Skip to navigation – Site map

HomeThemed issues23IssueMetropolitan mobility options and...


Metropolitan mobility options and individual identities: How much households' mobility choices are determined by residential-related identities?

Sébastien Lord, Benjamin Lamoureux and Philippe Gerber


Daily mobility is an ordinary problem that fits into the complexity of everyday household life. Mobility is also a dimension of identity, a predisposition which gives individuals a specific framework for dealing with activities and travels. Thus, households’ choices are made with a wide range of rational elements (economic, time, accessibility, etc.), but also with more sensible elements (preferences, past experiences, satisfaction, etc.). In addition, large-scale urban projects aim to improve modal split and to facilitate households’ commutes conditions. Since poles of employment generate the main concentrated flows of urban travel, those poles are at the core of strategic planning. Beyond improving mobility, changing job location may involve contrasted issues for households, positive (e.g., travel efficiency, sustainable mobilities, etc.) and negative (e.g., freedom of choice, undesirable mode change). This paper focuses on an exceptional workplace relocation (2015) in Montréal – McGill University Health Center – and explores the daily commute choices for more than 10,000 metropolitan workers.

While metropolitan accessibility is sharply improved, the new workplace offers a low-quality mobility environment compared to former ones (n=5) located downtown. What mobility choices do metropolitan employees make in order to rebuild their travel routines? What is the importance of both mobility habits and mobility identities in commute changes/satisfaction? Using an internet retrospective survey (n=1 977) conducted with the concerned workers, mobility strategies have been explored. Principal component analyses and clustering are used to take into account lifestyles typology into a logistic regression model in order to explain mode choice according to identities, workers’ socioeconomic characteristics, built environment and structural urban characteristics. Beyond the importance of accessibility, attitudes towards mobility have strong explanation power for specific workers' profiles. These variables should be better embraced by transportation plans.

Top of page

Full text


1Daily mobility is an "ordinary" problem that is embedded in the complexity of everyday household life (Yang et al., 2019). The collective accumulation of transportation mode choices, however, can have significant economic, social, and environmental consequences. Daily mobility is thus central to many planning and urban development policies. Mobility that is less dependent on the car is an "object of desire" in public policy. A good understanding of the mechanisms at work in household mobility choices is therefore crucial, both from a fundamental perspective on the construction of travel choices and for strategic urban planning adapted to their needs and preferences (Pritchard, Frøyen, 2019). Without this, the policy, program, or strategic direction is invalid, or at least limited in scope.

2If households evolve in a field of possibilities in terms of accessibility (Hägerstrand, 1970), the choices made in this constrained space fall under logics that go beyond functional and economic rationalities. Indeed, the development of routines contributes to the acquisition of spatial capital (Lévy, 2003; Kaufmann et al., 2004). Travel habits are the object of attachment and play a role in the definition of settlement or place identity (Feldman, 1990, 1996; Twigger-Ross, Uzzell, 1996). Understanding the mechanisms of preferences and decisions appears fundamental, both in terms of what facilitates or does not facilitate daily mobility routines, but also in the development of mobility options that go beyond spatiotemporal optimization. Our article proposes to approach the mechanisms at work in mobility choices while considering the residential-related identities that fuel, or do not fuel, these choices.

3First, we discuss the construction of household mobility choices and the different rationalities involved in the decision-making process. The hypothesis of residential-related identities as important influencing factors of mobility choices is explored in combination with well-known spatial and socioeconomic determinants. Next, the mobility choices of households are studied using an Internet survey of metropolitan workers (n=1 977). The objective is to evaluate the importance of the dimensions related to residential-related identity in the mobility choices compared to the classical socio-economic dimensions. To do so, a lifestyle classification involving variables on residential-related identities is first performed with the help of a principal component analysis and a hierarchical cluster analysis. Then, a multinomial logistic regression model puts into perspective the modal choices according to the lifestyles and socioeconomic characteristics of the workers as well as those of their residential environments. In conclusion, we discuss the results which show that spatial accessibility and lifestyles have a relatively strong explanatory power for specific profiles of workers, and even more than socio-economic characteristics.

1. How do households choose their modes of transport?

4Mobility choice is a complex notion encompassing many factors, both subjective and objective, interrelated in space and time. While there is a suitable mode of transport for every potential mobile person, many factors that vary from one individual to another must be considered to understand how mobility choices are formed (Kaufmann, 2004). Research shows the relative importance of several major types of factors: sociodemographic, instrumental and contextual, symbolic and psychological (De Witte et al., 2013), but also biographical (Lanzendorf, 2010; Scheiner, Holz-Rau, 2013). Instrumental and contextual factors are those related to the efficiency of one mode of transport compared to another (Steg et al., 2001), such as travel time, the comfort of a mode, its monetary cost, or, of course, its accessibility and the associated urban structure (e.g., Ewing and Cervero 2010; Yang et al., 2019). Objective factors are contrasted with symbolic or affective determinants, such as a sense of control or self-image – the use of the car as a social marker has been emphasized in numerous studies for many years (Fichelet et al., 1970; Jensen, 1999; Steg 2005).

