Through a social space lens – Interpreting migration of the tertiary educated

Abstract

Higher education institutions are thought to rejuvenate and replenish local labour markets with recent graduates, especially in rural locales. While some graduates stay in the area of their alma mater, others return home or to other areas. The aim of this paper is to contribute a synopsis of research results from current peer-reviewed literature on the topic of internal (domestic) migration of the tertiary educated as well as introduce the concept ‘social space’ as a method for facilitating a relational interpretation of graduates’ migration propensity and location choices. Findings from this literature review indicate that regional characteristics, human capital and demographic factors are most often cited as affecting migration propensity and location choices among the tertiary educated. However, the interpretation of these factors is often taken in isolation and from economic perspectives. ‘Social space’, from four philosophical starting points, is discussed as an approach to understanding the interdependence of these factors. Examples of how higher education migration research can develop into different directions by using social space theory from different philosophical perspectives are also discussed.

As the facets and roles of higher education institutions (HEIs) have transformed to meet the needs of their environment, so too has the mission of education changed. Globalization has brought a shift in the mission of education from that which focuses on philosophy, science and the development of high culture to developing the capabilities of students to meet the needs of local surroundings, such as labour markets (Kerr, 1995; Standing, 2009). In Sweden, for example, university colleges were established to meet the growing needs of local labour markets and to educate individuals who reside in more rural locations. Later there was intent for these institutions to stimulate local economies and the growth of innovative ideas through the establishment of new businesses (Danilda and Granat Thorslund, 2011). However, data from the Swedish Council for Higher Education shows that tertiary educated individuals tend to move to Sweden’s largest cities before beginning their studies and upon completion if they studied outside the larger metropolitan areas (HSV, 2011).

The movement of higher education graduates has been of great interest to educational researchers in recent decades. However, while there is a growing body of literature on international migration of higher education graduates (Solimano, 2008), less attention has been paid to national migration of graduates (Hansen and Niedomysl, 2009; Wikhall, 2002). Thus, this literature review explores the latter theme with a particular focus on factors important to the propensity for migration and locational choices among higher education graduates.

Migration of the tertiary educated differs from other types of migration in terms of individuals’ motivations and destinations. Under the umbrella of migration, there are many different types, such as labour market migration among skilled professionals, refugee migrations, youth migration (e.g. those seeking experience and adventure) and migrations for love (King, 2002). Each of these migration forms implies differences in motivations and outcomes of the migration, although sometimes the categories are blurred. The distinction of migration among recent higher education graduates from other forms of migrants is their motivation to find a job that is reflective of their recently achieved education. Therefore, for many recent graduates, migration decisions after completion of their studies are tied to the arrangement of industries in the labour market, as well as characteristics of their social position (Bourdieu, 1985, 1989), such as their chosen field of study.

I begin by describing the methodology in how I found and selected the articles used in this review before thematically presenting the main results of these articles in terms of the factors deemed most important for determining migration behaviours of the tertiary educated. This paper identifies a lack of consideration of the interrelationships between these factors and individuals’ abstract social knowledge in the interpretation of the results. Social space is then suggested as a theoretical concept for identifying these interrelationships. Then, examples of how social space can be used to interpret the relationship of multiple factors influencing migration of the tertiary educated are provided from four philosophical starting points. This paper concludes with a discussion of the potential contributions of using this concept in educational research on migration of the tertiary educated.

This review serves to shed light on migration patterns of different groups of graduates and researchers’ interpretations of their migration patterns. In addition, areas where relational thinking (i.e. accounting for the interrelationships of multiple factors) could contribute to further understanding these resultant migration behaviours are highlighted. The intention is that this paper will contribute a synopsis of research results on higher education graduate migration as well as introduce the concept of ‘social space’, which stems from socio-spatial theory, as a method for facilitating a relational interpretation of migration from an educational perspective.

