The Complex Identities of Latvians Abroad: What Shapes a Migrant’s Sense of Belonging?

Building on the data from The Emigrant Communities of Latvia survey, this chapter aims to reveal and describe the complex nature of the sense of geographic and social belonging of those who became emigrants. It will explore the factors affecting the maintenance of their identity and consider transformation processes among migrants. The starting point of this chapter is the assumption that the ‘sense of belonging’ is affected by a variety of different objective and subjective factors, and that identity is multi-faceted. Instead of facing a trade-off between feeling close to the home country or host country, or developing a supranational identity, different combinations of types of a ‘sense of belonging’ can be distinguished among Latvian migrants. Using cluster analysis, the author distinguishes four separate groups of emigrants based on their attitudes and self-identification. Among the findings is that most respondents feel closer to Latvia than their host country. However, people who left Latvia during the years of the Great Recession and its aftermath, and who left for economic reasons, are the most alienated from their home country. Among the factors important in shaping this sense of belonging are subjective life satisfaction and having friends and family back at home.

belonging among Latvian emigrants, and to explore the factors affecting their identity maintenance and transformation processes.

Theoretical Framework
The concept of belonging is used to explore relationships between the self and society for several reasons: it is oriented towards the individual, used in daily lives that are saturated with formal and informal relations, and allows the establishment of a perspective on complex relations between the self and society while capturing changes (May 2011). Research into belonging helps uncover identification processes; not so much identity itself, rather the result of it. Both these aspects are closely related and are hardly separable in daily thinking. Bisley (2007) extends the notion of identity to those of being and belonging; to the ways in which individuals make up an image of themselves. In his view, the process of identity formation involves individual cognitive mechanisms as well as political and economic forces that promote certain ideas about existence and belonging.
In diaspora research the concepts of identity and belonging acquire new contexts and meaning. As with concepts of nationalism and globalisation, belonging and migration initially seem to contain a contradiction. Operationalisation of the concept of belonging within the context of national identity and territory leads us to such concepts as 'rootedness', 'state of peace', 'balance' and 'traditionalism', while 'migration' is related to mobility and postmodernity, and the conditionality and uncertainty of borders that comes with globalisation. Nowadays, both belonging and migration have acquired a new meaning that calls for a new conceptual approach (Hedetoft 2004). During the past decades, these approaches have changed and shifted the focus towards action, individual activities and the meaning of choice. In the context of migration, it means that migrants actively choose and shape their identities. Identities are seen as life projects, while recognising that such processes never end. Moreover, belonging is not something that a person commits to just once and keeps forever. The world we live in is constantly changing, as are the people living in and adapting to that world, and 'belonging' is achieved through an active process: in other words, belonging can be imagined as a trajectory in time and space (De Certeau 1984).
Identities develop within a social context and are shaped by culture, time and place obstacles. They can overlap or conflict (Tajfel 1981;Roccas and Brewer 2002). In the case of migrant identities, Mieriņa and Koroļeva (2015) find that the correlation between the sense of belonging to Latvia and the new host country is extremely small (−0.06), which means that there is no conflict between these identities: they can co-exist.
Vanessa May (2011) points out that the concept of belonging is used to research relationships between the self and the society for several reasons. First, it is directed towards the person. Second, it is used in daily life when formal and informal relationships intertwine. Third, it allows for a complex view of the relationship between self and society. Finally, the dynamics of such relationships allow the capturing of social change (May 2011, p. 364). In The Emigrant Communities of Latvia survey, the identity characteristics of Latvians living abroad were based mainly on measurements of the strength of belonging to various social groups, communities, places and particular territories.
The strength of a 'sense of belonging' can vary. It can be weak or strong, firm or fragile, and this makes it possible to use quantitative measurements of its intensity. Due to these aspects, the belonging approach is widely used in sociological research and is applied to this project as well. Hedetoft (2004) points out four aspects of the belonging analysis that determine identity in several ways.
These aspects are: the source of belonging; the sense of belonging (shaped by socio-psychological necessities, identification with place and memories); the construction and institutionalisation of belonging; and the variability of belonging. All these intertwine in the process of forming a sense of identity.
The sense of belonging is never immediate or truly 'pure'. It always passes a mental process through personal and collective experience, over time and through psychological 'memory filters', each of which shapes individual images and perceptions of belonging and gives them depth and value. In this way new forms of belonging in emigration take shape: through interaction between memories, experience, future plans and opportunities (Hedetoft 2004). Life in Latvia as well as in emigration can be seen as a source for the formation of belonging.
The sense of belonging to social groups, communities and places -whether one's own or experienced in migration -is an a priori individual subjective feeling. Therefore, it is affected by a range of psychological or subjective features which can influence the intensity of different levels of 'belonging'.
The circumstances under which a migrant leaves their home country can impact on the affective feelings towards the home country and the host country (Mieriņa and Koroļeva 2015). In part, these circumstances are reflected in the particular year of emigration or the wave of emigration (see Hazans in this volume). Identity as well as attachment to the local community continues to develop with time. The migrant's personal success in the new host country can also be expected to affect their sense of belonging (Gustafson 2005;Zepa and Kļave 2011). Having economic and social ties in the home and host countries are other major factors that can affect migrant attachment to a community or country -particularly the social ties (Mieriņa and Koroļeva 2015;Ros 2010). Finally, attitudes towards state institutions both in the home and host countries are linked to attachment to the country itself (Mieriņa 2015).
The formation of a sense of belonging is a complex process, often inwardly complicated. A person who left their motherland and settled in another country searches for new objects of belonging. A fresh sense of belonging develops in relation to the people and places in their new life as they overcome the internal contradictions between wanting to be similar and belong to the 'others' and wishing to remain unique and different from them. The sense of belonging to a group and place provides an ontological sense of security and, as May (2011) notes, it bears a mostly positive connotation. If belonging is understood as a sense that helps an individual cope with the surrounding world, then it is inevitable that non-belonging can be characterised as a burden (May 2011, p. 373). Non-identification with a place can lead to negative consequences and create the feeling of detachment and emptiness. It facilitates the development of such symptoms as longing for home, depression, desolation and an unbearable feeling of emptiness.
The formation of the sense of belonging and its transformation in emigration has emerged as a significant research theme, because preservation of the sense of belonging to the motherland and the formation of new feelings towards the new land of residence would determine to a large extent the success of integration, as well as acting as an influence on a person's decision whether to settle in the new country or return home (Ķešāne 2011). However, previous studies have operationalised migrants' identity predominantly in terms of their belonging to a particular place or group of people (Alonso and Oiarzabal 2010;Brewer 1991), overlooking the fact that identity is multi-faceted, and various combinations of attitudes towards their home country, host country or the global community are possible. This study aims to fill this gap in the knowledge and shed new light on the complex identities of Latvian migrants. It also explores a diverse set of factors that impact the formation of the sense of belonging.

