Abstract
The increasing development of pan-national and supra-national institutions and ‘states’ such as the European Union implies that comparative cross-national research will become both a more frequent and fruitful research exercise. After all, there is the assumption that data will be consistent and easily available, thought policy contexts are increasingly shared and co-terminous. However, this may not always be the case. Reflecting on a European Union and Nuffield Foundation project, which considered the experience of migrant children and, conducted in Greece, Portugal, Sweden and the UK, the author highlights that, as with any study concerning childhood and youth, qualitative and quantitative methodologies remain ‘culture-bound’. Tracing examples from the developing sociology of childhood, this article suggests that in as much as we recognise the cultural specificity of childhood, so too must we acknowledge that research methodologies are a product of, and embedded within, particular national/cultural contexts. It concludes that, even at the fundamental level of analysing data, culture, ethics and research methodology are closely interconnected and cannot easily be separated into discrete universally understood categories.
Introduction
In recent years, cross-national research has become a major feature of academic activity. In particular, it has fed and been informed by the growth in comparative studies across most academic disciplines, which recognize that as a consequence of globalization, policies that affect individual citizens are no longer set exclusively by national governments. Within Europe, for example, the increase in cross-national research activity has been encouraged by the growth in potential funding sources, among which the European Union is the most significant. Indeed, a major feature of the EU's Sixth Framework is the desire to establish a range of research networks across a greater number of Member States. Yet, whilst cross-national research has developed, methodology and definitions have not remained static. Instead, they too have matured beyond the simplistic explanatory statements offered, for example, by Kohn (1989), who asserts that cross-national research is “any research which transcends national boundaries” (p. 20). Instead, researchers have increasingly been guided by the work of Hantrais and Mangen (1998), who argue that cross-national research involves the study of “particular issues of phenomena in two or more countries with the express intention of comparing their manifestations in different socio-cultural settings” (p. 1).
For Hantrais and Mangen (1998), a cross-national study also implies that “one or more units in two or more societies, cultures or countries are compared in respect of the same concepts and concerning the systematic analysis of phenomena, usually with the intention of explaining them and generalising from them” (p. 2).
While Hantrais and Mangen (1998) offer a clear sense of direction, it is, for those researching children in cross-national contexts, a route that leads into a number of methodological and theoretical minefields. In particular, researchers have become aware that the theories, definitions and concepts with which they are familiar are not universal. Instead, they are culturally specific and determined by particular national/cultural factors. Indeed, an initial challenge presented to those who attempt to conduct cross-national studies concerning childhood and adolescence is the historical, cultural and geographical specificity of these concepts. While children and young people are world wide, in that they exist in all forms of human society, it does not follow that all children and young people experience childhood and adolescence in the same way. Franklin (1995) identifies the problem noting
Definitions of children, as well as the varied childhoods which children experience are social constructs formed by a range of social, historical and cultural factors. Being a child is not a universal experience of any fixed duration, but is differently constructed expressing divergent gender, class, ethnic or historical locations of particular individuals (p. 7).
Thus, rather than being an ahistorical and universally experienced status implicit in Hazard's (1947) notion of ‘a world republic of childhood’, Franklin suggests that childhood is fashioned by the socio-political and historical contexts of particular societies. Thus, childhood is a stage in life, experienced differently by children, depending on when and where in the world they are born and raised.
Franklin's (1995) argument builds on the work of Archard (1993) who, in turn, draws upon Aries (1973), when explaining how childhood is an historical construct that is simultaneously influenced by the socio-political context it which it occurs. In particular, Archard (1993) suggests that childhood has emerged over the past three hundred years, structured under the burgeoning influence of education, health and social care professions. As a status, childhood commenced with the expansion of the education system, where middle-class young people found themselves removed from the adult realm of paid work. In explaining this, Firestone (1972) notes “If childhood was only an abstract concept, then the modern school was the institution that built it into a reality” (p. 81).
