Conducting Empirical Research With Older Migrants: Methodological and Ethical Issues.

This special section brings together a set of four articles containing novel quantitative and qualitative research on older migrants in Europe. Detailed reporting and reflection is presented on fieldwork decisions and how certain challenges were tackled, and their implications. This introductory article aims to lay the groundwork for a better understanding and awareness of methodological and ethical challenges researchers face when designing and conducting empirical studies involving older migrants. Highlighted are the main methodological issues and ethical dilemmas we observe in studying older migrants, which can serve as a wake-up call for researchers to be more critical throughout the process. We end with a plea for more collaboration between researchers in the field of older migrants, by sharing their data despite potential methodological and ethical problems.

Recent years have witnessed an increase of research on older migrants in Europe. This is a growing and diverse population, including first-generation migrants who migrated in early adulthood for work or in the context of family reunification and are aging in place; older refugees who have been in the country of destination for years or have recently arrived; and those who become mobile near or after their retirement in search for a better quality of life, economic opportunities, or affordable care (Ciobanu et al., 2017). Main research areas include transnational lifestyles (Gustafson, 2008; Klok et al., 2017a) and grandparenting (King et al., 2014; Nedelcu & Wyss, 2020), well-being (Bolzman et al., 2004; Fokkema & Naderi, 2013), and issues of identity and belonging (Burholt et al., 2016; Klok et al., 2017b; Näre et al., 2017; Zontini, 2015). Other issues revolving around aging and migration include health and health care use (Berdai Chaouni & De Donder, 2019; Carlsson, 2021; Reus-Pons et al., 2017), and active aging or aging well (Conkova & Lindenberg, 2020; Lulle & King, 2016; Torres, 2001). Edited collections devoted to aging in a foreign land investigate older migrants’ lived experiences of aging and the processes and meanings of old age across national borders (Ciobanu & Hunter, 2017; Ciobanu et al., 2017; Fokkema & Ciobanu, 2021; Horn & Schweppe, 2016; Karl & Torres, 2016; King et al., 2021). As older migrants are an extremely heterogeneous group with different motivations, life stages, and destinations of migration, this great volume of research pinpoints to their unique experiences and lifestyles in a variety of contexts.

And yet, so far none of the published empirical work on older migrants devotes exclusive attention to the methodological and ethical challenges faced by researchers. This article is exactly designed to do so. We observed that, in general, the data and methods sections give only a snapshot of the reality of the research process and the decisions made (see Torres, 2009 for an exception). In particular, methodological and ethical issues deserve detailed and critical reporting and reflection, as they can have huge implications on the coverage and quality of the data and its interpretation. We argue that the studies involving older migrants need to pay even more attention to the methods used and the ethical rules followed because of their specific characteristics and circumstances. Addressing this issue, all the contributors to this special section intensively engage in reporting their fieldwork decisions and how they overcame some of the challenges they faced, and their implications. In this introductory article we provide an overview of the main methodological and ethical challenges when conducting research with older migrants and sharing data between researchers, and in some cases link to the findings of the four articles in this special issue.

Methodological Challenges

It is beyond the scope of this article to acknowledge and review all the methodological challenges researchers face when conducting empirical research with older migrants. There are five main common issues in quantitative and qualitative research concerning this particular group: (1) defining the population of interest, (2) sampling, (3) selection of questions, (4) validity of the answers, and (5) composition of the research team.

As older migrants are such a heterogeneous group, a first key issue is defining the target population: who is considered old and a migrant? A clear definition of the population being studied is not only important for these researchers, it is also relevant for other researchers, practitioners, and policymakers who are the potential users of the findings. The vast majority of studies among older migrants do provide information about the age that is used as a marker of being old and which generation and ethnicity the population of interest belongs to. We observe that an age of 50 or 55 is commonly set as cutoff point, with the rationale that migrants tend to exhibit poorer physical and mental health in their earlier life stages based on their living and working conditions (van der Greft & Droogleever Fortuijn, 2017; Kristiansen et al., 2016) and thus feel older at a younger age than older adults without a migration background.

In terms of migrant generation, European studies are usually limited to older adults migrating themselves (the first generation, sometimes called the zero generation when parents of migrant children follow their adult children in migration). However, the age at which the migration took place is not often clear. To interpret the findings, it is especially important to know whether the research group only includes older migrants who migrated in adulthood or also those who arrived as children (the 1.5 generation) and were therefore partly socialized in the host society.

