Student Characteristics and Barriers to International Mobility: Evidence from the European Union

Abstract We use Eurobarometer data to examine barriers to international student mobility. Multivariate analysis is employed to study how individual characteristics are related to the obstacles preventing higher education students from participating in activities in another EU country. The results suggest that several demographic factors including area of residence, gender and age are associated with attitudinal, informational and financial barriers. These findings may help in the identification of measures to make international student mobility more inclusive.


Introduction
There is a growing number of studies pointing out the benefits that may result from participating in international mobility activities.While being abroad, participants have the opportunity to acquire a wide array of knowledge and skills that will enable them to successfully compete in the labor market.Many works show how studying and working abroad can enhance employment prospects and increase earnings potential (Deakin, 2014;Di Pietro, 2012, 2015, 2022;Kratz & Netz, 2018;Pinto, 2022).An experience abroad can also support people's personal growth and development (Zimmermann & Neyer, 2013).It may boost their self-confidence, may help them to become more independent and increase their adaptability to new situations (Bal� a� z & Williams, 2004).National governments and international organizations strongly encourage young individuals to participate in international mobility programmes.In a recent European Union Council Resolution (i.e., 2021/C 66/01), the need for continuing to expand international student mobility is stressed.
However, while expanding international mobility programmes, it is important to ensure that efforts are made to promote social inclusion and the participation of individuals with special needs or with fewer opportunities/resources (European Commission, 2014).Several papers show that more vulnerable people (e.g., people with mental and physical disabilities, minorities, people from more disadvantaged backgrounds) are less likely to take part in international mobility programmes (Di Pietro, 2020;Johnstone & Edwards, 2020;Van Mol & Timmerman, 2014).In order to ensure that the growth of participants in these programmes goes hand in hand with equal opportunities, one should attempt to identify and remove barriers deterring people from living an experience abroad.
In the literature, two distinct lines of research have been adopted to quantitatively study factors preventing students from going abroad.The first provides descriptive statistics on the relevance of students' perceived barriers to participation in study abroad programmes (e.g., Heirweg et al., 2020;Kmiotek-Meier et al., 2018;Wanger et al., 2020) or examines how the relative importance of specific barriers does vary between study abroad participants and non-participants (e.g., Souto-Otero et al., 2013;Sulejmanov et al., 2021).This investigation allows to assess the importance of different obstacles to international student mobility including, for instance, financial constraints, insufficient foreign language skills, lack of information about study abroad opportunities, personal obligations, work-related commitments, etc.Nevertheless, this approach does not give any indication about the traits of those students who are more likely to face these specific barriers.Such information, however, would be very useful as it would help policymakers to develop appropriate measures specifically targeted to these students in an attempt to encourage their participation in international mobility activities.For instance, who are the students more likely to report that they do not have the necessary foreign language skills to be able to participate in international mobility programmes?Are these students more likely to come from more disadvantaged backgrounds?Are they more likely to be older or younger students?Are they more likely to be female or male?
By contrast, the second line of research uses regression results on differences in individual characteristics between internationally mobile students (or students who plan to study abroad) and their non-mobile peers (or students who do not plan to study abroad) to infer about study abroad barriers (e.g., Netz, 2015).However, the interpretation of these results may not be straightforward.There may be more than one barrier associated with the same individual characteristic.For example, students from low socio-economic status may be less likely to take part in international mobility programmes because they have insufficient foreign language skills, because they lack the financial resources to study abroad, or because they lack information about study abroad opportunities.Furthermore, there is also the risk that misleading conclusions are drawn if different individual characteristics are not simultaneously considered in the analysis.This occurs when two closely related individual characteristics are associated with different barriers.For example, while having children and being in employment are likely to be correlated (i.e., students with children are more likely to have a job in order to support their family), they are associated with different barriers to study abroad participation (i.e., personal obligations and work-related commitments).Therefore, if one of these individual characteristics is not accounted for in the analysis, there is the possibility that a wrong barrier would be identified.Finally, including all categories of non-participants in international mobility programmes in the same analysis may be misleading.One should be able to separate students who are not interested in studying abroad from those who are keen to do so but face one or more obstacles.These two groups are different from a policy perspective, with the former perhaps needed to be informed about the expected benefits from going abroad.The importance of this distinction has been highlighted in a recent paper by Haldimann et al. (2021).They classify non-mobile young adults into two categories: 'stillness' (no desire to move) and 'stuckness' (unable to move).
