Between stuckness and stillness: Why do young adults not undertake temporary mobility?

Funding information Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Grant/ Award Number: 100017_165726 Abstract Youth temporary mobility occurs for purposes such as volunteering, Erasmus exchanges and linguistic stays. Although it is increasingly common, a large proportion of young adults are not mobile. This study is based on a large-scale survey among young adults in Switzerland. It draws on the concept of motility to analyse the barriers to temporary mobility, where motility may be defined as a set of mobility resources that refer to three dimensions: access (e.g., financial means or time), skills (e.g., languages and self-confidence) and appropriation (e.g., level of interest). A typology places nonmobile young adults on a continuum between ‘stillness’ (no desire to move) and ‘stuckness’ (unable to move) and identifies four groups: ‘the constrained’ and ‘the financially challenged’ are often constrained to a varying extent by their socio-economic background, their educational pathways and their family network. ‘The locally anchored’ and especially ‘the satisfied stayers’ face fewer constraints and draw more on their own agency to be nonmobile.

The literature on spatial mobility and on temporary mobility has neglected nonmobility and explained it as the consequence of constraints. Research and policies tend to pay little attention to the absence of action, as human agency is often recognised in terms of action (Schewel, 2020). However, scholars have started to address the agency of nonmobile individuals . Nonetheless, explaining nonmobility (or the nonoccurrence of any event) is more complex than explaining mobility because nonmobility can result either from constraints (such as limited finances or time) or from a choice driven by the desire to stay or by other plans. Until now, the literature has failed to differentiate groups of nonmobile individuals and to demonstrate that nonmobility also results from individual agency. However, this topic is highly relevant as societal pressures to become mobile are increasingly criticised (Mincke et al., 2019). In the next sections, we develop a theoretical framework based on Kaufmann's (2016) concept of motility to differentiate groups of nonmobile individuals and to classify them on a continuum between 'stillness' and 'stuckness' (Cresswell, 2012). A first analysis identifies the sociodemographic factors that differentiate the (non)mobile young adults; we then focus on the barriers to temporary mobility stated by young adults and create a typology of nonmobile young adults. We then conclude and discuss the findings.

| Motility and nonmobility
The literature on temporary mobility has paid little attention to nonmobile young adults and to the factors preventing them from being mobile (for an exception, see Souto-Otero et al. (2013) on Erasmus and non-Erasmus students). Given this lacuna and the many questions that remain open with respect to temporary mobility, our theoretical framework is informed by the more general literature on various spatial mobilities.
As conceptualised in the migration literature, nonmobility is not a decision taken at one specific point in time but is the result of multiple influences and events over time (Carlson, 2013;De Jong & Gardner, 1981;Hollywood, 2002;Williams et al., 2018). As  and Hjälm (2014) state regarding residential moves and migration, nonmobility is renegotiated repeatedly throughout the life course. Several terms are used to describe nonmobile individuals; they either relate to a specific type of mobility, such as non-migrants or stayers (Fernández-Carro & Evandrou, 2014) or are loaded with negative connotations (left behind, rooted and tied to place) (Barcus & Brunn, 2009;Jónsson, 2011). Another commonly used term is immobility (Schewel, 2020). However, immobility is an overarching term denoting a lack of movement in general, while in this article, we are interested in the nonoccurrence of a specific type of mobility, that is, temporary mobility. We define this as nonmobility. Because not engaging in temporary mobility does not necessarily mean that individuals do not engage in other forms of mobility, such as residential or day-to-day mobility (e.g., commuting), we prefer the term nonmobility over immobility.
Concerning nonmobile individuals, scholars have attempted to separate the factors related to a choice to remain nonmobile from those related to constraints (Coulter et al., 2016;Rérat, 2014;. More recently, Carling and Schewel (2018) proposed the 'aspiration-capability framework', a two-dimensional framework that considers the capability to be mobile as one dimension and mobility aspirations as another dimension. Both dimensions need to be considered jointly to explain why individuals become mobile or not.
In this article, we build upon Kaufmann et al.'s (2018) motility framework. Motility represents a 'set of characteristics that enables people to be mobile, including physical capacities, social conditions of access to existing technological and transportation systems, and acquired skills' (Kaufmann et al., 2018, p. 199). Motility represents a mobility potential which may or may not be activated and is composed of three dimensions, each of which may explain why an individual is mobile or not: access (being able to be mobile), skills (knowing how to be mobile) and appropriation (wanting to be mobile). These three dimensions are used to structure the article and to explain the barriers to temporary mobility.
In addition to motility, we build on Cresswell's (2012) definition of nonmobility as a continuum between 'stillness' (choosing not to move) and 'stuckness' (being prevented from moving). We assume that stuckness results from a lack of access or skills, whereas stillness is more likely due to a lack of appropriation or interest in temporary mobility.
However, interest in temporary mobility also depends on habitus and therefore on the integration of social norms. Rye's (2011) analysis of the 'structured freedom' concerning rural migration allows us to understand how interest depends on cultural and economic capital.
Rye draws on Bourdieu's concept of habitus (Bourdieu, 1979) and analyses the link between social class, social constructions and individuals' migration plans and preferences. Habitus is a disguise for individuals, hiding the social mechanisms for interest in a particular phenomenon. We assume that this may also apply to temporary mobility. Some social groups may have internalised the norm of nonmobility and young adults within these groups may therefore be less motivated to move and may experience stuckness, as their low interest may stem from (unconscious) constraints. We define stillness as nonmobility that does not result from constraints but from an individual choice.