1.1 Modal choices arbitrated between determinants and fictional rationalities

5Whether or not there is an existing transport alternative, the decision to use one mode or another is not generally the subject of a conscious weighing of rational interests. In the complex process of mental construction of the best possible trip, the multiple biases of personal representations place the individual and their household in a false situation of choice. From the representation of space to the value attributed to time, all these factors make the choice of mobility, in a spatiotemporal framework of possibilities, a highly subjective element (Kaufmann, 2004). Similarly, once a modal habit has been generated, it is difficult to break it (Aarts and Dijksterhuis, 2000; Verplanken et al., 2008); and this habit will tend to reproduce itself, in the logic of the trajectory, since it avoids a mental reconstruction that can sometimes prove complex (Brisbois, 2010).

6In this richness and complexity of the multiple determinants of modal choice, accessibility is an instrumental factor that has been integrated into many analyses and its central importance in modal choice has been widely highlighted (Barff R., Mackay D., Olshavsky, 1982; Ewing, Cervero, 2010; Beimborg et al., 2003). Indeed, as several authors have noted, among the many instrumental factors, accessibility – to a destination in particular – is one of the most crucial. Density, on the other hand, is only a proxy for other variables (Cervero, Kockelman, 1991). Even so, a direct link between modal choice and accessibility may be emerging, although it has not been formally demonstrated. Other factors, such as distance to work, have often been put forward as a major determinant of modal choice (Godefroy & Morency, 2010).

7The logics and rationalities of the activity programs of individuals and of households (Amar, 2010; Kaufmann, 2004; Kesselring, 2006) are of great interest with respect to modal choice. The construction of activity programs (Bourdin, 2005; Meissonnier, Richer, 2015) considers the combination of individuals' needs and aspirations to move within a field of spatiotemporal possibilities of instrumental factors. Households develop routines that are part of a problem-solving logic (Juan, 1997). They are adjusted more or less continuously, according to the hazards of daily life, and according to different, more or less conscious rationalities (Ascher, 2006; Énaux, 2009). In this context, a balance determined by the economic resources of the household and the daily constraints of its members, marked preferences for the car are observable (Lucas et al., 2011). The car allows a sequence of activities that is perceived as easy. In addition to the comfort and control of daily mobility, it offers flexibility in the conduct of activities (i.e., Pradel et al., 2005; Lesteven, 2015).

1.2 Modal choices constructed and combined with residential-related identities

8In the development of travel routines, we observe the central place of the car in mobility habits and the fact that its use is rarely questioned, unless there is a significant event (e.g., moving, change of job, etc. (see Rau and Manton (2016)). In fact, by changing their mode of transport, individuals, and sometimes their households, must reconsider their activities, and therefore the locations where these activities take place and the people associated with them. This forces a review of routines because the accessibility of the car and of public transport, for example, are not the same. Mobility changes are thus gradual and can lead to car use in incompatible contexts or generate considerable costs (time and money) for households (i.e., Lesteven, 2015; Lucas et al., 2011; Dupuy, 2006) without them realizing it. Thus, households evolve in a field of possibilities in terms of accessibility (Hägerstrand, 1970), and the choices made in this constrained space fall under logics that go beyond the rationalities of instrumental factors. Indeed, the development of routines contributes to the acquisition of spatial capital (Lévy, 2003; Kaufmann et al., 2004). Travel habits are the object of attachments and partially defines residential-related identities (Feldman, 1990, 1996; Twigger-Ross, Uzzell, 1996). How, then, can one make a choice that is consistent with one's identities?

9Sociologists and psychologists Burke and Stryker (2000) and Schwartz et al. (2011) have developed an analytical understanding of the complex concept of identity by referring to the answers to three simple questions (at three levels), "Who are you?" (Relational level), "Who am I?" (Individual level) and "Who are we?" (Social level). In addition to these three groups of entities, there is also an aspect of identity that needs to be considered, namely the material dimension that considers the material artifacts that influence behavior. For example, Murtagh et al. (2012) and Heinen (2016) found that transport mode-related identities (e.g., being a car driver or using public transport) remain significant in explaining travel mode choice, all factors being equal. Thus, understanding the mechanisms of preferences and decisions appears fundamental, both in understanding what facilitates or does not facilitate mobility routines, but also in developing sustainable mobility options that would go beyond spatiotemporal optimization. This is what we explore in the remainder of this paper.

2. Methodological framework

10An internet survey of workers at the McGill University Health Center (MUHC) was conducted in May 2018. A self-administered web questionnaire (LimeSurvey) of about 100 questions was constructed and validated. Six themes were targeted (for more details, please see Zarabi et al., 2019; Gerber et al., 2020; Ma et al., 2021): 1) daily activities / commuting; 2) trajectory / residential project; 3) housing / commuting costs; 4) perceptions of changes; 5) identities, values and lifestyles; 6) attitudes towards transportation modes. The daily mobility of workers could be examined objectively (e.g., costs, time) and subjectively (e.g., satisfaction indices, attitudes and preferences). The mobility choices analyzed were those of workers to their place of work only, e.g., between the McGill University Health Centre (MUHC). Located on the Glen site in the middle West of Montreal Island, this major employment pole, with more than 10 000 jobs, is accessible by all modes of transportation present in the Montreal metropolis. Located at a major highway junction in the agglomeration, it has a multimodal station (commuter train and metro) and is served by several bus routes and a bicycle path.