Methodology

In order to identify relevant and current peer-reviewed literature for this review, combinations of keywords were entered into social science peer-reviewed databases accessible through the University of Gothenburg libraries. Since graduate migration is an interdisciplinary topic, searching a broad base of journals and articles was important for obtaining a holistic understanding of the common themes uncovered in this research area. Different disciplines place varying amounts of emphasis on the influential factors of migration. Therefore, an interdisciplinary approach is necessary to avoid a biased perspective of migration. Keywords searched included higher education/college/university/tertiary education, migration/mobility, geography and graduate. The assumption was that such a process would generate relevant texts discussing or referencing these topics. The review also targeted the most recent texts (from the year 2000 to present) and focused on peer-reviewed journal articles written in English.

Through the search for articles it became evident that most articles addressing migration propensity and locational choice among the tertiary educated originated from economics and human geography fields. Moreover, while the type of methods used within the articles were not discriminated against, all articles used quantitative methods – either surveys or register data. The titles and abstracts of the search results were reviewed for relevance to the objectives of this review. This initial review resulted in identifying a moderate number of articles. Upon reviewing the reference lists for these articles, additional relevant articles were selected.

These reviews resulted in 18 articles that were selected for inclusion in this literature review, representing research in four European countries: Sweden (eight), Italy (two), the Netherlands (two) and the UK (six). This selection removed any articles where the focus was not on tertiary educated individuals or their decisions to migrate following their participation in higher education. Many articles included education as a variable but did not make higher education a focus of the article. In addition, several articles focused more on the outcomes of migration and regional employment rather than factors influencing graduates’ migration behaviour. The articles included in this review are not meant to be exhaustive but rather are meant to provide a sample of the findings on current higher education graduate migration research across four European countries.

Empirical review

Research on graduates’ migration has typically combined human capital migration models with spatial job-search models (Faggian et al., 2006). This combination suggests that migration is related to both individuals’ demographics and areal characteristics, such as economic and employment conditions, of their origin and destination regions. Within demographics, an individual’s human capital is often deemed the dominant determinant of migration behaviour with other demographic characteristics, such as gender, age, social class and civil status, also contributing to migration behaviours (Faggian et al., 2006). With regard to regional characteristics, local unemployment rates, industry, amenities, crime and salary levels often influence migration behaviours. Since human capital and spatial job-search models are often used in graduate migration research, three common themes emerged within the results of the reviewed articles – regional characteristics, human capital and individuals’ demographics. Therefore, a summary of outcomes from the articles is illustrated in the following sections, which are organized according to these themes.

Regional characteristics

The relationship between regional characteristics and migration among higher education graduates is examined in 15 of the studies reviewed in this paper. Regional characteristics found to be significant in attracting higher education graduates can be set into three categories: the local economy and employment, cost of living and wages, and relationships.

As for the local economy and employment, the Swedish literature cited work and career opportunities as the most important regional aspect to attract graduates (Niedomysl, 2008; Niedomysl and Hansen, 2010). Research from the Netherlands cited large labour markets as important (Venhorst et al., 2011). Studies from the UK took the research a step further by analysing the types of job markets graduates are drawn towards. They found that areas with higher numbers of public sector jobs tend to attract more higher education graduates (Faggian and McCann, 2009). On the opposite side, areas with locally specialized business sectors tend to not attract as many higher education graduates (Faggian et al., 2007b). Unemployment rates are also significantly related to the migration decisions of the tertiary educated (Faggian et al., 2007b; Venhorst et al., 2010) as are higher gross domestic products (GDPs; Nifo and Vecchione, 2014) and strong tax bases (Tano, 2014).

In terms of cost of living and wages, graduates generally move away from expensive regions (Venhorst et al., 2011) and areas with high wage gaps (Ciriaci, 2014) and towards regions where higher wages can be obtained (Faggian et al., 2007a; Nifo and Vecchione, 2014). In addition, higher education graduates like being near people – like-minded people and their relatives. Graduates are attracted to areas with high population densities (Faggian et al., 2007b; Tano, 2014) and areas where there are catchments of tertiary educated people (Faggian et al., 2007a; Tano, 2014). Furthermore, graduates are attracted to areas in proximity to their family, friends and relatives (Wikhall, 2002).