This paper adopts a novel approach to look at the issue of international mobility barriers by bringing together these two lines of research.Using data from a recent Eurobarometer survey, a large cross-national individual-level survey conducted on behalf of the European Commission, we use multivariate analysis in order to examine how individual characteristics are related to the different barriers.The advantage of such approach is that it enables to gain a better understanding about who are the students primarily affected by each of these barriers, and this may help toward the identification of targeted measures that may ensure a more inclusive access to experiences abroad.We focus our attention on students who have not taken part in international mobility programmes and are able to consider various types of barriers that are commonly cited throughout the study abroad literature.Additionally, we study separately students who report not to be interested in going abroad and students who indicate specific reasons that prevented them from participating in international mobility activities.
The remainder of the paper is organized as follows.It starts with a description of the dataset, the variables used in the study, and the different hypotheses emerging from the literature that are empirically tested.This is followed by a presentation of the empirical results.Finally, some conclusions are presented together with their policy implications.

The Dataset
We Only respondents who report not to have taken part in any of the above activities (i.e., (f)) are asked about the reasons that prevented them from doing so.These reasons include: "not interested", "insufficient language skills", "lack of financial means", "lack of information about possibilities to go abroad", "my parents/family/partner/friends are discouraging me from going abroad", and "lack of opportunities that would match my interests".Respondents could select more than one option.
One issue with the above question is that it does not specify any time period during which respondents had the opportunity to undertake a mobility experience in another EU country, but they did not use it.Ideally, one would want to have such information to be able to link the reasons behind the nonparticipation of respondents with their individual characteristics observed during this time period.In an attempt to address this problem, and given that respondents' individual characteristics are recorded at the time of the survey, we decided to focus our analysis only on respondents who are tertiary education students.The rationale for this choice is that, since a vast array of international mobility programmes are typically offered by tertiary education institutions, it can be expected that young people have a higher chance of going abroad especially during tertiary education.As claimed by Smith et al. (2014), education is a crucial enabler of temporary mobility and, in particular, higher education does offer more access opportunities (Haldimann et al., 2021).Our choice of focusing on tertiary education students appears also to be supported by the fact that "Studying, training or apprenticeship in another EU country" (i.e., (a)) is the most popular answer to the aforementioned question on the different types of international mobility experiences.For instance, the Erasmus programme, which accounts for most of the short-term mobility in Europe (Song & Kim, 2022), gives many tertiary education students the opportunity to study, train and work in another European country.

Variables
In addition to the tertiary education student's country of residence, the survey collects also information about other individual characteristics such as age, gender, immigrant status (i.e., students who do not have the nationality of the country they live in), area of residence (rural area/village, small/medium sized-town, large town/city), employment status (part-time, full-time or not in employment), marital status (i.e., married or cohabiting, single) and the presence of dependent children.These characteristics are relevant because, as shown in the next section, the literature suggests that they may be related to some of the barriers preventing tertiary education students from taking part in international mobility activities.
In this paper, these individual characteristics are regressed against a series of dummy variables, representing different reasons behind the nonparticipation in international mobility programmes.Given the binary nature of our dependent variables, we use both logit models and linear probability models.Country-fixed effects, which account for time-invariant country specific factors, are included in all the models.They pick up all of the average differences in the dependent variables across countries and mitigate the problem of omitted variable bias (Mårtensson et al., 2023).
Table 1 shows the descriptive statistics of all the variables used in the empirical analysis.Our sample consists of 4,019 observations.In line with several previous cross-country studies (e.g., Beerkens et al., 2016;Orr et al., 2011), financial barriers appear to be the major obstacles preventing tertiary education students from participating in international mobility activities.More precisely, 41 percent of the students included in our sample report that they were unable to go abroad because of lack of financial means.