2.1 The MUHC workers’ sample

11Our analyses rely on a sample of 1 977 metropolitan workers from the Greater Montreal area in Quebec, Canada. Figure 1 presents their geographic distribution and Table 1 presents the independent variables. They are distributed throughout the agglomeration, with a greater concentration on the island of Montreal, especially near the metro network. Women (78%) are over-represented, with a strong presence of nurses (34%). The job categories are relatively varied, with administrative, health and care-related or maintenance jobs. This variety is consistently observed in respondents' incomes, where there is a high degree of variability, with 14% earning less than $60,000 per year and 14% making over $180,000. More than 3 out of 4 respondents were born in Canada, while 12% came from abroad. Over 85% of respondents own at least one car and 19% live in a neighborhood from which access to the Glen site is faster by public transit than by car.

Figure 1: Distribution of employees’ place of residence and main mode of transport

Figure 1: Distribution of employees’ place of residence and main mode of transport

Table 1: Independent variables

12The dependent variable we selected to analyze mobility choices is presented in Table 2 according to the available modalities. We explored the factors of influence on this dependent variable according to the characteristics of the participants, their jobs and their place of residence (Table 1), and a typology of lifestyles based on residential-related identities that we created. How then does one make a choice based on one's residential-related identities? (section below).

Table 2: Dependent variables

2.2 Lifestyles of the metropolitan workers

13The first step consisted of conducting a principal component analysis (PCA). This technique allows us to obtain a synthetic and global vision of a series of variables associated with the residential-related identities of the survey respondents (Table 3). This step allows us to create lifestyles.

Table 3: Variables of residential-related identities

Table 3: Variables of residential-related identities

14Using principal component analysis (PCA), the identity and value scales were merged into 7 components that defined the employees' lifestyles after using hierarchical clustering analysis (HCA). This two-step approach is used in modal choice analysis (e.g. Gao et al., 2022). It allows to synthesize information from latent variables into clusters that are easy to interpret in a model. Each lifestyle was then described in terms of socio-economic dimensions as well as their mobility and residential status.

15Identities are not directly observable but can be inferred using Likert scales. Latent variable models are commonly used when an unobservable or latent construct is indirectly measured by observable exogenous variables (Washington et al, 2003). Likert scales of 5 levels of measurement (1-5) (Table 3) were considered for all identity and value variables. Participants were asked to give a score and not a qualitative response. We can assume that there is no difference between perceived and numerical distances. Other work has applied PCA to non-numerical data, even for more subjective and entirely arbitrary scales (Gitelman et al., 2010; Papadimitriou et al., 2012). When performing a PCA, the variables must be at least approximately normally distributed. The descriptive statistics (Table 1) of the sample provide no evidence that the distribution assumption is violated. PCA suffers from a limitation that assumes that the relationships between variables are linear. We tested categorical PCA for nonlinearity in structure, without convincing results.

16Table 4 presents the results of two PCAs, one conducted with the identity items, the other with the residental-related value items. Four components explained 56% of the variance in identities with 1) a work and health component (18.4%), 2) a family component (16.8%), 3) a self-mobility component (11.9%), and 4) an urban component (9.0%). Three components explain 47.0% of the variance in values with 1) a neighborhood attachment component (21.0%), 2) an urban flexibility component (16.4%), and 3) a residential ownership component (10.4%).

Table 4: Results of principal component analysis (PCA)

Table 4: Results of principal component analysis (PCA)

2.3 Creating the Lifestyle Typology

17PCAs provided new latent quantitative variables, which were used to define lifestyles groups through hierarchical ascending classification (HAC). The Ward's method is used which increases in the sum of squared errors that would result from a merger. The number of groups is chosen using a dendrogram (Figure 2). This diagram summarizes the different steps of the analysis by representing the successive groups. The number of groups chosen depends on a social and not only statistical logic. The significant associations (Chi-square and Anova) between the groups and the questionnaire variables were used to construct exploratory multinomial logistic regression models.

Figure 2: Dendrogram of the hierarchical cluster analysis

Figure 2: Dendrogram of the hierarchical cluster analysis

18Figure 3 shows the four major lifestyles that emerged from the hierarchical bottom-up classification: (1) Motorized Family Lifestyle, (2) Unrooted Lifestyle, (3) Active and Flexible Lifestyle, (4) Rooted Metropolitan Family.