While most of these articles suggest a significant relationship between the regional context and migration behaviours among the tertiary educated, the study by Faggian et al. (2006) speculates on this relationship. They determined that regional context may be less important to the migration decisions of the tertiary educated than the general population because in general they have lower unemployment propensities. Venhorst et al. (2011) also found that regional amenities tend not to affect the migration patterns of graduates. However, Swedish studies have found mixed results in terms of amenities influencing migration behaviour. Niedomysl (2008) and Niedomysl and Hansen (2010) state that the tertiary educated have a preference for areas with good cultural and entertainment facilities, coastal areas and places with a view of the water, while they are less interested in healthcare and shopping. These observations suggest there are distinct differences between the migration patterns of the tertiary educated and the general population; an area that could be explored further would be to compare the preferences of higher education graduates with their peers who chose not to formally continue their education.

Human capital

Human capital in terms of education was discussed in all the articles. This dimension was stratified into four predominant themes that emerged in the literature: HEI type and quality, field of study, grades and employer relationships during education.

Within HEI type and quality, one way HEIs are often differentiated in terms of migration propensity among graduates is by their mission. Generally university graduates in the Netherlands have a tendency to migrate more often than college graduates (Venhorst et al., 2010, 2011). College refers to those institutions with a more regional or vocational mission. This was consistent in the UK as well. In Faggian and McCann (2009), a HEI’s membership in the Coalition of the Modern Universities (CMU) is used to infer institutional mission, since CMU institutions are more strongly focused towards their local economies. Their results showed that graduates of these institutions are less likely to migrate post-graduation. Institutional size in Italy also appears to have a role in graduate migration; attendance at mega-sized universities increases the likelihood of migration post-graduation (Ciriaci, 2014). Finally, in both Italian and UK studies, attendance at a HEI of high quality is found to reduce the probability of migration post-graduation (Ciriaci, 2014; Faggian et al., 2007a, 2007b).

As for field of study, there are mixed results in terms of migration propensity among graduates. In both the UK and the Netherlands, science graduates are more likely to stay in the same region as where they graduated (Venhorst et al., 2010). However, studying medicine in the UK promotes migration (Faggian et al., 2007a). Among university graduates from peripheral regions in the Netherlands, economics and agriculture graduates are most likely to move to another region, while healthcare and social and behavioural science graduates are more likely to stay in the peripheral region. Overall graduates in economics, law and agriculture tend to have the highest propensity towards migration (Venhorst et al., 2010). In Italy, a specialist degree increases migration probability (Nifo and Vecchione, 2014). Also, UK graduates with arts degrees, which are generally less specific to employment needs, tend to reduce migration propensity (Faggian et al., 2007b). What these mixed results imply is that the migration decisions of graduates from some fields are dominated by human capital while others are dominated by the labour market, meaning that in some fields the most capable graduates stay in competitive labour market regions as they are recruited by local employers and tend to have the first choice of opportunities available to them, while others are forced to migrate in order to find work based on the needs of the labour market (Faggian et al., 2007b; Venhorst et al., 2010).

The importance of the labour market is evidenced in Swedish studies involving field of study and graduate migration. For instance, the labour market is more important for civil engineers and media-communications graduates and is less important for graduates with specialist teaching degrees. In addition, interest in cultural activities can also vary by field of study. For example, Wikhall (2002) discussed how cultural facilities are most important for those with an artistic education and in media-communication fields. This is one example of how the field one studies can be representative of preferences that shape migration behaviours.