With regard to ethnicity, European researchers often narrow their population of interest to one or a few of the largest groups of non-Western older migrants (e.g., Moroccans and Turks in the Netherlands, Indians and Pakistanis in the United Kingdom). The reasons for this are substantive (culturally most distant groups may be expected to have a high risk of vulnerability) as well as practical (by far the most publicly accessible data on these groups). Among qualitative researchers with a migration background, however, it is not uncommon to study older migrant groups smaller in size who are ethnically similar to them. In general, no further selection criteria on migration-related characteristics are applied or raised, such as migration motive, timing of migration in the life course, migration context in terms of historical time, or migration status. A good example of an ethnic group for which the (implicit) assumption of homogeneity in these respects does not hold, and therefore alternative explanations for within-group differences in outcomes remain underexposed, are Chinese older migrants in the Netherlands. The first among them was mainly from the former colonies of Indonesia and Suriname, followed successively by people from Hong Kong for economic reasons and people from China for political reasons in response to the Cultural Revolution and because of family reunification (Cheung, 2022).

A second methodological issue is sampling. It is well-known that it is not easy to entice older migrants to participate in research. As a result, survey response rates among this group tend to be rather low and participant recruitment in qualitative research can be very time-consuming and costly. Difficulties researchers may face when recruiting older migrants are manifold (Areán et al., 2003; McHenry et al., 2015; O’Brien et al., 2006). Often-mentioned reasons for refusal are fear of their privacy being breached and mistrust about use of their personal data (especially for older refugees), stigma of health-related issues they might be going through, and skepticism about the research outcomes. Another frequently cited reason is fatigue, not only due to health or other problems but also because of being interviewed for the umpteenth time and yet not seeing this lead to any improvement in their own lives. Refusal may also be due to language or cultural barriers, or instrumental barriers such as transportation or scheduling issues. In this issue, Tomás and Ravazzini (2021) clearly show that the response rate was particularly low among those older migrants for whom the questionnaire was not available in their mother tongue. Response rates among older migrants were also relatively low in this issue’s study of Seibel and Gerner-Haan (2022), yet older cohorts of migrants were more likely to cooperate and participate in the survey study than their younger counterparts. The authors explained this finding by the fact that older migrants are less mobile than younger ones and are accordingly easier to reach.

In an attempt to increase the response rate in a survey among older migrants, the fieldwork is sometimes outsourced to survey agencies that are specially equipped for this group (staff who speak the languages, know the recruitment channels, etc.), with the disadvantage being that the research team has less control over and information about how data are collected. For qualitative studies, literature highlights the importance of working with gatekeepers such as influential community members and religious leaders. For example, Lörinc and colleagues in this issue made extensive use of migrant associations in recruiting their participants for conducting walking interviews, which substantially increased the participation rate (information based on correspondence with the first author). However, gatekeepers gaining access to the target population does not automatically translate into participation because of issues like fear of gossip (Ryan et al., 2011; Sin, 2004). Research involving older migrants living in institutions such as asylum centers and nursing homes requires cooperation with those institutions in the first place.

Also crucial is to gain and give insight into the representativeness of the sample, in order to assess the generalizability (in quantitative research) and scope (in qualitative research) of the findings. Information on the representativeness of the studied group of older migrants is the exception rather than the rule. An important reason is the lack of available or reliable data on sociodemographic characteristics of the target group from official sources (population registers, census data). One example is the invisibility of a nonnegligible share of international retirement migrants in the official figures, namely those who have not registered as residents in the country of destination (Böcker & Balkir, 2012; Bolzman et al., 2021). Qualitative researchers should be especially aware that involving gatekeepers may lead to selection bias. The participants will very likely consist of those who take an active part in community life and who benefit the most from the activities gatekeepers offer.

Third, the selection of questions is crucial—after all, one only gets an insight into matters on which questions have been asked. Theoretical assumptions together with the findings of previous research need to guide the questions asked in a study. However, there might be researcher biases and agendas in research and data collection, which sometimes stem from the funding agency’s guidelines. Although there has been some shift in focus recently, research questions targeting older migrants tend to revolve around issues of vulnerabilities and less on their resilience or their caregiving roles (but see Ciobanu & Fokkema, 2021; Klokgieters et al., 2020). A notable exception is international retirement migrants. For this group of older migrants, it is often assumed that enjoying the sunny side of life is the incentive to move and that their migration leads to a better quality of life, with the risk of ignoring other migration motives and forms of social, economic, and health vulnerabilities (but see Bolzman et al., 2022; Botterill, 2017). Selection of questions based on these assumptions can lead to distorted presentations of the older migrant groups in question, thus reinforcing stereotypes and stigma by depicting unidimensional images (Ciobanu et al., 2017).