Figure 3: Typology of residential-related identities

Figure 3: Typology of residential-related identities

(1) Motorized Family Lifestyle

19The first type of lifestyle is the Motorized Family Lifestyle. It serves as a reference and is representative of the living patterns of the Montreal metropolitan area in several urban forms, in the near and far suburbs. It is strongly marked by individual car mobility. This profile stands out, globally, by being more female, more Canadian and having the highest proportion of nurses. This profile has higher incomes than the sample average, but no more children and shows a lower rate of residential mobility. This group has a very high proportion of car users and the lowest proportion of metro or bus users, but the suburban train is used more than other groups. Car ownership is consequently higher. Their commuting time is higher.

(2) Unrooted Lifestyle

20The second profile is the (2) Unrooted Lifestyle profile and seems to correspond to respondents at the beginning of their professional career. This group is made up of younger people with lower incomes, fewer qualifications, less weekday work, and more nursing jobs. Their residential mobility is higher than the sample average. They do not show significant differences in mode choice but have higher commuting times. This profile also shows more often people born outside Canada who work in research positions with lower incomes. This group tend to favour public transit and has low levels of car ownership.

(3) Active and Flexible Lifestyle

21The (3) Active and Flexible Lifestyle is characterized primarily by the highest proportion of people under 35 years of age and the lowest proportion of households with children. Research positions are slightly overrepresented, while service positions are more underrepresented. These workers have higher residential mobility than the sample average and use public transportation more. This lifestyle is the one that uses the most active modes of transportation. These respondents own the fewest cars and have the shortest commute times.

(4) Rooted Metropolitan Family Lifestyle

22The fourth lifestyle is the (4) Rooted Metropolitan Family Lifestyle. It is structured according to activity and mobility components of respondents who are attached to their residential environment. These respondents are the oldest and have the highest proportion of physicians, with consequently higher incomes. They have low residential mobility. These respondents have varied mobility profiles with all modes represented, although they use active modes more. Nevertheless, they own on average more cars than the rest of the sample. This group stands out as being more female and European. It is also the group with the fewest people who use or own a car. The travel time of these workers is about average.

3. Analysis of mobility choices

23In order to better understand the role of classical variables in the analysis of mobility choices, but also of residential-related identities grouped into lifestyles, a multinomial logistic regression model was constructed. This analysis makes it possible to highlight, ceteris paribus, factors likely to influence modal choices while isolating possible structural effects.

24The multinomial logistic regression model was constructed to predict the mobility choices to reach work (n=1 071) (Pseudo R-two Nagelkerke = 0.617) according to the available independent variables (Table 1): 1) socio-demographic characteristics of the individuals, 2) characteristics of the jobs held, 3) mobility characteristics, 4) characteristics of the residential environments and 5) lifestyles. The model allows us to approach the complexity of transport mode choices.

25Overall, there is a relatively small impact of socio-economic variables on the choice of mobility to work. Variables related to employment and residential environment prevail in the observed mobility choices. Lifestyles based on residential-related identities are the most significant variables in the model.

Table 5: Multinomial logistic model for choice of transport modes

26Choosing the bus, compared to the car, is almost 3 times more likely (2.7) for women than for men. Researchers (5.2), administrative professionals (3.6), and health technicians (4.2), all of whom have weekday schedules, are more likely than physicians to choose the bus. Income is not significant. The respondent mobility variable is significantly stronger. Reporting commutes of more than 50 minutes (4.6) implies a greater likelihood of commuting by bus compared to 30-to-40-minute commutes. Working on weekdays (4.1) compared to weekends increases the likelihood of choosing the bus over the car. Of course, households without a car are more likely to use the bus than the car (5.5), the opposite is also true for households with one or more cars, which are less likely to use the bus than the car (0.4). Compared with Motorized Families, all other lifestyles have high probabilities of choosing the bus over to the car: Rooted Metropolitan Family (7.2), Unrooted (5.3), Active and Flexible (5.0).

27The metro represents a stronger choice but, unsurprisingly, is common in a very specific territory. Choosing the metro, compared to the car, involves three socio-demographic variables. In this sense, employees born in Europe and in Africa/Asia are respectively 4.5 and 3.0 times more likely to take the metro than Canadian respondents. Age and sex play no significant role. Females are 1.7 times more likely to use the metro. Income is not significant. Professional health jobs (3.2), health technicians (4.6), and administrative professionals (2.6), who work weekday hours, are more likely than doctors to choose the metro. The perceived travel time variable shows that metro users are significantly more likely to report commutes of more than 50 minutes (3.3) compared to 30-40 minutes. Working on weekdays (8.3) compared to weekends shows greater likelihood of choosing the metro compared to the car. Compared with the Motorized Family style, all other lifestyles are likely to choose the metro over the car, namely the Unrooted style (6.6), the Active and Flexible style (14.1) and the Rooted Metropolitan Family style (5.6).