Educational achievement or grades can also determine graduates’ migration behaviour. Generally, graduates who earned good grades during their studies are more prone to migration (Ciriaci, 2014; Faggian et al., 2006; Nifo and Vecchione, 2014). In Sweden, high achievers choose more ‘well-off’ regions and if a high achiever lives in a region with a strong tax base, high population density and well-educated population, they are less likely to migrate. Actually, the top 25 per cent of achievers are less likely to migrate in comparison to the other quartiles if the person lives in a well-off region. This holds true for the UK as well, particularly among graduates from London (Faggian and McCann, 2009). High achievers move from university towns (outside urban areas) to urban areas or other towns (Tano, 2014). Lower achievers in Sweden tend not to live in urban regions but they do live in university towns and rural areas (Tano, 2014). Low-achieving graduates in the UK also have a greater tendency of being recruited to peripheral regions, and low-achieving graduates who did not initially move to attend higher education are more likely to be recruited locally (Hoare and Corver, 2010).

Finally, having completed an internship or having prior work experience facilitates migration away from the region from which individuals graduated (Venhorst et al., 2011). Employer sponsorship of studies also facilitates migration post-graduation (Faggian et al., 2007b). This might be opposite to what one might expect, but perhaps these relationships contribute to building one’s human capital levels, thus easing their spatial mobility.

Taken as a whole, the research shows that human capital in terms of higher education attainment and graduate migration are positively linked. However, the direction and nature of the relationship between graduate migration and the different stratifications within higher education is not entirely clear across all countries reviewed. Further research is needed to understand how the human capital and labour market models relate to the recruitment and migration of different groups of graduates.

Demographics

As for other demographics affecting migration behaviour, four were most often explored in the studies: gender and family commitments, ethnicity, age and previous migration. There were mixed results for the impact of gender across the four countries explored. In the Netherlands, UK and Italy, women are more prone to migration than men (Faggian et al., 2007b; Nifo and Vecchione, 2014; Venhorst et al., 2010). Venhorst et al. (2010) speculate that women migrate in order to gain access to more and better jobs and not due to being in a relationship.

However, studies of Swedish women with higher education show that they are less likely to migrate after completion of their studies than their male counterparts (Lundholm, 2007), but when they do migrate, they place more emphasis on family, communications (transportation), shopping, cultural opportunities and neighbourhood reputation when making a choice on location (Niedomysl, 2008; Wikhall, 2002) rather than on work or career opportunities (Niedomysl and Hansen, 2010). With reference to cultural opportunities, Niedomysl’s (2008) conclusion differs from Niedomysl and Hansen’s (2010), who say that culture and entertainment are less important in a location for women. They speculate that this is because women have less leisure time than men.

Several of the Swedish articles explored the influence of friends and family as an important pull factor to a region. Most references related to spouse/partner relationships. For instance, individuals in dual-earner relationships have a lower propensity to migrate (Branden, 2013; Lundholm, 2007), except for couples where both partners’ jobs are geographically ubiquitous (Branden, 2013). Generally, men’s careers facilitate migration, but when high-powered females are in a relationship with a spouse of a lower position they tend to have more influence on choice of location when migrating (Tano et al., 2014). Also, single individuals who enter high-powered positions at a young age are drawn to large labour markets, presumably to where other high-powered singles reside (Tano et al., 2014).

In addition, Swedish families with young children of a school age have a greater preference for living near family and relatives and a lower propensity to migrate (Branden, 2013; Niedomysl, 2008). However, this is contrary to an Italian study, which found that family commitments actually increase migration probability (Nifo and Vecchione, 2014). The researchers speculated that this may be due to a heightened need to ensure a steady income and support for the family. In Sweden, the highest migration propensity is among highly educated couples where neither lives in the region of their birth (Branden, 2013).

One Swedish article also considered the influence of income among higher education graduates who are in dual-earner relationships. The more women contribute to household incomes, the less likely migration becomes. However, if there are wage differences in her type of job between regions, migration propensity increases. On the other hand, this does not hold true for men (Branden, 2013).