The emphasis on highlighting childhood and children as ‘different’ from adulthood and adults accelerated during the late nineteenth and early twentieth century. This period witnessed both the wholesale removal of children from the public sphere of paid employment, and the development of a philosophy that suggested that children required particular protection via the influence of an expanding social and health service. Childhood ‘came of age’ following the establishment of welfare regimes, which resulted in the birth of new powerful professional groups such as paediatricians and social workers. Informed by Piaget's notions of child development and competence, these professional groups emphasised the specificity of childhood and the vulnerability of children. Indeed, Mayall (1999) comments that “certain psychological theories” (p. 10) dominated the early debates concerning children and child development. At the centre of these psychological theories was the belief that children lack adult characteristics and, because of this deficit, adults need to help children become competent, fully functioning reasoning beings. On the one hand, these approaches regarded children as being in need of protection, while on the other they argue that children require educating and integrating into the world of the adult (De Winter, 1997). Such ideas have influenced the role of children in research and often the rationale that lay behind such research. Archard (1993) suggests that an enduring influence of these professional groups has been the establishment a broad framework, which has defined and legitimised childhood. This framework suggests:
Childhood is a status, which is the opposite of being an adult; Childhood can be regarded as a condition marked by the absence of adulthood; and Childhood is a stage on the way to full adulthood, where the gaining of adult status is an advantage to the individual.
Archard (1993) therefore implies that Franklin's (1995) notions of cultural and historical difference between national groups take place within these three areas. This is because childhood and youth is the result of a “dialogue between young people, their families, their peers and institutions of wider society” (Jones & Wallace, 1992, p. 4). Reflecting the ways in which this ‘dialogue’ may manifest itself, Archard suggests that divergence between countries is displayed in three main ways:
The boundaries of childhood – when it starts and ends; The nature of childhood – what precisely distinguishes it from adulthood; and The significance of childhood – how important are differences between child and adult.
These definitions bring us to Shamgar-Handelman (1994), who reflects Franklin's (1995) comments, noting that “every society crystallises its own set of norms, rules and regulations which dictate attitudes towards the category of its members defined as children” (p. 250).
These points of divergence in the experience of childhood, as explained by Archard (1993), Franklin (1995) and Shamgar-Handelman (1994), reinforce the idea that children will experience childhood differently across the EU or any other cross-national context. This is because the cultural, social, economic and political factors that are unique to individual nation-states define and regulate the boundaries of childhood. It is a situation exacerbated by the fact that whilst childhood ends at the point on which a young person reaches the formal age of adulthood some ‘adult’ rights and responsibilities accrue throughout adolescence. Indeed, the “period from birth to adulthood…[is] sub-divided into a number of different periods” (Archard, p. 26). Here there are “different needs, rights and responsibilities, being judged appropriate for the different age groups” (Franklin, p. 7.
The researcher, when conducting cross-national studies, needs to note that the stages between childhood and adolescence may also vary between countries. While at either end of the childhood spectrum there is infancy and adolescence, the cultural differences between nation states can result in the move from childhood through adolescence into adulthood taking place at different stages. In addition to this, the move from childhood to adulthood can have different meanings and implications. Not only do the ages at which rights to keep pets and enter a public house, for example, vary across European Union Member States, so do those rights related to marriage, leaving school, and homosexual or heterosexual intercourse. Such variation demonstrates that the perceived competency of a child is rooted in national contexts. Indeed, as Penn (1998) notes “childhood is not a universalistic phenomenon but a highly specific and local one” (p. 21).
The discussion above highlights the challenges that researchers have to confront when attempting to establish a cross-national research framework that recognizes the tensions surrounding childhood. However, when we extend the context and apply these issues to the subject of young people and migration, the situation becomes more complex. In particular, throughout their daily lives, migrant young people are required to straddle and balance the different experiences and expectations placed on children and childhood in the different countries (home and host-states) with which they are associated. This is because the route that young people find themselves embarking on, when moving from adolescence toward full citizenship, may be signposted by different phenomenon such as social and legal rights, depending, for example, on on whether a young person lives in Greece or Sweden. Thus, the experience of being a young person and becoming an adult citizen, and the differences in what this may mean between the Home-State and the Host-State, can be a difficult encounter for the young migrant to negotiate.
Toward a ‘Pan-European’ Understanding of Childhood?
While the discussion above recognizes that children's lives and the experience of childhood is fashioned by the various socio/ legal/ economic policies that individual nation states adopt, nation states are increasingly located within a globalized environment. In particular, in addition to structuring trade and financial regimes, pan-national and international bodies are also incrementally establishing a platform of socio-legal rights for nationals, which include children and young people. Some of these rights bypass national governments and are offered direct to the individual citizen. At the same time other universal rights are afforded to individual citizens as a consequence of national governments signing international declarations and conventions. Therefore, the socio-legal context that governs children's rights is simultaneously national and international. Faist (2001) reflects on this issue by focusing on the subject of citizenship and nationality. He argues that in globalized environments individual nation-states are no longer able to determine the socio-legal rights and responsibilities associated with citizenship and nationality, without recourse to other nations or to supra-national and international bodies. Faist (2001) implies that ‘nested citizenship’, means that aspects of citizenship, which an individual national enjoys, are located in different sites in order to facilitate a greater co-ordination of citizenship rights and to manage the global economy. If we extend Faist's argument into the realm of children, and specifically a new concept of ‘nested childhood’, as experienced by both non-migrant and migrant young people alike, then children may increasingly find significant points of commonality across national and cultural borders, in pan-national settings such as the European Union.