28Choosing the train, compared to the car, is observed within specific employee and territory profiles. The train network serves relatively remote suburban areas of the agglomeration, with limited access points to the network. This time, the youngest group, 20-34-year-olds (0.3), have a very low probability of choosing the train compared to 45-54-year-olds. Members of the high-income category of $140,000-$179,999 are 2.4 times more likely to choose the train than the reference category of $100,000-$139,000. Researchers (8.3), health professionals (11.3) and technicians (10.1) and administrative professionals (6.9), all of whom work weekday hours, are more likely than physicians to choose the train. The mobility variable for respondents is also significant. Commuting by train implies higher odds of spending more than 50 minutes (2.8) compared to 30-40 minutes, and expectedly very few 0-20 minutes (0.3) and 20-30 minutes (0.4). Choosing the train implies high probabilities of working on weekdays (9.7) compared to weekends versus the car. All other lifestyles have a higher probability of choosing the train over the car compared with Motorized Families. Unrooted (3.3), Active and Flexible (5.5) and Rooted Metropolitan Family (2.8) all follow this trend.

29Choosing active modes, compared to driving, is limited to areas surrounding the Glen site. Nearby neighborhoods are relatively privileged residential environments in terms of socioeconomic status. Respondents over the age of 55 are more than twice as likely (2.5) to choose an active mode compared to driving. Households with a child under 13 are 2.6 times more likely to choose these modes compared to those without a child. Income was not significant, nor were other sociodemographic variables. Nurses (0.3) as well as volunteers (0.2) and administrative technicians (0.1) had very low odds of choosing an active mode compared to physicians. None of the respondents' mobility variables were significant except for not owning a car (6.0) or owning a car (0.1). Some lifestyles were significant. Active and Flexible (9.7) and Rooted Metropolitan families (3.8) are more likely to choose active modes than Motorized Families. The Unrooted style is not significant.

4. Discussion

30The analysis of residential-related identities provided a deeper understanding of mobility choices in relation to household characteristics and their residential environment. Several previous studies have mostly examined these dimensions separately and rarely were able to effectively integrate the dimensions of identities and residential attitudes. Three points of discussion can be developed.

4.1. The importance of functional dimensions

31Our results clearly renew the role of the spatial context and especially of the functional configuration of metropolitan spaces when it comes to mobility choices. In the previously explored model of mobility choices, we note the importance of accessibility variables and car ownership, leading to some field of possibilities. The same observation can be made regarding the employment context, where the work schedule and the position held determine the field of possibilities at the spatiotemporal level. Working during weekdays rather than in evening, and in an administrative position rather than in a job with atypical hours, has just as much weight as the functional dimensions. In this sense, the employment location of the Glen site shows good multimodal accessibility, which gives households interesting choices for public transport, as long as their members work on weekdays.

32The metropolitan functional space from Montreal to the investigated employment location gives employees several options. This same spatial accessibility shows, however, that favorable accessibility to public transit creates certain opportunities. In this sense, certain urban sectors (proximity and accessibility of the place of employment) or points of access to the public transport network (metro stations, train stations, ends of metro/commuter lines) allow, or even support, more sustainable choices of public transport. In general, our analyses show the marginal influence of socioeconomic variables in mobility choices. As in other research (see Van Acker, Witlox, 2010), lifestyles seem to provide here a better approach to how mobility choices are made or in which type of residential environment it is possible to choose one or more modal options.

4.2 Between socio-demographic and psycho-sociological dimensions

33Socio-demographic variables are closely linked to metropolitan functional spaces and to the dispersion of types of employment. They show a strong structure of metropolitan living spaces, particularly spread out and specialized for North American environments. When inserted into a model of mobility choices, however, the socio-demographic variables are only marginally significant. On the other hand, the variables of residential-related identities are much more salient. The psychosociological disposition or sense of residential environment thus seems to take precedence over socio-economic status. Without necessarily giving more possibilities, lifestyles seem to orient choices within a field of possibilities at a given socio-economic level, or the past residential choices driven by lifestyles that may give the possible modal options. A similar observation about the lesser importance of socio-economic variables can also be made regarding the concept of satisfaction (Gerber et al., 2020). Like the residential-related identities, residential satisfaction is more structuring than socio-economic dimensions. This would confirm the precedence of individually and socially constructed predispositions in the life course.

34The respondents belonging to the Motorized Family lifestyle constitute the most motorized households. The car and the functional space of Montreal agglomeration could allow them to adapt continuously, without having to make different mobility choices. This elasticity brought about by car, in the type of space analyzed, is sufficient not to bring about a rupture in lifestyle and residential location. This lifestyle is definitely associated with a way of living structured by representations of the family in a suburban environment and where one depends on the car. Their commuting time is the highest and they can be very attached, both literally and figuratively, to their car and their residential setting. The analyses with most nurses were however conducted with individuals who could adjust their work schedules during the day, evening, and night, which is not possible for many other types of jobs.

35So, who are the people who choose a mode of transport other than the car? They seem to involve contrasting career and residential paths, in which lifestyles are intertwined. First, people from the Rooted Metropolitan Family stand out. These workers seem to be the most established in the professional or residential pathway. They adapt their mobilities to their residential choice. People in the Rooted Metropolitan Family and Active and Flexible lifestyles show a very high probability of opting for transit stability when the functional space of their residential environment is favorable. And this is precisely the case for many of the workers in these lifestyles. The "health" component of the Rooted Metropolitan Family does not appear to have a relationship with mobility, either for soft mobility or transit, where walking is present.

