , therefore, increasing awareness of traffic-free routes through education and promotion may improve ‘journey literacy’ about the availability of these cycle-friendly routes in the local environment. Participants also raised the issue of potential conflicts between pedestrians, cyclists and drivers of motor vehicles, but as well as trying to avoid these conflict areas, experienced cyclists also knew how to negotiate the road with other users (in particular with motorists), either to get out of their way or to be dominant in these situations. Although our findings would need to be replicated in a larger sample, they suggest that interventions to promote cycling which focus on improving cycling skills and confidence through cycle training should be continued. These views are in line with recent commitments made by the Department for Transport, which will continue to support cycle training (‘bikeability’) schemes until 2015 . Despite this, Cambridge cyclists in this sample did report feeling somewhat inferior and had to yield to motorists, suggesting that even in a city known for cycling, people still report feeling unsafe in both questionnaires and interviews.
Consistent with other studies  and other analyses from this study [23, 31, 32], and with some of the predictions of conventional economic theories of travel behaviour , we found that commuters made the decision to walk or cycle pragmatically based on a combination of other reasons unrelated to the route environment, such as convenience, personal preference and both domestic and workplace constraints, including the convenience and cost of alternative modes and the availability of car parking at work. A number of these are modifiable factors and could form components of an intervention strategy to promote walking and cycling. Previous research in this sample [26, 31] and elsewhere  has highlighted the importance of parking subsidies or charges, finding that those who have to pay for parking are less likely to use the car for commuting. Interventions focusing on restricting on-site parking at work may encourage commuters to consider making the journey on foot or by bike, while providing free or subsidised car parks off-site but within walking or cycling distance may also encourage walking and cycling for part of the journey. At the same time, we found evidence that there are factors that prevent individuals from cycling; these are fluid, vary according to people’s personal circumstances and preferences and may evolve over time .
Walkers and cyclists also talked about the additional benefits of walking and cycling beyond that of cost, such as enjoyment and exercise. They perceived themselves to have ‘greater control’ over journey times, were less reliant on others and enjoyed the experience, although in some cases the latter appeared to be a by-product of the barriers to alternative travel modes (such as the inconvenience of sitting in traffic or the availability of car parking). Promoting these secondary benefits may also form an important component of an intervention strategy to promote walking and cycling in some settings, particularly in congested city centres.
Finally, in interpretive qualitative analysis we also probed more deeply into less explicit explanations of the discrepancy between reported perception and behaviour. Most significantly, some interviewees — largely experienced cyclists — distinguished between their own perceptions that were more positive towards their commuting environment and the perceptions of others who did not cope as well with unsupportive environments. Other participants used their interviews as an opportunity to make a political statement that environments should be made more supportive for walking or cycling. Assuming that these interviews affirmed, contextualised and differentiated their survey responses, we suggest that their survey responses should also be read as potentially reflecting such public opinion and political statements. Indeed, a recent sociological study also conducted in Cambridge confirms the importance of cycling citizenship and cycling activism in this setting  and thus the possibility that survey responses might be shaped by an increasing public awareness of cycle campaigning e.g. how media discourse on nuclear power provides an essential context for interpreting survey results on nuclear power . While qualitative research often includes the exploration of public discourse in its research agenda , survey data seem to be another potential source for such analysis.
Mixed-method research projects increasingly aim to achieve both breadth and depth to address their research problem, but there is an emergent debate as to whether such projects are simply a response to a current ‘fashion’ or reflect a true attempt to use this approach to yield new research outputs . Our analysis enabled us to probe behind survey responses and find explanations and contextual data. Surveys gather data in separate sections – here, travel behaviour and environmental perceptions – without the opportunity for a participant to justify their choices. In contrast, open-ended, in-depth interviews record contextual narratives on experiences and attitudes and enable interrelations to be identified; but as qualitative inquiry does not produce representative data, the extent of coping strategies or practices of resilience should not be generalised.
The limitation of this mixed-method analysis lies in the inherent difficulty of consolidating both methods and datasets and producing a research output that allows for both qualitative and quantitative terminology. This article presents mainly qualitative findings, as our analysis aimed to explain quantitative questionnaire results with the help of contextual data from the qualitative interviews. Moreover, the study focused on the very particular setting of Cambridge and drew on a small sample of highly educated and eloquent participants who volunteered to participate in interviews. Nonetheless, as with most qualitative social research the aim of this work was to produce in-depth, exploratory and explanatory findings rather than to address expectations of representativeness or generalisability. While the integration of qualitative and quantitative methods requires careful consideration of the design and approach to analyses , the accumulation of ‘mixed’ bodies of evidence should be encouraged to inform a more comprehensive understanding of behaviour. For example, while the combination of qualitative interview data  with quantitative survey data  can provide greater understanding of the reporting and meaning of ‘perceptions’ of the environment, as in the current paper, this could be complemented with objective measures of behaviour such as those collected using global positioning systems (GPS)  to provide a more comprehensive understanding of route selection and other tactics used by cyclists to negotiate an apparently unsupportive environment. Further research should consider exploring the reasons for non-participation in walking and cycling in other settings in order to inform interventions to encourage people to choose active modes of travel in environments that are often less than ideal.