It is widely recognized that migrants maintain connections to their home countries, including social, economic, cultural, political, and familial ties.1 It is also known that these transnational connections have impacts, both positive and negative, for the migrants and the families and societies in the sending countries.1 One area of research that has received significant attention in this field is transnational family support, defined as the exchange of goods, money and information, and social and emotional interactions, between families across borders.2–4 This research, however, has primarily focused on support giving by migrants towards their home country, including distance-mothering, caring for elderly across borders and sending remittances.4–6 There has been less emphasis in the literature on transnational family support for migrants, especially from the perspective of the family members in the home country who are providing the transnational support.
The limited research that exists on transnational family support for migrants suggests that support received from family back home tends to primarily be in the form of psychological and social support, which may contribute to a sense of belonging, connection and well-being for migrants.3,7–10 These studies have mainly focused on the perspectives of the migrants and those that include family back home were living in European or more developed countries. Perspectives have also mostly been limited to those of grandparents. Other research in the field, has focused on grandparents and other family kin, mostly women, who provide support by caring for children who remain behind and/or by making extended visits to the migrant families and caring for children in the host country.3,11 Findings from this body of research, show that for family members providing the support, they may feel honored and satisfaction in helping their migrant family members, however they may also experience stress, and/or feel support is not adequately reciprocated.9 For migrant parents, they may experience difficulty negotiating their parenting role in a context where children are mainly being raised by family caregivers in the home country.9 The detailed experiences of what it is like for family members in the home country to provide other types of support from a distance, and their perspectives on how they feel they are contributing to their migrant family members’ life in the new country, the emotions that they experience, and the impact on their lives and relationships with family (both locally and abroad), however, remains relatively unexplored. To better understand transnational family dynamics and the support networks maintained across borders, more in depth documentation is needed on the support giving provided to migrants by their family members in the home country, especially in different migration contexts (eg, forced versus non-forced migration; migration from low- and middle-income countries as well as high-income countries). Moreover, to fully capture the transnational dynamic and to comprehend the impact on the lives of migrants as well as those back home, studies on transnational family support should include the perspectives of different family members, both in the home and migration countries.1,12,13
To begin to address the gap on transnational family support for migrants, a study is being planned. The objective of this study will be to gather data from a diverse group of migrant families living in Montreal, Canada and also from their family members back home, who are living in various countries around the world. Cross-country research traditionally consists of travelling across sites and gathering data in person.14–17 However, given the planned study intends to include migrants from a multitude of source countries, interviews with family members back home will only be feasible if conducted through phone or internet (video) calls. Given limited research in this area and the proposed distance interviewing for data collection with those living abroad, the current study therefore aimed to determine the feasibility of conducting transnational research, particularly in the context of a low-income country, via the use of communication technology. Specifically, the current pilot study aimed to address the following questions: 1) Are family members in the home country interested and willing to engage in this type of research? 2) Is it feasible to gather data using communication technology, including observational data? and 3) Who (which family members) should be included and what questions should be explored? We also used this opportunity to gather some preliminary data on the experience of providing transnational family support from the perspective of family members back home.
This small, qualitative descriptive, pilot study was conducted in Kisumu, Kenya.18 The location was selected because the Canadian researchers had existing collaborations with researchers in Kenya, and because Kenya is a low-income country where migration is common. The setting was therefore deemed ideal for determining the feasibility of conducting transnational research on the phenomenon of transnational family support and for gathering pilot data.
DM and DO, locally based in Kenya, recruited (via posters and word of mouth) a convenience, purposive sample of nine adults (6 men and 3 women), who self-identified as providing support to family members who migrated to another country. ‘Support’ was left open to the interpretation of the participants, but was broadly defined as emotional, instrumental, financial and/or material help, as well as providing advice and/or information. The study was explained at the time of recruitment and consent forms were signed in Kisumu. The study was briefly re-explained and consent reconfirmed verbally by phone at the time of the interview. We gathered data through semi-structured interviews using Skype to call participants (on their cell phones) in Kenya; LM conducted the interviews from Canada between May and July 2018. This approach was selected as it did not require the participants to have access to internet, it allowed for flexibility in scheduling the interviews, since all participants had cell phones, and because it ensured no costs were incurred by participants. Using Skype also facilitated the recording (Skype has an application for this), and was relatively inexpensive.