Data and Methods
We used The Emigrant Communities of Latvia weighted survey data to analyse the sense of belonging and characterise a person's identity. Our definition of 'emigrants' in this chapter included all ethnic Latvians and Latvian nationals outside Latvia regardless of the year of emigration, ethnicity or citizenship (n = 14,051; for detailed information on survey methodology and the design of statistical weights see Goldmanis (2015) and Mieriņa in this volume).
Drawing from the previously described theoretical insights, we formulated the following hypothetical assumptions regarding factors upon which the sense of belonging and identification depends: • Background factors: the length of time a person has lived abroad, the wave of emigration, the aim of that emigration, the occupation in the host country, the relation to family and friends in Latvia, the social ties in Latvia and abroad and whether or not the migrant had a property in Latvia; • Subjective factors: satisfaction with different areas of one's life, feelings of trust and attitude towards the institutions of Latvia and the host country and the reasons for emigrating in the first place.
A five-point Likert scale was used for measuring responses characterising a sense of belonging (to ten social and territorial groups), including the categories 'I feel strongly that I belong', 'I tend to feel that I belong', 'I neither feel I belong nor do I feel I do not belong', 'I tend to feel that I do not belong', 'I feel strongly that I do not belong'. The sense of 'closeness of belonging' to a specific territory was measured on a four-point scale in categories such as 'very close', 'close', 'not very close', and 'not close at all'. The Emigrant Communities of Latvia study explored notions of belonging in three circles: the primary circle, formed by family ties and contacts with family members; the secondary circle of friends and the tertiary circle of contacts with groups in society and participation in social organisations.
In order to reveal the various types of belonging and to arrive at a relatively homogeneous groups of respondents we first conducted cluster analysis using the K-means method of merging clusters. The analysis was conducted on all variables mentioned above related to the sense of belonging. Cluster membership, distance information and F statistics were saved, to provide information about the contribution of each variable to the separation of the groups (see Table A1 and A2 in the Appendix). This allowed the clustering of a large number of observations into a small number of categories, characterised by the mean points of belonging in one category being relatively close to one other -that is, a grouping based on similarity in scoring. However, the distances between these categories or groups was significantly larger than the distances between the points within one individual defined category.
Considering the differences mentioned previously in the intensity of the sense of belonging scales, standardised values were included in the cluster analysis. Initial cluster centres for the first round were established as a means for cases equal to the number of clusters and then iterated.
For choosing an appropriate number of clusters, two, three and four-cluster models were compared, looking for a solution with the most proportional distribution of respondents between clusters, the biggest distances between clusters and distribution of F values.
In the next step of the analysis we used multinomial regression models that allowed the identification of factors that are important for the formation of a sense of belonging.
Among the independent variables we included: 1. Socio-demographic characteristics: age, gender, ethnicity, level of education, current occupation, financial situation, type of settlement; 2. Experience of migration: year of last emigration (i.e., emigration wave), main reason for emigration, aim of emigration; 3. Social networks: family members or friends that still live in Latvia, having friends who are natives of the country; 4. Subjective evaluations: satisfaction with different areas of life, changes in satisfaction since emigrating from Latvia, trust in the Latvian and host country's government.
The quantitative data of the emigrant survey included very detailed measurements of practically all areas of life. 1 However, on only very few occasions were the original questions kept in the analysis. Several nominal variables as well as several categorical scales were recoded into dichotomous variables; for example, the statement that 'the respondent has close friends that live in Latvia -yes or no?' If the nominal variable had more than two values and one could serve as a reference category for others, we used the system of coding indicators. In other cases, based on the initial measurements obtained in the survey, new variables were calculated.