Only the UK studies considered the role of ethnicity on graduate migration behaviours. They found that being non-white decreased the likelihood of migration (Faggian and McCann, 2009; Faggian et al., 2006, 2007b). As for age, there was consensus that older age reduces migration post-graduation (Ciriaci, 2014; Faggian et al., 2006, 2007b; Nifo and Vecchione, 2014). Older graduates may be less migratory because they may have already built-up networks in the region of their studies or began working simultaneously while they were studying. In the Netherlands, older university graduates are more likely to move to central regions if they graduated from a HEI in the periphery (Venhorst et al., 2010). In addition to these characteristics, individuals who have a history of migrating, for instance moving to attend higher education studies, are more apt to migrate again when they graduate (Ciriaci, 2014; Faggian et al., 2006; Hoare and Corver, 2010; Nifo and Vecchione, 2014; Tano, 2014).

Empirical discussion

From the reviewed literature, many important regional, human capital and demographic factors are identified as affecting migration behaviours among the tertiary educated. However, the influence of these factors is often considered in isolation when in actuality these factors are interconnected with individuals’ social knowledge and perceptions (e.g. knowledge from prior social experiences or perceptions of broader societal ideologies that guide behaviour). The human geographers who have typically studied graduate migration tend to focus on regional attributes and the geographical arrangement of labour markets in terms of their economic appeal to graduates as explaining migration. Some have addressed job ubiquity in relation with educational choices and the arrangement of the labour market. For instance, the degree programme and field of study students pursue have implications for the possible job locations that are available to them. Students who pursue agriculture or marine biology are limited by regional landscapes and resources, while graduates from interdisciplinary disciplines, for instance, can find employment in many different labour market contexts. However, these links between field of study and the labour market are often related back to economic explanations by inferring that graduates choose to migrate to locations that are most economically beneficial for the degree they pursued. What the researchers fail to acknowledge are the overarching cultural and social associations (Bourdieu, 1985, 1989) of these economic explanations.

Local economies and labour markets contribute to the spatial division of labour, where work is often distributed and implied as either men’s or women’s work (Massey, 1994). Regional norms and culture often develop under these economic conditions, which can create social, political and cultural climates appealing to certain segments of the population. This appeal relates to individuals’ social knowledge and perceptions, which suggests explanations for the perceived value of a man or woman moving to a particular place.

In relation to the gendered regional labour markets, educational pathways are also gendered. Prior research indicates that young men and women typically choose different fields of study and specializations (UNESCO, 2012). However, the stratification of educational pathways between the genders is not accounted for in the reviewed literature. Recognizing these initial gendered educational choices and their relationship to social perceptions and knowledge and the arrangement of jobs in the labour market is important to understanding graduates’ choice of region with which to establish themselves and of the potential benefits they might reap from migrating to a different region post-graduation. None of these elements individually can explain migration; they must be considered in combination with each other.

This discussion implies that current research in the field of graduate migration often does not take a relational approach to understanding migration propensity and location choice. Rather this research has a tendency to perceive direct observations as reality instead of taking into consideration the relational aspects of geography, human capital and demographics. The examples above describe areas where relational thinking could be used to further understand the basis behind graduate migration beyond direct economic or geographical explanations. A social space theoretical perspective, which recognizes the interdependence of competing factors and abstract social knowledge, is thus proposed as a theory that could be used in research on graduates’ migration propensity and location choices to further understand the relationships among the many factors influencing their behaviours.

Considering a social space perspective

Social space is part of a broader spatial turn in educational research (Middleton, 2014) and was first introduced as a concept by Emile Durkheim in the 1890s (Buttimer, 1969). Since the establishment of the concept, it has been modified in a number of ways by different theorists. Theorists who have contributed much to the development of the concept include Pierre Bourdieu, Michel Foucault, Henri Lefebvre and British realists such as Andrew Sayer, each of whom have taken different philosophical positions towards social space – structuralist, genealogical, Marxist and realist, respectively (Shields, 1991). Some research has already addressed the potential contributions and use of social space theory in the educational discipline generally. For instance, Robertson (2010) used the Marxist social space perspectives of Henri Lefebvre and David Harvey to ‘spatialize’ educational markets and pathways through education. However, social space theory is also appropriate for educational research on graduate migration, as it draws connections between the geographical and the social.