Interviews were in English, lasted between 20 and 45 minutes, and were audio-recorded. We asked participants to first describe their context (who migrated and when, and the circumstances of migration). We then asked them to discuss their relationship to their family member(s) who migrated; the nature and kind of support they gave and how they maintained contact with their family; and their thoughts and feelings, both positive and negative, about their experience of providing support. Following these questions, we asked participants to share their opinions regarding the relevance, feasibility and how best to conduct this type of research (eg, who we should speak with, what questions to ask, feasibility aspects and potential barriers to participation). We collected some basic socio-demographic information from each participant and kept field-notes, including general details and impressions about the interview process (i.e., what worked and yielded more data and what was more challenging).
Interviews were transcribed verbatim. Transcripts and field-notes were compiled and analysed over the data collection process. To address the feasibility questions, LM reviewed and extracted relevant information from the field-notes and the direct feedback received from participants during the interviews regarding conducting research on transnational family support from family members back home. This information was then used to generate a summary of the key points to consider when conducting this type of research.
To begin to identify themes related to the experiences of providing transnational support to migrant families living in another country we also performed a content analysis of the interview data.19 This process was led by the first author (LM), and supported by NE. LM read and reviewed all transcripts multiple times to identify patterns in the data; NE read a selection of three interviews. We then abstracted sections of the transcripts and grouped them into categories based on the patterns that we had identified. Both LM and NE reviewed the categories and regrouped them into larger themes; this was an iterative process and was done until agreement was reached. We then refined the labels and selected quotes that best supported and illustrated the categories and themes. DO and DM were not involved in the analysis of the raw data given they were acquainted with some of the participants. They did however, review and confirm their agreement with the final themes. Participants’ socio-demographic and contextual information were summarized.
We received ethical approval from the University of Ottawa and from the Great Lakes University of Kisumu. Approval was also obtained from the National Commission for Science, Technology and Innovation (NACOSTI) and from the advisory institutions of the Government of Kenya on matters of research.
The participants’ characteristics are reported in Table 1; they described a range of migration situations and relationships to the family members who were living abroad. Seven participants spoke about only one family member who had migrated, while three participants spoke about more than one family member (i.e., three brothers, three cousins, and their wife and children respectively). Most of the migrant family members were living in the US; for some the move was recent, while for others it was many years ago; one family member was born in the US (her parents had migrated). The reason for migration was mainly to study; almost all of the participants’ family members planned to stay abroad permanently. Two of the participants had been migrants themselves (to the US) but had since returned to Kenya.
Feasibility of conducting transnational research
Participants were generally positive about the study and thought that the research was relevant. One participant stated: “Whatever comes out of this study should be utilized, not kept in a shelf somewhere… to assist families.” While another requested to have the results: “I just want to know how we get the research findings of the study you are conducting. I may be interested to get the findings.” Overall participants were open and willing to share their experiences and most felt that others like them would also be interested and receptive to participating in such research. Table 2 presents the factors that participants believed might influence participation and data collection. They highlighted the importance to speak with different family members and to ask a range of questions during the interviews in order to obtain diverse perspectives and more detailed accounts of experiences especially on more negative aspects. They warned that family dynamics will also affect what is shared by each family member. They also expressed that privacy concerns and emotions could affect one’s willingness to participate and/or what information a participant would feel comfortable disclosing.