Identity of Latvian Migrants
In line with national surveys of people living in Latvia, the emigrants' answers showed that people felt a stronger sense of belonging to their closest social groups (family and friends), rather than to social categories (e.g., Latvia, the host country, inhabitants of Europe) (Koroļeva and Rungule 2013). Almost all respondents (93%) felt close (i.e. 'very close' and 'close') to their family, 85% felt close to their friends (Table 4.2, column 1). The next strongest association was with their ethnic group: 73% felt close to people of their own ethnicity. Approximately half the respondents (47-57%) felt close to other emigrants from Latvia living abroad, to inhabitants of Latvia, as well as to Europeans, world citizens, inhabitants of the host country and Source: The author, based on The Emigrant Communities of Latvia survey Differences between clusters significant at p < 0.001 level the neighbourhood they live in. The sense of belonging to a religious group or community was somewhat weaker (Table 4.1). It is clear that national identity still dominates the 'foreign' or 'new' identity among members of the Latvian diaspora abroad, highlighted by the fact that most respondents felt closer to Latvia than to their host country. Overall, 73-74% felt close or very close to the place they lived in Latvia and the place where they spent their childhood. The sense of belonging to Latvia in general indirectly implies not just a territorial belonging but also other dimensions of national identity such as political, cultural or psychological belonging. This might explain a relatively weaker attachment to 'Latvia' than to a particular place in Latvia: only 63% of emigrants felt closely or very closely attached to Latvia (Table 4.2). At the same time, 58% of members of the Latvian diaspora felt attached to the country they currently lived in, and 51% to the place (city, village) they currently lived in.
Of course, without in-depth analysis, these numbers illustrate only the hierarchy of the sense of belonging. Therefore, we then used cluster analysis to distinguish typological groups of belonging. Out of all the versions, a four-cluster solution was selected as the most optimal, based on statistical characteristics and the interpretation of the results. The socio-demographic characteristics of these groups is shown in Table A1 in the Appendix. Based on their characteristics, the typological groups can be labelled as follows: Cluster 1: 'well integrated'; Cluster 2: 'home-rejecting'; Cluster 3: 'host-rejecting'; Cluster 4: 'home-leaning'. The size of clusters is shown in Table 4.3. Clusters 1, 2, and 4 are similarly-sized (around 30%) while 13% of respondents belong to Cluster 3.