In general, social space offers a way to understand the relationship of physical and social environments, as well as abstract social structures, and individuals’ perceptions thereof. The combination of these features produces and reproduces social status and broader power relationships in society, including migration trajectories. The social space perspectives outlined above offer different approaches for achieving a relational understanding of graduate migration. The following are examples of how graduate migration research can develop in different directions by using social space from differing philosophical perspectives. The examples below also discuss relational thinking in terms of the different social space perspectives.

The structuralist approach to social space is based on the view that there are objective cultural structures in the social world, independent of human will or consciousness, that impose on individuals’ perceptions of reality. This standpoint relates human agency to structure. Bourdieu, a major contributor to the structuralist approach to social space, suggests that the social world is constructed by individuals’ ‘habitus’, that is, their perceptions, thoughts and actions. Therefore, the construction of social spaces is guided and constrained by societal structures through cultural and economic capital (Bourdieu, 1989) as well as other principles of division, such as ethnicity or gender (Bourdieu, 1985). In terms of graduate migration, behaviours are determined by an individuals’ perception of the value of a particular move. This perception is assumed to be shaped by larger structural forces in relation to the graduate’s position in society. For example, a highly educated young man who originated from a rural area may choose to establish himself in a metropolitan area because he perceives societal expectations of earning a high income in preparation for being a household breadwinner. He moves because he perceives greater opportunities to fulfil this expectation in large city regions. Relational thinking is evidenced by considering perceptions of gender roles, as defined by society, to elaborate upon economic explanations for establishing oneself in a particular geographic location.

The genealogical perspective assumes individuals behave according to common principles formed by earlier knowledge and communicative exchanges that were prevalent in their lives. These exchanges articulate appropriate roles and positions for different individuals in society. In this sense, social space is a fundamental element of an individual’s relationship to the world (Shields, 1991). With regards to the highly educated young man, he would move to the metropolitan area based on his personal history and knowledge of work opportunities appropriate for his status and position in society, which were gained by communicating with those nearest to him. From these communicative exchanges he would infer greater opportunities outside his rural hometown. In this case, the relation between personal background and the communication of social norms about what is appropriate and what is not for different individuals in a particular society is stressed in terms of graduates’ choice of migration.

The focus of the realist perspective, specifically from Sayer’s point of view, is on the contingent, causative role of physical space in social relations (Sayer, 1992). That is, to understand social stratification across different physical spaces, a consideration of the spatial configuration of these physical spaces is needed (Sayer, 2000). These characteristics could include distance and proximity of physical objects, areal differentiation (Sayer, 2000) and the particular social structure of the physical space (e.g. labour market areas) (Sayer, 1992). With regards to the example of the young man, he would move to the metropolitan area based on a myriad of spatial configurations related to the metropolitan area. Some examples include the metropolitan area’s close proximity to his place of origin or the high relevance of the industrial structure of the labour market within the metropolis to characteristics of the young man’s social position, such as his education and skillset. These spatial configurations unique to the metropolitan area, in relation to his social position, provide context for his migration decision. Thus, the realist perspective is relational in the sense that it relates various spatial configurations of a physical space to contextualizing the social behaviour of individuals, such as their migration tendencies, within and between physical spaces.