Lack of Interest
The human capital theory predicts that older students are less likely to participate in international mobility programmes because they will be able to take advantage of the benefits of this experience (for instance in terms of higher earnings potential) for a shorter time period compared to their younger peers (Netz et al., 2021).For this reason, mature students may be keen to finish their higher education studies quickly and they may perceive a mobility experience abroad as delaying their graduation (Taylor & Rivera, 2011).Pope et al. (2014) provide evidence showing that younger students are significantly more likely to intend to study abroad than their older peers.Some studies suggest that interest in studying abroad can also be related to gender, as female students are more likely to see the opportunity to take part in overseas mobility programmes as a positive and (Presley et al., 2010) valuable experience.(Stroud, 2010) and Tompkins et al. (2017) find that women are more likely to express an interest in developing an understanding of other countries and cultures.Girls are also more likely than boys to take the risk of living and studying overseas (Raj, 2022).Goldstein and Kim (2006) argue that the reasons for women's higher interest in studying abroad can be found in them being less ethnocentric, more open-minded and less intercultural communication apprehensive relative to men.Using a sample of students from Oregon State University, King and Young (1994) show that almost twice as many men than women reported not to be interested in studying abroad.
Additionally, having a partner and having dependent children may both discourage students from undertaking a mobility experience abroad.Students may be reluctant to leave their partner and their children for a long time.Naffziger et al. (2008) show that if students feel that social obligations (e.g., "missing significant other", "significant other would not want me to go") are too great they are less likely to consider studying abroad.Similarly, the research by Cairns (2014) also emphasizes the role of family relationships in discouraging international mobility.
Finally, it is also possible that students living in urban areas have more incentives to participate in international mobility activities than those living in rural areas.This is because, as stated by Porter and Porter (2020), the benefits of an education experience abroad are perceived to be more relevant in terms of future career opportunities by the former relative to the latter.Compared to students from rural areas, students from urban areas are more likely to see themselves pursuing a career that rewards those who have had an international experience.Additionally, Li et al. (2017) argue that students from rural communities are more likely to have parents with little or no international exposure who tend not to encourage their children to study abroad.This argument is also in line with the finding that people living in rural areas are less likely to have positive views on the EU relative to those living in large towns ( € Oz &Van Praag, 2023).
Hence, the following hypotheses are proposed.
Hypothesis 1: Female students are more likely to show an interest in participating in an education experience in another EU country than their male counterparts.
Hypothesis 2: Older students are less likely to be motivated to study, train or do an apprenticeship in another EU country than their younger peers.
Hypothesis 3: Students with family responsibilities (i.e., married/cohabiting students or students with dependent children) are more likely to have a negative approach toward international mobility.
Hypothesis 4: Students from rural areas are less likely to be keen to study, train or do an apprenticeship in another EU country compared to those from urban areas.

Lack of Financial Means
There is some evidence suggesting that concerns about the cost are more likely to dissuade women than men from engaging in an experience abroad.Shirley (2006) finds that female students are more likely to indicate the cost of studying abroad as a barrier preventing them from participating in international study programmes.Similarly, Presley et al. (2010) show that female students are more concerned about money and cost factors when considering the possibility of studying abroad.
On the other hand, older students may be less likely to lack the financial means to undertake an international mobility experience compared to their younger counterparts.Pimpa (2003) concludes that mature students tend to be less financially reliant on their parents to study abroad.Similarly, working students may be more likely to afford an education abroad experience as they may have some savings that can be applied toward the cost of living abroad.
Therefore, it is hypothesized as follows.
Hypothesis 5: Financial barriers are more likely to discourage female students from participating in an education experience in another EU country relative to their male counterparts.
Hypothesis 6: Older students and those with a job are less likely to have financial problems that could prevent them from studying, training or doing an apprenticeship in another EU country.