36Second, we observe that workers in the Unrooted lifestyle and Active and Flexible lifestyle are workers who seem to be beginning their career or residential pathways. Would this give them more flexibility in their mobility choices? We do not observe any significant difference in the choice of transport mode in these categories, except perhaps for higher commuting times. Cost could be the motivator behind the decisions of these groups. These individuals also appear to prioritize urban living. With these groups adopting an active, urban lifestyle, one wonders if their residential choice is likely to change with the arrival of children, characteristic of the Rooted Metropolitan Family. That said, higher incomes, as observed by Lin et al. (2018) seem to allow them to choose a lifestyle that is unencumbered, well connected to the workplace, or even in close proximity to it. These lifestyles are situated in urban settings less likely to cause car dependency.

4.3. Which dimensions should be targeted in the analysis and intervention on mobility?

37The analyses we have carried out allow us to propose various avenues for research and intervention in the area of mobility. Obviously, the lifestyles where we observe the greatest likelihood of choosing public transport should be targeted. They represent those with more urban characteristics, where conditions are most favorable to sustainable mobility. By being more flexible in terms of transport, they show that a wider range of mobility choices does indeed lead to choices in this direction.

38Compared to motorized workers who have settled in the suburbs or in the metropolitan peripheries, the three other lifestyles stand out. That said, two of them, the Unrooted and the Active and Flexible lifestyles, attract people at the beginning of their residential or professional career. That said, urban and flexible lifestyles may also turn to the car (Lotfi et al., 2018). More research on these specific logics would be relevant. Here, it seems appropriate to directly target individuals and households that have already experienced public transport use. The same logic applies to policies or programs that attempt to induce or encourage modal shifts from the car to more sustainable modes. Constructing proposals, subsidies, or benefits by targeting lifestyles that are more likely to change seems to be more promising than merely targeting socio-economic characteristics (e.g., age, incomes, etc.).

4.4 Towards modal choice dynamics

39This survey certainly considers many of the determinants of modal choice already observed in the literature (see especially section 1.1). Thus, functional, psychosocial, and demographic aspects have been clearly established in the model presented. It is regrettable that biographical elements and key events, as advocated by mobility biographies (e.g., Lanzendorf 2003), were not considered. The fact that people have moved does not affect modal choice in our regression model. Without going into the details of the results, it is nevertheless worth citing other work related to this same survey that indicates changes in travel behavior following the workplace relocation (Zarabi et al., 2019; Gerber et al., 2020; Ma et al., 2021). In addition, ICT use was not included in the survey, although it also elicits some differences in behavior (e.g., Arroyo et al. 2020 ; Olde-Kalter et al., 2021).


40For the case of the 1 977 MUHC workers located on the Glen site on the outskirts of downtown Montreal, we have explored in this article several categories of variables of mobility analysis that appear to be significant and useful for understanding mobility choices. The characteristics of living environments, their accessibility, and the characteristics of jobs are crucial here. Socioeconomic variables are only marginally significant, although they are structural in the exploratory logistic regression model developed. The lifestyles constructed based on residential-related identities make it possible to better understand the scope of the usual variables. Lifestyles appear to be strategic in the understanding of the mobility choice patterns of households, more so than socioeconomic variables. These variables should be better taken into account in transport plans and major urban projects, bringing the focus on people and their living environment rather than only on territories and their accessibilities.

41A vast majority of employees surveyed still use the car, probably because of accessibility constraints and atypical working hours, but also because of the dispositions induced by their lifestyle and living environment. Choosing a living environment is synonymous with choosing mobility, but lifestyles seem to predispose, to be an unavoidable intermediate variable. This observation is in line with several objectives and orientations of organizations working in planning and transportation in Montreal (e.g.: CMM, ARTM, STM), as in other metropolises around the world. Although daily mobility is central to many development and transportation policies, there is often little understanding and control of the real levers of action on this mobility, without even involving land use planning or urban development. Detailed knowledge of the factors that influence choices for a more sustainable daily mobility thus seems to require a better understanding of daily life and of the pathways, not of individuals who travel, but of several profiles of people and households who make mobility choices that are intertwined with other life choices, particularly residential ones.

Top of page


Arroyo R., Ruiz T., Mars L., Rasouli S., Timmermans H. 2020. Influence of values, attitudes towards transport modes and companions on travel behavior. Transportation Research Part F 71: 8-22.

Aarts H., Dijksterhuis A. 2000. The automatic activation of goal-directed behaviour: The case of travel habit. Journal of Environmental Psychology, 20(1), 75-82.

Amar G. 2010. Homo mobilis, le nouvel âge de la mobilité. Paris, Éditions FYP.

Ascher F. 2006 Le futur au quotidien. De la fin des routines à l'individualisation des espaces-temps quotidiens. In L'individu hypermoderne, Aubert N. (Ed.). Toulouse, ERES Sociologie clinique (273-290).