Interviewing participants by phone from Canada proved feasible, however there were some challenges. These included communication issues due to language barriers (English was their second language and it was difficult to understand their accents) and background noise; coordinating a time to speak, especially given geographic time differences; and technical difficulties (cell phone battery dying, phone not being turned on when the interviewer called). Not speaking face to face (and without the video function of Skype) was less personal and also meant that non-verbal expressions could not be captured. For those participants who had access to internet, they said that they would have been open to a WhatsApp video call (participants were less familiar with Skype), although they admitted that the quality of the internet connection may have prohibited this approach. When asked about the possibility of conducting group interviews with their migrant family members, all participants said that they would be comfortable with this as long as their family members were in agreement.
Pilot results: Experiences of providing transnational family support
We identified four main themes related to the experience of providing transnational family support: 1) Nature, means and type of support; 2) Shifts in support over time and influences on nature and type of support; 3) Expectations of those back home and bi-directional support; and 4) Strain, stresses and tensions (Table 3). Results showed that multiple family members tend to be involved in giving support to family who migrate. The support may take different forms, may be local, and often shifts over time, with assistance being more intense during the early resettlement period and in times of financial difficulty. In some instances, support (e.g., childrearing advice) may be less welcomed or challenging to give. It also appears that the context and age at migration affect the amount of support needed while the relationship and degree of closeness, gender, age and life circumstances shape the nature and intensity of support given by individual family members. The amount of contact and how communication is maintained (e.g., Facebook, WhatsApp, phone) and with whom, also varies. Important factors affecting communication include access to technology, logistics, closeness to the migrant family member(s) and financial means.
Overall the families were generally positive about their family members’ migration, however, the transnational family context did present some challenges. Family members missed and worried about each other, many back home had expectations of return support, which was not always fulfilled, and there were situations that caused family strife. For some, the support-giving to migrant family members also caused financial strain, family conflict and tension among those back home.
The key findings from this study are that families back home are interested and willing to participate in a study on transnational family support and that it is feasible to conduct interviews and to collect data from family members back home using communication technology. The pilot data gathered also shows that obtaining the perspectives from family back home who are providing the support to their migrant family members yields informative data on the phenomenon of transnational family support.
Implications for conducting transnational research with family back home
To gather a diversity of perspectives and experiences, a range of family members within and across families (various relationships and closeness, gender, age, circumstances) in different migration contexts, including both recent and non-recent migration, should be recruited. To elicit data on both positive and negative experiences, a variety of questions that explicitly ask about different kinds of support provided, challenges, less welcomed support, negative impacts and tensions, as well as expectations, benefits, and emotions, must be asked. There appears to be an openness to conducting family group interviews, although to collect information on more sensitive topics family members should be interviewed individually since they may be reluctant to share personal feelings with other family members present. Family dynamics should also be considered during the interviews and when interpreting the findings.
Flexibility is essential when coordinating interviews, including accommodating different schedules and time zones and adapting to unforeseen situations (e.g., background noise, phone dying) that may arise during interviews. While this is relevant to any cross-country study, the constraints are likely greater for those in a low-income country. For example, access to electricity to charge a phone or finding a quiet, private location to hold a discussion may be more of an issue. Another consideration with regards to the timing the interviews, is also access to technology. We decided to conduct the interviews by phone calls since we knew that participants all had cell phones and this would broaden the scheduling options. Had we decided to use the internet, for those with no home access, we would have had to coordinate interviews based on when they could go to the University (organized by our colleagues locally). This would have incurred additional costs, been more burdensome for the participants, and likely would have reduced participation. In our planned Canadian study where participants back home will be in multiple countries, coordinating a location where participants can go to access internet will not really be a reasonable option, as it will be too unwieldly to organize across so many sites.