Characteristics of Groups
In this section we look at the characteristic traits of the groups distinguished as a result of this cluster analysis, including providing a socio-demographic description of these groups and identifying the differences in their opinions and behaviour. In detecting the relationships between variables, we use correlation tables and calculations of standardised residuals (see Table A3 in Appendix). We also explore the main factors affecting the development of certain combinations of feelings of attachment.

Well Integrated
The first cluster includes respondents that are well integrated into their current country of residence and felt a very strong sense of belonging to it. Almost all respondents in this cluster (95%) felt closely or very closely attached to the host country and the place (city, town, etc.) where they currently lived (91%). Moreover, a large proportion of respondents representing this cluster felt closely or very closely attached to people living in their current country of residence (85% compared to the sample average of 50%) and to people living in their neighbourhood or local area (84% compared to the sample average of 47%) (Tables 4.2 and 4.3).
Based on a very distinct host country identity, we can characterise this group of emigrants as 'well integrated' into the host country. However, a huge proportion of the representatives of this cluster also felt strongly attached to Latvia, to inhabitants of Latvia and to people from Latvia abroad. Compared to other groups in the typology, most representatives of this group (70% compared to a sample average of 56%) felt attached to inhabitants of Latvia, and -at 88% -the identification with one's ethnic group was even stronger. Most (74% compared to a sample average of 53%) felt close to other immigrants from Latvia, as well as to Europeans and inhabitants of the world as a whole. 78% felt close or very close to Latvia, and the attachment was even higher to the place they lived before emigrating (86%) or where they spent their childhood (85%). Analysis of the socio-demographic characteristics of clusters confirms that the timing and amount of time spent abroad has a significant impact on the formation of an individual's sense of belonging. A sizeable percentage of those who were well integrated had spent a considerable amount of time in the host country (Table 4.4). Among the well integrated there were a few who left Latvia because they could not find a job there, but a significant number (22%) who joined their family or started a family abroad. Particularly for those who married 'locals', the family itself can become a very important bridge to integration.
As shown by the socio-demographic analysis of clusters, family is the most important factor in the differentiation of dimensions of belonging. There were a few 'well-adapted' migrants among those respondents whose family members -for example, a spouse or children -still lived in Latvia, but a comparatively high number among them with nobody in Latvia.
Thus, family ties can either facilitate or hold back integration by tying an individual emotionally to a particular place where the family lives. In line with their inclusive identity, the well integrated migrants tend to have friends that include both locals and Latvians in the host country and to be active in both the host country's organisations and the Latvian community.
The results confirm that satisfaction with life is another important factor for integration. The 'well integrated' were more satisfied with life after leaving Latvia than others. Of all groups, they were the most satisfied with their job, family, relationships with people, home, education, standard of living and life as a whole.
Several dimensions of trust were covered in the survey: trust in the government of Latvia, its mass media, police and courts; trust in the host country's government and in the European Parliament. The results revealed that, in general, the well integrated migrants tended to have higher levels of trust in the host country's government (Table 4.5).
Finally, in the country with the largest flow of emigration from Latvia -the United Kingdom -the probability is significantly lower of finding 'well integrated' types of migrant. This could be explained by migrants having less need to socialise Interestingly, the 'well integrated' included more women than men (62% vs. 38%).