As Europe's citizens increasingly find themselves participating in a European social space where national-based socio-political and economic barriers have declined, to be replaced by a context defined by European wide economic and political structures, the question arises as to whether the potential trend, identified above, will accelerate. Will, for example, a pan-European sense of childhood emerge, fashioned, for example, by the policies, directives and decisions emanating from the European Commission, Parliament or Court of Justice? Such a scenario could hasten a modest realisation of Hazard's (1947) idea of a world republic of childhood.
Though a credible answer to this question will only come about after decades of closer integration, it is evident that supra-national bodies such as the European Union currently create barriers and problems for the researcher when conducting cross-national studies. Often these problems occur at the most fundamental level. Thus, though one of the most basic problems to emerge for the researcher exploring children and childhood is that Member States do not agree on a suitable age range that constitutes youth, and in which rights are consistently given at certain stages, there is a similar lack of consistency between and within European Union institutions. Various parts of the polity refer to the ages of 18, 25 and 30 years, for example, as appropriate boundaries of youth. Such a situation presents problems in the compilation and subsequent analysis of statistics relating to children and young people. It renders the work undertaken by quantitative researchers more than problematic, and suggests that it may even become a fruitless task. Indeed, Ruxton's (1996) analysis of the figures produced by Eurostat, the official statistical and information agency of the European Union, identifies children and young people as being absent, hidden or ignored in the data. The responsibility for this shortcoming, however, does not lie squarely at the door of the EU.
While the presence of cross-national data, collated by Eurostat, offers the researcher access to statistics, there is a need to remember that these figures derive from a variety of sources. These may include the various government agencies located in individual Member States. This situation means that within the EU, figures presented as European Union-wide statistics are, in effect, secondary data reflecting information collected by individual Member States for a variety of different reasons. Here, not only may individual governments not share the same purpose for collecting the information, they may not use the same variables, such as age ranges, particularly when collected, for example, under the guise of children of pre-school, primary, secondary and tertiary school age. These educational definitions will mean different things to particular EU Member States, and they will lack a consistent application of age ranges. Such a situation returns us to the theme of understanding the cultural context in which research data is gathered.
Primary quantitative data collected by the European Union is not exempt from this problem, even if such information relates to European Union policies that Member States are required to monitor. Ackers (1999) notes that whilst EU policy implies the notion of consistency in application of that policy across the Union, they are likely to be interpreted in what she refers to as the ‘recipient end’. For Ackers, the ‘recipient end’ is precisely the cultural context and the wider socio-political circumstances of the individual Member State. Such a situation suggests that one EU policy may have as many interpretations as there are Member States. A ‘maturing’ methodological approach to cross-national research within the EU has yet to address this dilemma.
Cultural Challenges in Qualitative Methodologies
Given the constraints that surround quantitative cross-national research, it would appear that qualitative studies have a greater chance of producing reliable and consistent information, particularly because the methodology rests on generating primary data. Indeed, reflecting the discussion highlighted above, cross-national qualitative methods have developed and matured over the past few years, both recognizing and quite often celebrating the fact that data is located within a specific cultural context. As a consequence, the ‘traditional’ ‘safari’ method of data gathering which has typically involved the researcher and/or the research team going ‘native’ in order to undertake research within particular national contexts has increasingly been discredited (Room, 1986). This is because such practice has a danger of producing a one-sided ethnocentric focus within the research findings. In its place, cross-national studies, based on mixed nationality research teams undertaking data gathering processes within their own national/ cultural contexts, have increasingly gained legitimacy and are now recognised as a better way forward for cross-national research. The team members in such projects can draw on their understanding of the specific socio-cultural setting in which the research is undertaken. Such an approach assists in the grounding of the research in terms of the research context, methodology and findings. Room (1986) offers a clear sense of direction, suggesting that cross-national research should “involve a network of innovators, engaged in local research and learning from each other through cross-national collaboration. It is he [sic] rather than any international expert who alone, is able to specify the relevant practical questions which his local work poses for that learning network” (p. 111)
Our project, which set out to examine the experience of young migrants who move across the European Union as a consequence of the rights offered by the European Union's Free Movement Provisions, adopted the strategy recommended by Room (1986). In particular, the cross-national research team comprised partners who were both nationals of, and based in, Greece, Portugal, Sweden and the UK. As a consequence they were grounded in the cultural contexts of particular national settings. This team worked co-operatively, devising and agreeing on the methodology adopted in the study. Having agreed on the nature of the sample group (recognising variables such as age, nationality and family composition) each partner was responsible for certain aspects of the research activity, not least identifying and interviewing a set number of young people. Finally, each partner was also required to tape record and transcribe his or her interviews, and have the transcript translated into English before sending it to the UK for analysis.