Insufficient Foreign Language Skills
It is often suggested that women are better at foreign languages than men (e.g., Banegas & Govender, 2022).A possible explanation for this is that female students are more likely to study humanities, social sciences, and general education subjects including foreign languages (Speer, 2017).Furthermore, King and Young (1994) find that among the reasons reported by students who have no plans to study abroad, anxiety about learning a foreign language was more frequently cited by men than women.
Several studies (e.g., Hansen-Thomas et al., 2016;Izquierdo et al., 2021;Samanhudi, 2019) indicate that rural school students face a lot of challenges in learning foreign languages.This is related to the characteristics of the schools, teachers and parents.Many rural schools lack appropriate educational resources and facilities.A large number of their teachers do not have the credentials to teach a foreign language.In rural communities, there is only a small proportion of parents who are able to help their children in learning a foreign language.Finally, it is also possible that students from small villages are not good at foreign languages due to a lack in motivation.They may perceive learning a foreign language to be neither necessary nor required in their immediate environment (Botturi et al., 2018).
The great majority of immigrant students are multilingual as they tend to be exposed to two languages (i.e., language of their host country and language of their home country).Results from PISA 2018 data show that students with an immigrant background are more likely to report that they speak more than one language and are learning foreign languages at school (Covacevich & Vargas, 2021).
The following hypotheses are therefore proposed.
Hypothesis 7: Compared to students from urban areas, those from rural communities are more likely to identify lack of foreign language skills as a barrier preventing them from studying, training or doing an apprenticeship in another EU country.
Hypothesis 8: Compared to female students, male students are more likely to consider their foreign language skills as inadequate to participate in an education experience in another EU country Hypothesis 9: Compared to native students, students with an immigrant background are less likely to perceive a lack of foreign language skills as an obstacle for international mobility.
Lack of Information about Possibilities to Go Abroad Chieffo (2001) as well as King and Young (1994) find that older students are more likely to be aware of study abroad programmes.These students are likely to have more friends at university, which is an important source of information about study abroad programmes.Similarly, older students are also likely to be in contact with more professors and advisors, which is another valuable source of information about international mobility.
Students from rural areas and small villages may not have enough information to make an informed decision about taking part in international mobility programmes.Again, it is possible that these students lack direct contact with people who participated or are knowledgeable about study abroad programmes.The results of the Eurodesk 2022 survey (Eurodesk, 2022) show that young individuals from urban areas are more likely to have friends who undertook an education abroad experience relative to those from rural areas.Vossensteyn et al. (2010) conclude that in Finland students in the capital region are more aware of the Erasmus programme and its expected impact on participants, compared to those living in more rural regions.
The following hypotheses are thus formulated.
Hypothesis 10: Older students are more likely to have access to information on opportunities to study, train or do an apprenticeship in another EU country, compared to their younger peers.
Hypothesis 11: Compared to students from urban areas, those from rural communities are more likely to lack information about the possibility of participating in an education experience in another EU country.

Empirical Results
Table 2 shows the logit estimates for the relationships between individual characteristics and the different barriers to international student mobility.As reported at the bottom of such a Table, country-fixed effects are consistently highly statistically significant.This result supports their inclusion in all the regressions.We use the VIF (Variance Inflation Factor) to check whether there is a multicollinearity problem among the independent variables of our model.The findings of this analysis indicate that there is no multicollinearity problem as the VIF values are all well below the critical value of 10 (Hair et al., 1995).In all the regressions, the likelihood ratio chisquare (LR chi2) and its p-value indicate that our model as a whole fits significantly better than an empty model (i.e., a model with no predictors).The results do provide support for all our hypotheses except three.We find that lack of interest in EU international mobility activities is higher among male students compared to female students, thereby confirming H1.Specifically, men are found to be 11 percentage points more likely not to be interested in doing an education abroad experience than women.Additionally, consistent with the human capital theory, this interest seems to decline with students' age, thus lending support to H2.The estimates show also the relevance of home ties as a discouraging factor, hence supporting H3.However, only being married or cohabiting (and not having dependent children) is found to increase the likelihood of not being interested in a mobility experience in another EU country.Finally, students living in large towns/cities appear to be more motivated than their peers living either in rural areas/villages or in small/medium-sized towns.However, while this finding is line with H4, it needs to be interpreted cautiously as both the relevant coefficients are only marginally statistically significant (at the 10 percent level).