| Barriers 1 to temporary mobility
Research has shown that young adults from higher social backgrounds are more likely to be temporarily mobile (Findlay et al., 2012;R. King et al., 2011;Waters & Brooks, 2010). This may be because parents with higher education (often from a higher social class) are more likely to value temporary mobility for their children's personal development and employability (Carlson et al., 2017).
Furthermore, mobility experiences provide in themselves the opportunity to develop skills that enable further engagement in temporary mobility, and so children from higher social classes are more likely to have those skills. Young adults from lower social classes tend to have fewer resources (access) to enable mobility (Findlay et al., 2006;Williams et al., 2018) and so are less likely to be temporarily mobile and less likely to develop the skills involved.
Finally, as mentioned above, interest (appropriation) can depend in part on social norms and therefore on social class. Thus, social class relates directly to the three components of motility.
Education is a major enabler of temporary mobility (Smith et al., 2014). In particular, higher education offers more access opportunities: compared with their peers who have entered the labour market, university students usually have more time available (e.g., during holidays) and more opportunities (through programmes such as Erasmus) (Findlay et al., 2006;Souto-Otero et al., 2013).
Moreover, temporary mobility experiences may be more valued as a part of their career paths (Frändberg, 2015;Holdsworth, 2017).
Education can also provide the opportunity to acquire better language skills, which relates to the skills dimension of motility.
Additional skills are also required to become mobile, such as self-confidence and not being afraid or reluctant to stay in another environment (A. King, 2011).
Motility, as well as concepts of mobility capital or spatial capital (Lévy, 2000;Murphy-Lejeune, 2003;Rérat, 2018;Schäfer, 2020), implies that mobility is a capital that may be accumulated through experiences (migration, temporary mobility, holidays) that favour future mobility (e.g., by learning to cope in different contexts). In this sense, mobility reproduces itself (Carlson, 2013) and a lack of experience may lead to a lower propensity to engage in temporary mobility. Whether an individual becomes mobile also depends on their wider familial context (Mulder, 2007) and social network (Beech, 2015;Souto-Otero et al., 2013). Related barriers include being reluctant to leave their family, friends or partner behind or a lack of support (Cairns, 2014;Kehm, 2005). Therefore, understanding nonmobility requires taking into account individuals' histories by considering different kinds of (non)mobility.
Finally, barriers to temporary mobility may differ by gender.
Female students are more likely to be temporarily mobile (OFS, 2017a) and are overrepresented among Erasmus students (Böttcher et al., 2016). Deakin (2014) shows that women put more emphasis on issues of language or cultural discovery. Gender differences concerning the barriers have so far been given little attention by the research.