A fourth methodological issue is the validity of the answers given by older migrants. The research group might only present themselves in a particular way and in an interview situation that takes place only once, so researchers might not see the complete situation migrants live in or experience. Migrants may also withhold information. Recent older refugees, for example, might opt out by not revealing all their experiences because of fear for their own and their family’s safety—information that might be useful or necessary to answer the initial research questions. Plus, the presence of other people such as spouses or children during an interview cannot always be avoided. This might not only affect interviewees’ answers and thus the findings, it also raises ethical challenges which we discuss in the next section. Last but not least, researchers should be mindful of the effect of using interviewers of the same ethnic origin and involving gatekeepers on interviewees’ responses. Besides the various advantages, being interviewed by people from the same ethnic group might increase respondents’ tendency to give socially desirable responses and, in qualitative research, reduce interviewers’ inclination to ask follow-up questions because they take certain answers for granted. Kappelhof (2015), for example, found that respondents who were interviewed by people of the same ethnic origin were more likely to report more traditional attitudes and to belong to their own migrant group. Gatekeepers might only invite participants for an interview who portray the image of the group gatekeepers want to show, such as those who can best articulate the needs of older migrants and who speak positively of the organization and are very satisfied with its activities. As a result, the same stories get retold while other stories remain untold.

Last, the composition of the research team is critical as it has an impact on all stages of the research process. Generally speaking, having a team that is culturally and linguistically diverse is advisable for conducting research with older migrants. It is equally important to also have research staff with a nonmigration background. In this way, research teams will not take certain beliefs, assumptions, and previous findings for granted and will ask for further elaboration on culturally relevant aspects when interpreting the results. Hence research team members with and without a migration background can complement each other, keep each other on track, and avoid getting stuck in stereotypes. Another common advantage of a culturally and linguistically diverse research team in qualitative studies is that researchers or research assistants with mobility experiences can conduct the empirical fieldwork themselves. Usually first-generation older migrants prefer to be studied in their native language and place more trust in interviewers of the same ethnic origin; regardless, commonalities between the researchers and the researched are always negotiated at different phases of the research (Leung, 2015; Nowicka & Ryan, 2015).

It is not always feasible to have such a diverse research team. In such cases, translators or interpreters can play an important role, although caution should be exercised mainly for two reasons. First, languages evolve and there is not one single way of translating from one language to another. While direct word-by-word translation might seem technically correct, some meanings and thus information might get lost (Temple, 2002). For quantitative research, using previous translations of questions and scales is recommended, especially if these have been validated for the target group in question. Second, translators or interpreters might have their own ideas or perspectives about the research, as they are also considered active producers of knowledge, just like researchers; this might not always be desirable, especially when interpretation of certain questions is done in suggestive ways. While translators or interpreters can also act as conduits of culture, as they can help the researchers understand culturally specific expressions and meanings, their contribution might not necessarily reflect the way the interview questions are meant to be asked (Berman & Tyyskä, 2011; Temple, 2002).

Ethical Challenges

Research ethics has three core principles: respect of persons, beneficence, and justice as defined in the Belmont Report (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978). These principles lead to important ethical guidelines researchers follow: anonymity, confidentiality, informed consent, self-determination, and minimization of harm (Hennink et al., 2020). They are generally translated to migration research as informed consent that makes clear to the research group: the research objectives; the right not to answer certain questions; the right to withdraw from the study at any time without any negative consequences; the voluntary nature of the answers given; and what happens to personal data such as anonymity, confidentiality, research team’s access, and secure storage.

Older migrants are a research group that should be treated with extreme caution. Researchers might not be able to assess the cognitive state of participants beforehand; therefore, the issue of informed consent can be jeopardized. As demonstrated in the contribution of Tezcan-Güntekin and colleagues (2021) in this issue, sometimes other people (family members, caregivers) are present during the interview, which may complicate the informed consent procedure to participate or call into question the voluntary nature of the participation. This latter can also occur when researchers need to rely on gatekeepers. Migrant organizations and other relevant bodies might have their own agendas and perspectives, which might go against the principle of voluntary participation (Ryan et al., 2011). In the case of health or mobility problems, researchers need to reevaluate their decisions to conduct an interview, or at least try to minimize physical and mental harm to participants. For example, in the study by Lőrinc and colleagues (2021) in this issue, the participants were accompanied in the walking interviews by two researchers in order to minimize potential physical risks of walking and talking.