Marxist perspectives such as Lefebvre’s combine two elements to define social space – physical or geographical elements and social elements; however, the focus of the Marxist perspective is on how these elements merge to produce different outcomes for individuals based on power relationships and social relations. Thus, individuals behave in ways that are most rewarding with regards to their particular social space. Essentially, this ideology suggests social space is an abstraction of its contents and its value to society (Lefebvre, 1991). In reference to the example of the young man, his choice to move to the metropolitan areas would occur as a result of the intermingling of several elements in his life, including his daily routines and physical surroundings, his social position in society and his conception of reality through his lived spatializations (such as in the genealogical perspective). These elements combine to produce his social space, which influences his choices and behaviours. His choice to migrate and live in the metropolis then reflects his particular standing in social space and the situation most advantageous for his position in social space. Relational thinking is exhibited in the Marxist perspective of social space by connecting geographical (i.e. physical) space, social relations and personal history or background.

These four social space perspectives provide different approaches for understanding the relationship of social and geographical elements as they influence graduates’ migration behaviours and choice of residential location. Incorporating a social space approach in migration research of the tertiary educated would extend the commonplace economic explanations for migration to include social explanations. Acknowledging social explanations provides reasoning for why certain groups of graduates may perceive the cost or benefit of residing in a particular location differently when economic conditions are held constant. Social space therefore provides a lens for recognizing and understanding the interrelationships of multiple factors, including regional characteristics, human capital and individuals’ demographics. Consequently, social space promotes much-needed relational thinking in the analysis of graduates’ migration behaviours and choices.

Relational analyses of higher education graduate migration are particularly relevant for an educational perspective to research on this topic. While the relationship of sociality and geography on migration after higher education is important and has been the focal point among economic and human geographers, educational research is usually directed towards investigating the political, social and cultural institutions as well as human dynamics associated with education and education outcomes (Ranis and Walters, 2004). Therefore, in terms of educational research, a social space theory that places greater emphasis on the social than the geographical may be preferable.

Lefebvre and Harvey are commonly cited socio-spatial theorists who centre on geographical and physical space (Reed-Danahay, 2015). Conversely, Bourdieu’s social space has much to contribute to a socio-spatial perspective of education and has often been overlooked in research by geographers (Bridge, 2003) and by social science researchers where ‘social field’ has been emphasized and the differences between social field and social space have been blurred (Reed-Danahay, 2015). Geographical or physical space is not Bourdieu’s primary focus (Creswell, 2002). Instead, he centres on sociality in relation to geography and often draws connections with education. In addition, a key tenet across Bourdieu’s theories is his promotion of relational thinking (Bourdieu, 1989). Thus, there is much work that can be done to develop an educational perspective of higher education migration using social space theory, particularly from the perspective of Bourdieu.

Concluding remarks

This literature review illustrates that while regional characteristics, human capital and demographics are common themes explaining migration behaviours among the tertiary educated, there is room to develop relational perceptions of these explanations by recognizing their interrelationships. In addition, the research did not address broader abstract social knowledge such as country-specific or macro-level social contexts from which society, and graduates specifically, shape their perceptions of places, regions and opportunities. Furthermore, research on higher education graduate migration has primarily been pursued by economic and human geographers, as indicated by the prevalence of articles from such researchers. Naturally, this has implications for the lens through which they interpret graduates’ migration behaviours. An educational approach to exploring issues in higher education graduate migration, while incorporating socio-spatial theory, is much needed in order to bring fresh perspectives and challenge familiar economic explanations so pervasive in this research area.

Given that socio-spatial theory is still an emerging trend in educational research, there is much these perspectives can contribute to the study of education in general and higher education graduate migration specifically. The introduction of social space theory in educational research would be especially advantageous in broadening the research focus of educational spaces from an emphasis on their surrounding regional geography and economic situation and the physical boundaries (i.e. architectural boundaries) of HEIs to incorporate relations with sociality. Higher education migration research using social space as a theoretical lens could, for example, contribute to understanding how the stratified relationships among students in HEIs, in terms of their backgrounds and choice of study programme, reflect their social space positioning, which is then translated to their migration propensity and location preferences upon completion of their studies.