Barff R., Mackay D., Olshavsky R.W., 1982, A Selective Review of Travel-Mode Choice Model. Journal of Consumer Research 8(4): 370-80.

Beimborn E., Greenwald M., Jin X., 2003. Accessibility, Connectivity, and Captivity: Impacts on Transit Choice. Transportation Research Record 1835: 1-9.

Brisbois X. 2010. Le processus de décision dans le choix modal : importance des déterminants individuels, symboliques et cognitifs. Psychologie. Université Pierre Mendès-France — Grenoble II.

Bourdin A. 2005. La métropole des individus. Paris, Édition de l’aube.

Burke P.J., Stryker S. 2000. The Past, Present, and Future of an Identity Theory. Social Psychology Quarterly 63(4), 284–297.

Cervero R. Kockelman K. 1997. Travel demand and the 3Ds: Density, diversity, and design. Transportation Research Part D: Transport and Environment 2: 199-219.

De Witte A., Hollevoet J., Dobruszkes F., Hubert M., Macharis C. 2013. Linking modal choice to motility: A comprehensive review. Transportation Research Part A: Policy and Practice 49: 329–341.

Dupuy G. 2006. La dépendance à l’égard de l’automobile. Paris, Documentation française.

Énaux C. 2009 Processus de décision et espace d’activité / déplacements. Une approche articulant routine cognitive et adaptation évènementielle. Cybergeo [URL :].

Énaux C., Gerber P. 2014. Beliefs about energy, a factor in daily ecological mobility? Journal of Transport Geography 41: 154-162.

Ewing R., Cervero R. 2010. Travel and the built environment: A meta-analysis. Journal of the American Planning Association 76: 87-114.

Feldman R.M. 1990. Settlement-identity, psychological bonds with home places in a mobile society. Environment and Behavior 22(2): 183-229.

Feldman R.M. 1996. Constancy and change in attachments to types of settlements. Environment and Behavior 28(4): 419-445.

Fichelet M., Fichelet R., May N. 1970. Contribution à une psychosociologie des comportements urbains. Pour une approche écologique de l’utilisation des moyens de transport. Paris, COPEDITH, Délégation Générale à la Recherche Scientifique et Technique.

Gao J., Yamamoto T., Helbich M. 2022. The role of residential consonance and dissonance between couples in travel behavior. Transportation Research Part D: Transport and Environment 104: 103196.

Gerber P., Thériault M., Enaux C., Carpentier-Postel S. 2020. Links between attitudes, mode choice, and travel satisfaction: A cross-border long-commute case study. Sustainability 12(21): 9203.

Gerber P., El-Geneidy A., Manaugh K., Lord S. 2020. From workplace attachment to commuter satisfaction before and after a workplace relocation. Transportation Research Part F: Traffic Psychology and Behaviour 71: 168-181.

Godefroy F., Morency C. 2010. Estimating latent cycling trips in Montreal, Polytechnique Montréal.

Hägerstrand T. 1970. What about people in regional science? Papers of the Regional Science Association (24): 7-21.

Heinen E. 2016. Identity and travel behaviour: A cross-sectional study on commute mode choice and intention to change. Transportation Research Part F: Traffic Psychology and Behaviour 43: 238-253.

Jensen M. 1999. Passion and heart in transport – a sociological analysis on transport behaviour. Transport Policy 6(1): 19-33.

Juan S. (Ed.) 1997. Les sentiers du quotidien – rigidité, fluidité des espaces sociaux et trajets routiniers en ville. Paris, L’harmattan.

Kaufmann V., Bergman M.M., Joye D. 2004. Motility: Mobility as Capital. International Journal of Urban and Regional Research 28(4): 745-756.

Kesselring S. 2006. Pioneering mobilities: new patterns of movement and motility in a mobile world. Environment and Planning A 38(2): 269-279.

Lanzendorf M. 2010. Key events and their effect on mobility biographies: the case of childbirth. International Journal of Sustainable Transport 4 : 272-292.

Lesteven G. 2015. La congestion automobile perturbe-t-elle la routine des ménages motorisés ? Illustration à partir de l’Île-de-France. Espace populations sociétés [].

Lévy J. 2003. Capital spatial. In Dictionnaire de la géographie et de l’espace et des sociétés, Lévy J., Lussault M. (Eds.). Paris, Belin (125-126).

Lin T., Wang D., Zhou M. 2018. Residential relocation and changes in travel behavior: what is the role of social context change? Transportation Research Part A 111 : 360-374.

Lotfi S., Després C., Lord S. 2018 Are sustainable residential choice also desirable? A study of household satisfaction and aspirations with regard to current and future residential location. Journal of Housing and the Built Environment 34(1): 283-311.

Lucas K, Blumenberg E, Weinberger R. (dir.) (2011) Auto Motives: Understanding Car Use Behaviours. New Milford, Emerald.

Ma, T., Van Acker, V., Lord, S., & Gerber, P. (2021). Dissonance and commute satisfaction: Which reference point to use? Transportation Research Part D: Transport and Environment, 100, 103046.