| Data and definitions
This study focuses on Switzerland, a small country located in the centre of Europe that has a high GDP per capita. Its four linguistic regions and its highly globalised economy make it important for young adults to have a good command of several languages to succeed in the labour market. 2 There is a high proportion of apprentices and a relatively low proportion of university students compared with neighbouring European countries: a third of the population has completed an apprenticeship, and only 18% has completed a tertiary degree (OFS, 2017b). Swiss students are among the most mobile among OECD countries: eighth out of 35 (OECD, 2018).
We use secondary data from the 'Swiss Federal Surveys of Adolescents' 3 (FORS, 2020), which have a history dating back to the 19th century. Each edition of the surveys focuses on a topic of relevance to young adults. The 2016-2017 edition covers life-course trajectories and mobility experiences and looks at past and planned temporary mobility as well as addressing the absence of mobility among some young adults and the reasons put forth for their (non)mobility.
In Switzerland, reversible mobilities such as temporary mobility are significantly more popular among young adults than irreversible mobility (moves to another region or country) (Stam & Rérat, 2019).
The data were collected in two stages by means of a paper questionnaire. The first part of the data collection took place during the recruitment procedure for a basic military service; this is a particularity of Switzerland in which all Swiss men between age 18 and 20 4 have to participate in a procedure that assesses their potential fit for a mili- The second part of the survey involved sending the same questionnaire to a representative sample of 2126 18-and 19-year-old Swiss women. 6 The territory was separated into regions and the questionnaires were sent to experts in each of these regions, who were responsible for administering the questionnaire to a representative sample of young women drawn from the registry. This sample allows us to make comparisons with the male population and to assess the generalisability of the results. To this end, in the part of the analysis that includes men and women, the population of men is weighted to obtain a similar sample size to that of women.
Only young adults with Swiss nationality were recruited, and so foreigners were excluded. Yet a large proportion of young Swiss adults have a migration background: 37% of the respondents have at least one parent born abroad, and therefore, based on the dataset used, we can draw some conclusions on the influence of migration history on temporary mobility.
In the survey, temporary mobility refers to both past and planned sojourns without parents, which take place abroad or in another linguistic region of Switzerland. A temporary move to another linguistic region of Switzerland is considered an experience of living in another linguistic and cultural context, and thus is an addition to the existing literature, which usually pays little attention to intranational mobility in multilingual countries (such as Belgium, Canada or India). Stays in another linguistic region may not have a purely touristic purpose: they can be educational (62% and 71% of past stays carried out by men and women), professional (10% and 12%) or cultural (28% and 17%).
The survey distinguishes short (between 1 and 3 weeks) and past experiences of a longer duration (more than 3 weeks). The cut-off of 3 weeks was used in the questionnaire given the age group considered here and the Swiss context: students in high schools have around 6 weeks (depending on the canton) of summer holidays, and apprentices usually have around 5-week holiday per year.
Due to the young age of the respondents, planned stays of at least 3 weeks in the three following years are also addressed. The sample is subdivided into four groups: (1) specific mobility plan, (2) wish to be mobile but without a specific plan, (3) uncertain mobility (may or may not become mobile, depending on opportunities) or (4) no plans (nonmobile). 7 This article focuses on the last group, representing 32.6% of men and 15.5% of women (Table 1). A logistic regression will identify a potential gender difference in terms of probability of becoming temporarily mobile when all other variables are controlled for.

| Sociodemographic factors and (non)mobility
The first part of the analysis focuses on the sociodemographic factors that may influence nonmobility as identified in the literature (Table 2).
We categorise the variables along the three components of motility.
The respondents' professional status and educational level influence access (e.g., opportunities) and skills (e.g., languages). The number of countries visited provides an indication of the willingness to be mobile (which is linked to appropriation) and of the skills. We also take into account whether respondents are in a relationship, which is expected to be negatively related to their willingness to become temporarily mobile. In addition, we control for respondents' age and linguistic region, which capture differences in cultural characteristics and in educational systems within Switzerland. Finally, we explore social class through the parents' highest level of education and the financial situation in childhood.
We compare nonmobile young adults to their mobile counterparts (defined as either having been mobile or having a plan to do so) in terms of sociodemographic factors in a logistic regression. 8 Logistic regressions assess whether each factor is statistically significant in predicting nonmobility all other things being equal. This analysis includes 37,000 male respondents (93% of the full population 9 ). We then validate the results for women (N = 1873 for past stays and N = 1865 for planned stays) with a similar logistic regression model that includes the female sample and the weighted male population (N = 4252).

| Stated barriers to temporary mobility
In the second stage, we zoom in on the nonmobile participants and compare the barriers to temporary mobility put forward by the men to those mentioned by the women. Respondents who do not plan to be mobile were asked about the extent to which different barriers applied to them on a 5-point Likert scale. 10 Table 3 shows the barriers in the questionnaire and how we operationalised them into the different motility components. Barriers related to access include financial resources and time available, whereas skills include language skills, self-confidence and (not) being afraid of going abroad. Finally, appropriation is defined by willingness to leave their friends, family or partner 11 and by the extent of personal or professional interest in temporary mobility.
Finally, we standardise the nine barriers and apply a cluster analysis using Ward's method to create a typology of nonmobile young men (N = 10,363; 80% of the nonmobile respondents 12 ). We then compare each group to the rest of the nonmobile population and apply four logistic regressions using the sociodemographic factors (each group is successively compared with the others) ( Table 2). We exclude the sample of nonmobile women, which is too small for this detailed analysis.
On the whole, these analyses explore the sociodemographic factors influencing (non)mobility and the stated barriers, in order to obtain a broad picture of nonmobility between agency and constraints, stillness and stuckness.  perceptions of mobility when entering the labour market. Yet individuals who have already entered the labour market are more often nonmobile than those who are still in education. This suggests that they face professional constraints, such as not being able to take time off work. The financial situation during childhood is not significantly related to mobility, which may be due to the imprecise and retrospective nature of this indicator.
A higher number of languages spoken, as well as having visited more countries, increase the propensity to mobility, whereas being in a relationship increases the likelihood of being nonmobile. Age is not significant because we consider past and planned mobility experiences jointly. Finally, there are differences between linguistic regions of Switzerland. German speakers are the least mobile, followed by Italian and French speakers. This can be explained by a greater incentive to learn a foreign language for the latter two linguistic minorities. The sample of women reveals that they are more mobile than men all things being equal. 13 The significance of the other variables remains similar, even though standard deviations are larger due to the smaller sample size. The factors influencing young men's propensity to be mobile are valid for women as well.