Other issues that are specifically relevant for older migrants relate to how to deal with sensitive topics. Some conflicts within the family or in the community as well as very stressful or emotional events might surface during interviews. On the information letters, participants are informed that they have the right to not respond to questions they do not want to answer or to withdraw their answers at any time, even after the interview. Nonetheless, sometimes during the interviews questions vital to the research might trigger emotions, such as those related to the migration history (especially if it involved a trauma caused by fleeing from the country of origin or forced marriage) or social life (especially when loved ones are missing or have passed away). In order not to harm the participants, the interviewing team needs to be very well trained to handle such situations.

One final issue we would like to raise is how the coronavirus pandemic poses ethical challenges to conducting research with older migrants. Fieldwork studies had to stop for a long while because older age was recognized as a high risk factor for being hospitalized or dying from coronavirus disease 2019 (COVID-19; United Nations, 2020), so researchers had to avoid exposing older adults to the virus. After large-scale vaccination, older adults might still fear the virus and therefore not want to participate in research. This may be even truer for older migrants of non-Western background, as they have lower vaccination rates (Robertson et al., 2021) and are more likely to suffer adverse effects after infection due to underlying health problems like heart disease, diabetes, and obesity (El Fakiri & Bouwman-Notenboom, 2015). If they do give their consent to be involved in the study, both parties must be extremely cautious to not put participants’ health at risk. This situation of voluntariness in participation has never been so extremely coupled with potential physical health risks for the participants, and yet this is not addressed at all in ethical challenges of contemporary studies.

Conclusion and Future Recommendations

We highlighted in this introductory article methodological and ethical challenges researchers face when designing and conducting empirical studies involving older migrants. We addressed five main methodological challenges that researchers face in both qualitative and quantitative studies, namely, (1) defining the population of interest, (2) sampling, (3) selection of questions, (4) validity of the answers, and (5) composition of the research team. Next, we addressed the main ethical challenges in the areas of informed consent, discussion of sensitive topics, and the impact of COVID-19. In so doing, we wanted not only to increase researchers’ awareness of the challenges they might face in their studies, but also to call for a greater debate on how to design and conduct empirical studies with older migrants, as well as more detailed reporting and critical reflection on the methodological and ethnical decisions they make. After all, each step in the design of a study has tremendous implications for the study participants and the findings.

The growing interest in older migrants among European researchers has resulted in a considerable wealth of data. The individual data sets, however, have limitations. They are usually limited to older migrants of a single or a few ethnicities, or the size of each ethnic group is so small that the groups have to be aggregated. Moreover, they usually pertain to a specific type of older migrants (e.g., former guest workers, postcolonial migrants, refugees, flying grandmothers, lifestyle migrants), one country of destination, and one moment in time. It is regrettable that, so far, hardly any attempts have been made to merge different data sets, leaving pressing and interesting research questions unanswered (but see Bolzman et al., 2021; King et al., 2014). To illustrate the potential benefits of merging data sets on older migrants: when done for same ethnicity or different ethnicities it increases the potential for presenting and explaining within- and between-group differences in the outcomes of interest; when done for different destination countries it facilitates examining the role of macro-factors (e.g., type of welfare state, attitudes of the population toward migrants); when done for different time points it allows for the study of trends in the outcomes of interest.

Thus, in addition to collecting new data on older migrants, we strongly advocate more collaboration between researchers sharing their data sets. Of course, we are also aware of other methodological and ethical challenges our argument for merging existing data sets might create. Methodological challenges may arise if questions are not asked in the same way or in different languages. Different interview modes or settings and differences in interviewer composition and training can create methodological biases. Additional methodological challenges include how to analyze the merged data sets, especially when there are culturally specific answers that were given as a response to particular interviewers. Data sharing also involves some ethical issues due to confidentiality reasons, legal issues associated with data protection, and issues of data ownership. Despite such challenges, if the aim is to understand the experiences of older migrants better it is worth trying to share existing data sets and analyze them anew.