Home-Rejecting
The second cluster of the typology -'home-rejecting' -is characterised by a moderately strong sense of belonging to the current country of residence and a strong negative attitude towards their country of origin, i.e., Latvia, and towards all dimensions of national identity. A distinct characteristic of the group was their sense of belonging to the inhabitants of the host country and the inhabitants of the neighbourhood or local area they currently lived in. Overall, 59% of this group felt close or very close to the inhabitants of the host country and 54% felt close or very close to people living in their neighbourhood or local area. At the same time, emigrants in this group demonstrated a comparatively alienated attitude towards everything related to Latvia, as well as displaying weak national identity. This was clearly demonstrated by the distribution of answers on their sense of belonging. Just 17% of respondents felt attached to the inhabitants of Latvia and 27% to other Latvians abroad. Just 19% of this group felt close to Latvia. Socio-demographic analysis of clusters showed that economic conditions forced many in this group to leave Latvia, due to social vulnerability or similar reasons. Thirty-three percent left Latvia during the years of economic crisis (2009)(2010)(2011). For them more than for others the main reasons for moving abroad were a desire to improve their quality of life and live in a country with better social guarantees. Such reasoning increases the probability of negative associations when thinking of Latvia (displaying a weaker emotional belonging, which in our data is identified by belonging to the 'home-rejecting' cluster).
'Home-rejecting' types of attitudes were more common among those who did not have any ongoing connection to Latvia, such as property or friends, yet did have friends among locals. One must note though that in this analysis we were unable to disentangle causal relationships.
Comparatively often the 'home-rejecting' migrant type could be found in the emigrant communities in Ireland and in Southern European countries. They had practically no trust at all in the Latvian government or the Latvian police or courts, yet comparatively high levels of trust in the host country's government (Table 4.5). One can conclude that there is clearly a link between belonging to this group and attitudes towards home and the host country's institutions. The exact mechanisms are not clear, but one possible explanation is that the sense of belonging to a country is closely intertwined with a sense of attachment or respect towards that country's institutions.

Host-Rejecting
The respondents who belonged to the third cluster -that of 'host-rejecting' -were characterised by weak belonging to the host country. That is, they did not feel 'very close' or 'close at all' to the host country and its inhabitants. Direct measurement of the sense of belonging showed that just 3% of this group felt close or very close to the inhabitants of the host country or the residents of their current neighbourhood or locality; 13-14% felt attached to their host country and the place they currently resided in (Tables 4.2

and 4.3).
This group felt weak subjective ties to the inhabitants of Europe or of the world. However, it possessed a significantly stronger national identity: 47% felt close or very close to the inhabitants of Latvia, and 31% to people from Latvia abroad. The ethnic identity was even stronger: 62% felt close or very close to their ethnic group. The sense of belonging to Latvia was also strong, at 62%. Nostalgic feelings towards Latvia seem to have been reflected in respondents feeling very strongly attached to the place where they spent their childhood and to the place they lived in Latvia before emigrating (84-85%).
Host-rejection was more common among those who emigrated very recently, after 2011. Thus, they had little time to develop attachments to their new host country.
The key characteristics for this group -slower adaptation and an incapability or unwillingness to integrate into the country of residence -can depend on several factors, including individual, psychological and personal characteristics. The sociodemographic analysis showed that more often than others, people who belong to this group emigrated because they could not find a job in Latvia or had financial difficulties, such as being unable to pay loans, i.e., they had no other choice.
The majority (60%) emigrated with the aim of finding work. It is possible that many of these emigrants had not had the time but also lacked the willingness to belong to their current host country and integrate into its society.
The 'host-rejecting' migrants often had friends or family members still living in Latvia, such as a spouse or children, parents or other close relatives. This precluded them from attaching themselves to another country. They were less likely to have friends among locals and were not active in any organisations either. Thus, their social ties with their new host country were weak too.
In this group there were comparatively large numbers of men, people with lower levels of education and ethnicities other than Latvian, as well as those struggling financially. Another characteristic of this group was their dissatisfaction with various areas of life, including their job, education, home and so on. These levels of dissatisfaction had increased after emigration. Accordingly, increased discontent with certain spheres of life can strengthen a migrant's sense of belonging to their country of origin.
The 'host-rejecting' migrants had very low levels of trust in any institutions (Table 4.5). It is likely that their reserved attitudes towards the host country are linked with feelings of insecurity, vulnerability and discomfort in an environment they did not know well and could not trust. The highest numbers of host-rejecting Latvian migrants were found in the UK and Ireland, with comparatively lower levels in other Western countries.