| Stated barriers to temporary mobility
We now address the barriers towards mobility among those who do not become mobile. Table 5 shows the means for the barriers as reported by nonmobile young men and women.
Overall, barriers related to access are rated quite high, especially by women. The most important one is difficulties around getting leave from studies or work. Skills are less frequently referred to by both men and women, although a social desirability bias cannot be excluded. Lack of self-confidence is less frequently mentioned by men. The most significant barriers are those related to appropriation, which are gender dependent. Reluctance to leave friends, family or partner is more important for women, whereas noninterest (personally or professionally) is more common among men.

| Typology of the nonmobile young men
Based on an analysis of the dendrogram, four groups have been identified according to the importance of the barriers to temporary mobility ( Table 6). As noted earlier, this analysis cannot be done on the sample of women because of its small size. However, we elaborate on nonmobile women at the end of Section 4.3 based on the gender difference shown in Section 4.2.
In the next sections, each group is described according to the deviations of each barrier from the mean of the group compared with the mean of the nonmobile young males. A score above average means that the barrier is more important for that group than for the others, whereas a lower score means that the barrier is less important.
In order to determine the profile of each group, we use logistic regressions where each group is compared with the rest of the nonmobile population using the same sociodemographic factors as in the first analysis (Table S1).

| The constrained
The constrained contains 22% of the nonmobile males (N = 2542) and regroups those with the highest constraints or, in other words, the lowest motility or the highest stuckness ( Figure 1). Compared with the mean, the barriers related to all three motility components are more strongly expressed. Access is more difficult (particularly with regard to financial resources), and skills are a particular barrier (lower selfconfidence and worry to live abroad). They also lack interest in temporary mobility and are more strongly anchored locally (reluctance to leave friends/family/partner).
These young adults are less likely to have university-educated parents, which reveals a link between a lower social class and T A B L E 5 Barriers to planned mobility for the nonmobile young men and women (5-point Likert scale, 1 = strongly disagree, 5 = strongly agree) constraints to temporary mobility. They are also more likely to be both in education and working at the same time (e.g., apprenticeship), implying that they have less free time and less opportunities to be mobile. Finally, they speak fewer languages and have visited fewer countries.

| The locally anchored
The locally anchored are the largest group (38%; N = 4335). They less frequently cite barriers related to access or skills, and lack of self-confidence is not important (Figure 2). Barriers related to appropriation are, however, more important. They have a strong local anchorage and are reluctant to be separated from their social network. On the one hand, this might mean that they feel in the right place and have made an independent decision not to become temporarily mobile; on the other hand, it might imply that their social network does not encourage temporary mobility.
The group also differs in its profile: they are more likely to be in a relationship, and they place high value on their existing social ties.
This group corresponds to a situation between stillness and stuckness.
Access and skills do not represent important barriers, but although their nonmobility may be a personal preference, it may also be the consequence of a social network that does not encourage temporary mobility.

| The satisfied stayers
The satisfied stayers (22%; N = 2394) are defined by a weak interest in temporary mobility (Figure 3). They give below-average importance to most of the other barriers in comparison with the nonmobile population, mention financial problems less than the other groups, and their social network does not restrict mobility. Therefore, they do not face strong constraints, and they draw on their own agency to be nonmobile.
F I G U R E 1 Deviations from the mean regarding the stated barriers of the 'constrained' in comparison with the nonmobile young men F I G U R E 2 Deviations from the mean regarding the stated barriers of the 'locally anchored' in comparison with the nonmobile young men Concerning their profile, they are more likely to be in secondary general education, to be single and to speak more than one language.
In fact, their profile seems favourable towards temporary mobility, further supporting the impression that nonmobility results from their agency and is voluntary. Therefore, the group corresponds most to a situation of stillness.