Moving on to financial barriers to international mobility, in line with H5, our results indicate that affordability is a more important concern for women than it is for men.Our estimates lend also support, although only marginally statistically significant (p-values between 0.05 and 0.1), to H6. Older students and those with a job are found to be less worried about the cost of a mobility experience in another EU country than younger students and those without a job, respectively.Surprisingly, the results show that students with dependent children are less likely to report financial problems as an obstacle preventing them from going abroad.Although the reason for this finding is unclear, one should note that parenting students applying for a scholarship/grant to study or work abroad (e.g., through the Erasmus programme) are often eligible to receive additional funding if they take their child or children with them.
Estimates shown in Column (3) of Table 2 provide support for H7 as they suggest that, other things being equal, lack of foreign language skills is a bigger barrier for students living in rural areas or villages than it is for students living in large towns/cities.On the other hand, our results do not support H8 given that there is no statistically significant gender difference in the probability of reporting insufficient foreign language skills as a reason for not going abroad.Similarly, there is no support for H9 given that this probability is also not found to statistically differ between native students and students with an immigrant background.
As for the lack of information about opportunities to study, train or do an apprenticeship abroad, consistent with H10, we find that this obstacle is more relevant for younger students than it is for older students.By contrast, our results do not support H11 as the probability of indicating lack of information as a reason for not participating in international mobility programmes does not statistically differ between the three areas of residence.
Estimates shown in Column (4) of Table 2 show that, unexpectedly, students with dependent children are found to be more likely to be aware of international mobility opportunities.This result would seem to go hand in hand with our earlier explanation about why parenting students are found to be less preoccupied about the cost of an international mobility experience.Not only may students with dependent children be more likely to know about the existence of international mobility programmes, but they may also be aware that they are entitled to receive extra funding if they are awarded a scholarship/grant.Finally, although earlier we did not formulate any hypothesis about gender differences in access to information about possibilities to go abroad, our results suggest that the information obstacle is more important for female students than it is for male students.While the reason behind this finding cannot be identified here, it is possible that women may be more likely to perceive than men that the available information about international mobility programmes is insufficient to make an informed decision.This argument is consistent with the idea that women tend to search for more information and need more time before they are ready to make a decision (Gill et al., 1987).
In order to check the robustness of our results, two additional analyses are performed.First, we use a linear probability model instead of a logit model.As argued by Angrist (2001), the issue of causal inference does not significantly vary between limited dependent variables and continuous outcomes.This implies that if there are no covariates or the covariates are sparse and discrete, then linear models can be employed to estimate models with limited dependent variables as well as models with other types of dependent variables.As shown in Table 3, linear probability estimates are very similar to our logit estimates.Second, we re-estimate our logit model but this time we cluster the standard errors at the EU country level.This allows us to account for the possible dependence among observations within countries.These estimates, which are reported in Table 4, are similar to those shown in Tables 2 and 3.There are, however, two exceptions.The first one concerns the effect of area of residence on the probability of not being interested in international mobility activities.Whilst in Tables 2 and 3 the coefficients on area of residence are both found to be marginally statistically significant, results depicted in Table 4 suggest that there is a statistically significant difference (at the 5 percent level) only between students living in rural areas or villages and those living in large towns/cities.The second exception is related to the effect of area of residence on the probability of reporting lack of information as a reason for not participating in international mobility programmes.Estimates shown in Table 4 go against H11 as students from rural areas or villages are found to be marginally less likely to face informational barriers than their peers from large towns/cities (p-value ¼ 0.09).

Conclusions
In this paper, we attempt to shed light on the individual-level factors behind barriers preventing young EU people to take part in mobility activities in another UE country.Data from a recent Eurobarometer survey are employed.This survey was carried out in 27 EU Member States in the first months of 2022.We focus our attention on tertiary education students.