Meissonnier J, Richer C (2015) Métro – boulot – dodo : quoi de neuf dans nos routines de mobilité ?, Espace populations sociétés [URL :].

Murtagh N., Gatersleben B., Uzzell D. 2012. Multiple identities and travel mode choice for regular journeys. Transportation Research Part F: Traffic Psychology and Behaviour 15(5): 514–524.

Nagelkerke N.J.D. 1991 A Note on a General Definition of the Coefficient of Determination. Biometrika 78(3): 691-692.

Papadimitriou E., Theofilatos A., Yannis G., Sardi G.M., Freeman R.P.J. 2012. Road Safety Attitudes and Perceptions of Pedestrians in Europe. Procedia - Social and Behavioral Sciences 48: 2490-2500.

Pradel B., Chardonnel S., Cailly L., Fourny M.-C. 2015. Les routines de déplacement dans les espaces périurbains : les dimensions collectives des agencements quotidiens. Espace populations sociétés [URL:].

Olde-Kalter M.-J., La Paix Puello L., Geurs K.T. 2021. Exploring the relationship between life events, mode preferences and mode use of young adults: A 3-year cross-lagged panel analysis in the Netherlands. Travel Behaviour and Society 24: 195-204.

Pritchard R. Frøyen Y. 2019. Location, location, relocation: how the relocation of offices from suburbs to the inner city impacts commuting on foot and by bike. European Transport Research Review 11(14) [https://doi:10.1186/s12544-019-0348-6].

Rau H., Manton R. 2016. Life events and mobility milestones: Advances in mobility biography theory and research. Journal of Transport Geography 52: 51-60 [].

Scheiner J., Holz-Rau C. 2013. Changes in travel mode use after residential relocation: a contribution to mobility biographies. Transportation 40: 431-458.

Schwartz, S. J., Luyckx, K., & Vignoles, V. L. (2011). Handbook of identity theory and research. Volume 1 Structures and Processes, Chapters 1–18. Volume 2 Domains and Categories, Chapters 19–39, Epilogue. (S. J. Schwartz, K. Luyckx, & V. L. Vignoles, Eds.). New York Dordrecht Heidelberg London: Springer Science + Business Media, Inc.

Steg L. 2004. Car use: Lust and must. Instrumental, symbolic and affective motives for car use. Transportation Research Part A: Policy and Practice 39: 147-162.

Twigger-Ross C.L., Uzzell D.L. 1996. Place and identity process. Journal of Environmental Psychology 16: 205-220.

Van Acker V., Witlox F. 2010. Car ownership as a mediating variable in car travel behaviour research using a structural equation modelling approach to identify its dual relationship. Journal of Transport Geography 18(1): 65-74.

Verplanken B., Walker I., Davis A., Jurasek M. 2008. Context change and travel mode choice: Combining the habit discontinuity and self-activation hypotheses. Journal of Environmental Psychology 28(2): 121-127.

Yang S., Fan Y., Deng W., Cheng L. 2019. Do built environment effects on travel behavior differ between household members? A case study of Nanjing, China Transport Policy 81: 360-370.

Zarabi Z., Gerber P., Lord S. 2019. Travel Satisfaction vs. Life Satisfaction: A Weighted Decision-Making Approach. Sustainability 11(5309) : 1-28.

Top of page

List of illustrations

Title Figure 1: Distribution of employees’ place of residence and main mode of transport
File image/jpeg, 311k
File image/png, 81k
File image/png, 12k
Title Table 3: Variables of residential-related identities
File image/png, 55k
Title Table 4: Results of principal component analysis (PCA)
File image/png, 75k
Title Figure 2: Dendrogram of the hierarchical cluster analysis
File image/png, 23k
Title Figure 3: Typology of residential-related identities
File image/png, 14k
File image/png, 91k
Top of page


Electronic reference

Sébastien Lord, Benjamin Lamoureux and Philippe Gerber, Metropolitan mobility options and individual identities: How much households' mobility choices are determined by residential-related identities?Articulo - Journal of Urban Research [Online], 23 | 2023, Online since 28 December 2022, connection on 23 April 2024. URL:; DOI:

Top of page

About the authors

Sébastien Lord

Sébastien Lord is Associate Professor at Faculty of Environmental Design at University of Montréal. He holds a PhD in Regional Planning, a Master’s Degree in Architecture from Laval University and a Bachelor’s Degree in Planning from the University of Montréal. His main research interests are: residential and daily mobilities, housing policies, demographic transition, and socio-spatial inequalities. E-mail:

By this author

Benjamin Lamoureux

Benjamin Lamoureux is a Master's graduate in Planning (M.Sc.) from the School of Planning and Landscape Architecture of the Faculty of Environmental Design at Université de Montréal. E-mail:

Philippe Gerber

Philippe Gerber is research fellow at Luxembourg Institute of Socio-Economic Research (LISER) in the field of geography, with an expertise in daily and residential mobility. He is developing research at Luxembourg Institute of Socio-Economic Research. The main objectives are modelling and simulating mobility behaviours considering urban constraints, especially in the context of cross-border regions. Email :

By this author

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search