Data collection approaches should also be flexible. We used Skype to call participants and this worked well, although we could not gather observational data. The range of ways that families maintain contact (Facebook, WhatsApp, phone, Skype), however, could offer various options (phone, video, texts, emails) for communicating with families and also different types of data (interviews, pictures, written exchanges, observations).20–22 Given that access and knowledge on how to use technology/internet varies, as well as capacity and comfort levels in disclosing personal information, using a mix of sources and approaches for gathering data may be the most beneficial. As with any cross-linguistic study, having interviewers who speak the maternal language of those back home would also improve communication by addressing language barriers, especially when there is no face-to-face (video) interaction and there is background noise, making it more difficult to understand what participants are saying.23,24
Given some of the issues (mistrust, communication barriers, privacy concerns, negative emotions) that may affect what participants share, it may be best to have more than one contact with participants. Multiple interviews may allow families to open up over time and feel more comfortable disclosing as they get to know the interviewer.23,24 Follow-up interviews would also provide an opportunity to ask for clarifications and to validate interpretations. Additional contacts may also allow the researchers to tap in to other data sources (e.g., pictures, text exchanges between family members) that participants discuss in their first interview.
The results also reveal certain ethical issues to consider when conducting transnational distance interviews. Firstly, the importance of ensuring that consent is informed and voluntary and that family members are clear on what participation involves, especially if consent is only obtained over the phone/internet and there are language barriers. In the current study participants were recruited by researchers locally, however in our future work we intend to recruit participants back home via the migrant families in Canada. In such research, it will therefore be imperative to have additional time to explain the study and to respond to questions, and to implement strategies in order to confirm that family members understand what is being asked of them, as well as the potential benefits and harms, if they choose to participate. This would include using known strategies such as, repetition, the use of simple language, allowing them to read the consent form in advance, asking them to explain back what they understand, and ensuring information is presented in their language.23–27 Given the vulnerability of certain families, i.e., some family members back home hoping to migrate, and some migrant family members still being in the process of filing their immigration application, or being in a precarious immigration situation, it must also be made explicit to participants that their participation or non-participation will have no impact on them or their migrant family members, and that the researchers have no role in immigration processes.23–27 The economic situation of families should also be considered; to avoid any potential conflicts or discomfort there should be complete transparency regarding any monetary compensation that may be given to either the migrants or the participants in the home country.
The second ethical issue is regarding how researchers may adequately respond from a distance, to families who may become emotional or distressed during interviews. The participants in this study expressed some sadness and also discussed tensions and conflicts related to their family members’ migration, and also due to expectations of return assistance not being met. Interviewers therefore need to be sensitive to these potential issues and be cautious to not cause or aggravate family stress. It is therefore critical to emphasize to participants that they have the right to not respond to questions that they prefer not to answer, and that they may withdraw their participation and data at any time; this is especially so if participants are unfamiliar with research, and their rights as participants. Additional time (or contacts) could also be allocated to ensure interviewers or a specialist on the team can provide psychological support (by phone or internet) as needed.23,24 If feasible, local supports should be identified for referrals as well. Research is also warranted to further develop guidelines for ethical issues that may arise when conducting research using communication technology in a transnational context, especially when multiple countries are involved and/or families back home may be living in unstable or war torn areas.28,29
Transnational family support and implications for population health
The pilot results, although preliminary, add to the body of literature on transnational family support for migrants. They showed that different kinds of support are provided, including financial, emotional, practical and spiritual, as well as various types of support back home, and that a range of factors influence its nature and intensity. They also showed that some support may be less welcomed by migrants, or families may find it challenging to give, or feel less inclined to provide it. Consistent with other research, we found that families are mostly positive about their family member’s migration but that there are also some negative effects, including strain, tensions, and family conflict that are directly related to the support-giving.3,9,10 Our results also showed that support is often bi-directional, although expectations of those back home can vary. Overall the findings highlighted that experiences are complex and diverse.