Home-Leaning
The dimensions of belonging within the fourth and final cluster -that of 'homeleaning' -are similar to the 'well integrated'. However, the sense of belonging both to the host country and to the country of origin is less intense. Representatives of this group, compared to others, more often identified with Europeans (65%) and with inhabitants of the world (66%). They also felt a sense of belonging, albeit a weaker one, to the inhabitants of the host country. However, their strongest feelings were towards Latvia and anything related to Latvia; 83% felt close or very close to the inhabitants of Latvia and 65% to other people from Latvia abroad (Tables 4.2 and 4.3). Their strong emotional ties to Latvia were confirmed by the fact that a large majority of respondents felt very close to the place in Latvia where they spent their childhood (90%) or the place they lived before emigrating (94%).
More often than others, the 'home-leaning' type of migrants had friends in Latvia and met or communicated with them regularly. Like the host-rejecting migrants, they also often had property in Latvia, but did not have friends among the local population. This demonstrates that having close friends back home increased feelings of nostalgia toward the home country.
Those who had become less satisfied with life after leaving Latvia were more likely to belong to this group, and in general evaluated their satisfaction with various areas of life at lower levels than other groups, except for the host-rejecting cluster.
As with the host-rejecting migrants, many had left Latvia recently. Comparatively often the main reason given for leaving was development, career opportunities (including for children) or family-related reasons. In terms of the occupation breakdown of this group, there were more students, unemployed and people taking care of family than in the other groups.
However, the 'home-leaning' migrants had more trust in Latvian institutions than other groups (see Table 4.5) -a conclusion that attests to their overall positive attitude towards anything Latvian.

Factors That Affect the Sense of Belonging
While the separation of emigrant 'types' into clusters points to different characteristics and various ways of belonging, it is important to understand which factors affect the development of certain kinds of belonging. Therefore, we performed multinomial regression analysis with 'well integrated' as the reference category (Table  A4 in the Appendix). The model fit was very good. The addition of the predictors to a model that contained only the intercept significantly improved the fit between the model and the data, χ 2 = 2717 (df = 180), Nagelkerke R 2 = .43, p < .001.
First, the analysis showed that having close friends in Latvia or friends from Latvia who lived in the host country, or having friends among the 'locals' was a significant pre-condition to a more successful adaptation in a foreign country and to maintaining both a sense of belonging to the host country and a bond with Latvia as well. Having close friends in Latvia reduced the probability of ending up in the group negatively disposed towards Latvia, and increased the probability of ending up in the 'host-rejecting' group -as opposed to being well-integrated. Not having Latvian friends abroad mattered too for those with 'host-rejecting' attitudes. Moreover, not having made friends with the 'locals' increased the risk of falling into the 'host-rejecting' or 'home-leaning' group. This result clearly demonstrates that social inclusion is an avenue to integration. The home-leaning respondents were more likely than the well-integrated to have extended family in Latvia, probably contributing to their more pronounced feelings of nostalgia. Besides friends and family, economic ties such as having a property in Latvia also strengthened attachment to the homeland. Participation in social organisations can serve as a stimulus or reason to foster feelings of belonging, and can also be a consequence of a sense of belonging that is forming or already formed. Based on the results of this analysis, one can conclude that the maintenance of a sense of belonging to Latvia was facilitated by involvement in the organisations of the Latvian diaspora or in both Latvian and international organisations in the host country. This also contributed to a more successful adaptation to the host country and integration into it. This increased the probability that the migrant would fall into the 'well integrated' group as opposed to the 'homerejecting' group. Significantly, it attests to the importance of diaspora organisations in preserving national identity.
As expected, one of the factors affecting identity formation was life satisfaction. Well integrated migrants tended to be more satisfied with life as a whole than others. They reported more often that their life satisfaction had increased, as opposed to 'host-rejecting' migrants who were more likely to see a deterioration in their life satisfaction.
One characteristic distinguishing the home-leaning migrants from others was their trust in the government of Latvia, whereas the well integrated had the highest level of trust in the host country's government. This result shows that positive beliefs about the government can facilitate the formation of overall positive feelings towards that country.
Both cluster analysis and statistical indicators of the regression model confirm that the probability of successful adaptation and integration into the host country is greater the longer a migrant is away from the country of origin. This is because closer ties and a sense of belonging develop to this new country of residence, as well as a sense of being part of the local neighbourhood (see Table 4.4). Interestingly, the home-rejecting migrants tended to live in larger cities, smaller cities or towns, while well integrated migrants were found more often in rural areas.
Significantly less often than both home-rejecting and host-rejecting migrants, the well integrated migrants emigrated due to economic reasons or dislike of the political processes in Latvia.
The probability of becoming a 'well integrated' migrant was higher among those for whom the main reason for emigration was marriage (to a foreigner) than it was among the home-rejecting migrants. Emigrants with secondary or lower levels of education tended to be less well adapted or integrated into the host country, as were men.
However, Latvians tended to be over-represented among the well integrated migrants while Russians were more often found in the 'home-rejecting' group of migrants and less often in the 'home-leaning' group.