| The financially challenged
The Several characteristics explain their nonmobility. Although they have a better knowledge of languages, have visited more countries and are less likely to be in an apprenticeship, they are more likely to report a modest financial situation in their childhood. This lack of financial resources may have prevented them from undertaking a temporary mobility experience. Hence, despite expressing fewer barriers, this group represents a case of stuckness due to financial constraints.
Based on young women's barriers, we may expect similar groups but with different sizes. Nonmobile women seem to be more constrained by access and their social network, but they are more interested in temporary mobility. Therefore, we expect fewer women in the group of the satisfied stayers, and more women in the other groups, particularly the constrained and the locally anchored. Temporary mobility is a selective process in relation to social class (Findlay et al., 2012;R. King et al., 2011), education and professional status (current and future university students are more mobile) and gender (temporary mobility is more popular among women). In addition, individuals who have been exposed to different contexts (including on holiday with their parents) seem to have developed a habit of mobility (appropriation) and learned to get along in different contexts (skills). Some nonmobile individuals are constrained by a lack of access or discouraged by their social network, and others simply choose not to be mobile.
Nonmobile young adults are not a homogeneous group, and the barriers they face represent a continuum ranging from stillness (not wanting to be mobile) to stuckness (not able to be mobile). Our analysis reveals four groups of nonmobile young men expressing different barriers and degrees of motility. The constrained are the closest to a situation of stuckness, with a low level of motility. They are the group least interested in mobility, although this is partly linked to social class.
This may be interpreted through Rye's (2011) perspective of 'structured freedom': in addition to the conscious barriers in terms of access and skills, the constrained may have developed a habitus that involves a lack of interest in temporary mobility. This may result from the internalisation of social norms (depreciating mobility and valuing local ties) and a lack of encouragement by parents and their social network.
The financially challenged are also predominantly in a situation of stuckness, even though they have a higher level of skills and appropriation than the other nonmobile groups. Their constraint stems from their economic capital, rather than their cultural capital (education and language). The locally anchored are somewhere in between stillness and stuckness: their barriers mainly refer to their local anchorage. Finally, only the satisfied stayers seem to be in a state of stillness as they would have a sufficient level of motility to experience temporary mobility.
Their nonmobility can be considered as an expression of agency in the sense that they want to remain nonmobile (Schewel, 2020), although this can be renegotiated during their life course.
The results highlight the importance of considering nonmobility not only scientifically but also for policy making. On the one hand, a large proportion of young adults face inequalities that prevent mobility (lack of financial means, time or opportunities and worry to live abroad or to leave friends and family). Therefore, policies aimed at promoting temporary mobility should focus on the various barriers leading to stuckness, identified in terms of access, skills and appropriation. The promotion of temporary mobility could involve grants, better information about opportunities, giving apprentices and young professionals more free time, and so on. On the other hand, despite societal pressures (Mincke et al., 2019), there are some young adults who do not want to be mobile. They should also be considered by political actors. Equality of access to temporary mobility is an important policy objective, but temporary mobility should not become an obligation.
Moreover, an absence of temporary mobility does not necessarily imply that people are immobile. They may have a strong mobility potential but have different aspirations.
Our study highlights some important avenues for future research on nonmobility. First, our study focuses on temporary mobility. It would be interesting to assess whether the same differences would also be found with respect to irreversible forms of mobility such as a move to another region or country. In addition, it would be valuable to compare the four groups later in their life course and to establish a longitudinal perspective. Moreover, echoing Schewel (2020), forced movers (without aspirations) should also be empirically identified. Quantitative data have shown us the prevalence of barriers among young adults, but a qualitative approach would allow us to identify relationships between variables and assess the causality and the hierarchy of factors leading to nonmobility. Finally, the internalisation of social norms linked to (non)mobility requires further ethnographic research. advice, as well as the editors and reviewers for their comments which greatly helped to improve the paper.

CONFLICT OF INTEREST
We do not have any conflicts of interest.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are openly available in  (Cresswell, 2014), constraints or barriers. In our analysis, we use the latter, which represents a broader definition.
2 Nearly 40% of the working population use at least two languages weekly (OFS, 2018). 3 Note that in this article, we use the word young adults because respondents are mostly between 18 and 20 years old. We distance ourselves from the word 'adolescents' used in the official translation of the survey. We acknowledge that the phase of youth can also be defined in different ways (UNESCO, 2017) and covers a wider age range.