Our results consistently show that, compared to students living in large towns/cities, students living in rural areas or villages are less likely to have an interest in participating in international mobility programmes, and when they do have it, they do not seem to face financial constraints but language barriers.Students from rural communities are more likely to believe than their peers from urban areas that they do not have adequate foreign language skills.This result has relevant policy implications in light of the EU's commitment to promote equal opportunities for young people in urban and rural settings.One of the EU's eleven European Youth Goals is called 'Moving Rural Youth Forward' and includes an aim to 'ensure equal access to high quality education for young people in rural areas' (Lodeserto, 2020).More recently, in the recommendations of the Conference on the Future of Europe, citizens highlighted the importance for the EU and its Member States to 'establish by 2025 an inclusive European Education Area within which all citizens have equal access to quality education and life-long learning, including those in rural and remote areas'. 1  Our findings would therefore seem to suggest that there is a need for (additional) measures designed to encourage young people in rural areas to participate in international mobility programmes.As many of these students may not know what to expect from an education abroad experience, possible actions may include hiring special tutors to discuss study abroad opportunities and the value of study abroad, as well as using local communication channels to promote the benefits of study abroad.At the same time, it would be important to prepare students from rural areas for this experience, paying special attention to their foreign language skills (e.g., providing free access to online language courses, increasing the number of qualified teachers in foreign languages in rural schools, promoting the establishment of more foreign language centers in rural areas).
One should also consider that several studies (e.g., Di Pietro, 2015; Waibel et al., 2018) conclude that students from less advantaged family backgrounds (in terms of parental education and occupation) are likely to benefit more from an education experience abroad than those from more advantaged family backgrounds.Given their poorer backgrounds, these students are less likely to be able to develop many marketable skills (such as, for instance, intercultural competence, global awareness, and foreign language skills) if they do not participate in international mobility programmes.This suggests that, since students from rural areas are more likely to have parents with a lower educational and occupational level than their urban peers, there 1 Conference on the Future of Europe, 'Report on the Final Outcome', May 2022, p. 88. https://www.europarl.europa.eu/resources/library/media/20220509RES29121/20220509RES29121.pdf could be considerable gains in increasing the international exposure of the former group of students.
Our estimates indicate also that younger students are more likely to lack information about international mobility opportunities than their older peers.This outcome stresses the importance of informing students early in their studies about the possibility to participate in an activity abroad.They need to be informed about the likely advantages of this experience as well as about available scholarships/grants to finance their stay abroad.This would help interested students to carefully plan their mobility so that they can go abroad without adding time to their studies.
The results of our empirical analysis show also that EU female students exhibit a higher motivation to live an experience in another EU country than their male counterparts.Such a finding is consistent with the well-known fact in the study abroad literature that women participate in international mobility programmes more often than men (Cordua & Netz, 2022;Di Pietro, 2022;Van Mol, 2022).Possible provisions in an attempt to encourage male participation include the creation of more international mobility opportunities in male-dominated subject areas (e.g., Engineering and Computer Sciences) as well as the expansion of short-term study programmes and internships abroad.
While, as argued above, EU female students are found to be keener on engaging in an activity in another EU country relative to their male peers, the former appear to be more likely to be held back by cost considerations and the lack of relevant information compared to the latter.These findings would seem to indicate that ensuring that students have access to detailed information about mobility experiences around the EU (including information on available scholarships/grants) is especially important for women.
One limitation of our analysis is that the results cannot be generalized to international student mobility given that only activities within the EU are considered.While intra-EU mobility accounts for a large share of international student mobility in the EU (Campus France's Study Department, 2020), EU students may also go outside the EU.Another shortcoming of our study is that the sample size is too small to perform meaningful analyses at the level of the individual EU Member States.Finally, due to data limitations, we were unable to include relevant individual characteristics (e.g., socio-economic status) in the empirical analysis.
use data from Flash Eurobarometer 502 (Youth and Democracy in the European Year of Youth), which was carried out in 27 EU Member States between the end of February and the beginning of March 2022.This survey targeted people aged 15-30.Such Flash Eurobarometer was conducted at the request of the European Commission's Directorate-General for Education, Youth, Sport and Culture to mark the 2022 European Year of Youth.