There is an emerging body of literature regarding transnational ties (social, economic, cultural) to the home country and the health and well-being of migrants, which indicates both beneficial and adverse effects on life-style behaviours, psycho-social health and disease management.10,30–34 The preliminary findings here suggest that financial and practical support from family back home help migrants with settling into the new country, while emotional and spiritual support may provide a sense of connectedness and motivation in the face of loneliness and discouragement due to resettlement challenges. Family assistance provided back home also helps with maintaining family and other commitments and may foster a close connection with home. It also helps migrants continue to sustain their lives in the home country (e.g., homes) for when they plan to return. Transnational family support therefore seems to be an important resource, in providing not only a psycho-social function, but also in providing tangible practical and economic assistance to migrants, that extends across borders.
Future work is needed to explore transnational family support (towards migrants) more in depth and to better understand the impacts, both positive and negative on health, for migrants as well as on their family members back home. For migrants, despite the distance, family back home are evidently having a role in shaping both the social and physical environments of migrants in the new country and they are likely also shaping day-to-day decisions and behaviours. Similarly, providing transnational support is having an impact on the living conditions (e.g., by reducing financial resources) and social environments (eg, affecting relationships between family members) of family in the home country. Receiving and providing transnational family support therefore has implications for both health risk and protection for migrants and their families in the sending countries.
Transnational family support also has implications for health and social care, programming and policies for migrants in receiving-countries. For example, policies that affect transnational transportation, communication, importation of goods, and banking and migration (e.g., travel visas) can facilitate or create barriers to the provision of transnational family support.35 Migrant employment policies and integration programs, can also reduce or increase the need for financial support from family back home. Support programs and care-providers working directly with migrants could also play a role in supporting and assisting migrants to mobilize transnational family support as a resource and/or in mitigating the negative effects for both migrants and for the family who are providing support from a distance. For the latter, studies will be needed to determine how this may be best achieved.
This was a pilot study meant to inform the methods for a larger study. The sample was small, highly educated, English speaking and represented the views of a select sample in Kenya whose family members had all moved for the purpose of education or employment. In other low-or middle-income countries, or migration contexts, there may be different feasibility issues that were not captured here. Different populations may also have different views and interest in participating in this type of research. The results presented here regarding the experience of providing transnational family support are also very preliminary since collecting pilot data was a secondary goal.
The planned Canadian study will build from this pilot work. We plan to focus on migrant families with children and will aim to better understand transnational family support specifically in relation to parenthood in a new country and the related health outcomes. We will recruit a diverse migrant population (various countries of origin and migration trajectories) as well as their family members back home, including different relationships. Based on the pilot experience, data collection will consist of a mix of group (migrants and their family back home together) and individual (migrants and family members back home separately) interviews, with more than one contact, and we will attempt to tap into other data sources as well (pictures, text exchanges). Questions will explore both positive and negative experiences from the perspectives of the migrant families as well as those providing support from a distance, including how they believe this may impact their health. To optimize participation we will use a mix of phone and internet, and will offer different modes for communication (WhatsApp, Skype). Interview times will be flexible, and we will be sure to allocate sufficient time for explaining the study and obtaining consent. The research team will be available to respond to any participants, local or abroad, in the case where they experience distress.
This pilot study shows that family members back home are willing to engage in research on transnational family support. It also shows that using communication technology is efficient and logistically feasible for interviewing and collecting data from migrants’ family members who remain in the home country. It also demonstrates that informative data on transnational family support can be generated by including family members’ back home in research.
The authors acknowledge the Great Lakes University of Kisumu for their logistical support and the School of Public Health at Mount Kenya University for facilitating the dissemination of the findings at their first Public Health Conference held on 23rd – 25th October, 2018 in Thika, Kenya.
LM was supported by a Post-doctoral award for healthcare professionals, by the Fonds de Recherche Québec- Santé (FRQ-S). The research activities were supported by NE’s general research funds at the University of Ottawa and LM’s start-up research funds at the University of Montreal.
The authors completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available upon request from the corresponding author), and declare no conflicts of interest.
Lisa Merry, RN, PhD
Faculty of Nursing, University of Montreal
Pavillon Marguerite-d’Youville, Room 6061
2375 Côte-Ste-Catherine road
Montreal, QC H3T 1A8