Conclusions
The empirical analysis demonstrates that the sense of belonging is multi-faceted and can be grouped depending on the intensity of identification with certain dimensions of belonging. The characteristics of each group of clusters in this chapter clearly points to the main elements that affect the maintenance or weakening of a sense of belonging to Latvia, as well as a sense of belonging to the host country. On the other hand, the results of the regression analysis point to the main factors that affect the formation of different types of belonging.
An important factor in the transformation of belonging abroad is the time period or wave in which emigration took place. The wave of emigration is closely linked to a migrant's reasons for leaving. Social ties are also important in forming a sense of belonging, for example, if the migrant lives alone or with their family and friends. National identity is affected by whether part of a respondent's family -or the family or friends of their parents -still live in Latvia, as well as by having a property in Latvia.
Subjective attitudes and evaluations are the second most important group of factors affecting belonging. Satisfaction with all areas of life -for example, good economic conditions and personal life -will strengthen the sense of belonging to the country of origin and the new host country alike. The same can be said about institutional trust. A stronger sense of belonging both to Latvia and the host country is linked to higher levels of trust in the government of the host country and Latvia. Social participation is important too: being involved in diaspora groups helps maintain feelings of national identity.
Belonging to the group of well integrated migrants is linked to a significant extent to marrying a 'foreigner', living with their family and satisfaction with all areas of life in the host country. Conversely, integration is hindered by all the economic reasons for emigration, such as difficulty finding a job, financial problems etc., as well as a dislike of the political processes in Latvia. Satisfaction with life in general and higher levels of trust facilitates the formation of a positive sense of belonging to the host country as well as to the country of origin. Dissatisfaction with work, family and socio-economic conditions significantly facilitates alienation and the lack of a sense of belonging.
The term 'distance nationalism' (Brubaker 1996) is the most overarching for characterising differences in the dimensions of belonging, as they are differentiated mainly by the sense of belonging to Latvia. Distance nationalism includes the result of the interaction between nationalism and migration characterised by the politically and socially specific relations of immigrants with regard to the host country as well as the country of origin. By using the term 'distance nationalism' in a broader sense than the political manifestations of national identity that most researchers link the term with (Anderson 1992) we can use 'national identity' in the widest sense as belonging to a country, its territory and inhabitants. Distance nationalism in a diaspora can facilitate diverse relationships with the country of origin, emphasising both a willingness to co-operate as well as to confront (Kaprāns 2015).
Maintaining a powerful sense of national identity regardless of the level of integration into and attachment to the host country is facilitated by the frequency of social contacts with Latvia through friends or family that make one feel a strong connection. Separation from loved ones, dissatisfaction with work and unhappiness with life in general are factors that hinder the development of a sense of attachment to the new place of residence. It is possible that adaptation into the host country is also hindered by a gap between expectations of emigration -such as improving the quality of life, earning more, finding a better job and so on -and its reality.
The willingness to distance oneself from the country of origin is manifested in a very weak sense of national belonging (the 'home-rejecting' group). Most of these people emigrated during or after the years of economic crisis and possibly left with a sense of resentment. The most important reason for leaving was a desire to improve their quality of life. This group often does not have anyone left in Latvia to keep in touch with, no property there and few friends among Latvian emigrants. They develop friendships with locals. In terms of their political views, they do not trust anyone except the host country's government. However, as our data shows, weaker ties with Latvia do not automatically result in a stronger attachment to the host country. In the globalised world national identity without a strong social, economic and civic dimension cannot compete with other kinds of identities which determine the diversity of belonging and identification (Ķešāne 2011).
The character of emigration has changed. Migration is no longer a one-directional, permanent and irreversible process, involving breaking ties with the country of origin. Modern technology makes it much easier to keep in touch with 'home', which increases the probability of some emigrants returning. It is clear, however, that most will not. Migrants leave their homeland for good reasons, and people search for and find new lives in new countries. As shown by our study, most Latvian emigrants are satisfied with their lives after leaving. They are happy with their work, their working conditions and their lives in general, leading to the conclusion that there is little hope the majority of those who left will return soon. A sense of belonging to Latvia is not enough to bring them back.      The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.