At first sight, such an approach appears to address and resolve many of the dilemmas and challenges identified above. However, as a consequence of the structure of the research team, the nations/ cultures represented, and the methodology we adopted, a number of surprising issues emerged, many of which will find a resonance in other cross-national studies irrespective of whether they take place in a European or other international context. Specifically, and as explored below, these differences highlight the cultural specificity of methodological approaches and, in particular, the different ways that methodology has developed and become embedded within the research traditions in individual national/cultural contexts. As noted below, the implication is that some methodological assumptions, when applied outside of the national context in which they have been developed, may be accused of representing ‘intellectual colonialism’. Here we have the notion that we know best how to do research, and that there are universal truths associated with methodological approaches which, though developed in and by one particular national/cultural context, can be applied in all national/cultural settings. Yet, it is important that researchers do not lose sight of the fact that though ethnography, grounded theory and hermeneutics, for example, have become recognised as distinct approaches within qualitative research, they may be understood differently, at different stages of development within individual national/cultural contexts, or even following different trajectories. Similarly, as is explained below, disciplines in some countries appear as rejecting the suitability of one approach while developing and extending another, setting guidelines and definitions of good/poor practice. As such, methodological approaches need to be understood within the context of the wider culture of the country in which researchers are located. However, given this scenario, there is a sense of inevitability for disagreements to arise in cross-national teams, in how ‘best’ to conduct qualitative research. Such disagreements became evident in the team exploring child migration in Europe. This was especially the case with regard to what questions to ask, and not to ask, children and young people.
The dilemma of whether to ask young people ‘Why?’ in response to answers they might offer provides an example of the problems the research team encountered when attempting to operate a methodology that is reliant on, and embedded within, conversation and dialogue. For those research partners based in Greece, Portugal, and the UK, asking ‘Why?’ was not considered as being problematic, especially given the assumption that our unstructured interviews were to adopt the philosophy of being a “conversation with a purpose” (Burgess, 1984, p. 102). Thus, Greek, Portuguese and British team members argued that if they did not fully understand what a respondent had said, or their rationale for saying something, ‘Why?’ seemed an obvious question to ask. Indeed, the term appeared as representing a ‘natural’ conversational response and a ‘typical’ form of enquiry that is evident in everyday dialogue. Yet, before we accept this viewpoint, cross-national researchers need to resolve a further a series of questions. These are associated with ‘Natural and typical for whom?’ Indeed, though unstructured interviews might adopt Burgess' philosophy, we must acknowledge that conversation, dialogue and the nature of the language we use are culturally determined and located within a cultural context. Indeed, what the Greek, Portuguese and British team members felt was appropriate was not shared by Swedish colleagues. Instead the Swedish partners found asking ‘Why?’ presented as a significant methodological problem. In particular, they acknowledged that though asking ‘Why?’ might be a natural response for some, it was not always phrased as a neutral form of enquiry, nor was it always polite. Indeed, they argued that it could be interpreted as questioning the legitimacy of the respondent's statement, and therefore ran counter to the purpose of the study and the theoretical underpinnings behind a methodology, which focused on understanding the lived experience of the respondents. Instead of asking ‘Why?’, the Swedish partners suggested that a response could be made in an alternative manner, such as ‘What are the reasons for this?’ or ‘Tell me more’. In discussing their position, the Swedish partners referred to the ethical codes of guidance that social science disciplines in Sweden have adopted. The partners suggested that in following the proposed methodology of asking ‘Why?’ the Swedish members would invite scorn, if not censure, from their colleagues.
Whether it is appropriate to ask ‘Why’ raises another set of questions, which all cross-national research teams might consider, and though these also relate to the cultural context of language and conversation, they also have an impact on the relationship between the interviewer and the respondent. Thus, the interviewer who asks ‘Why?’ might be using the term in an attempt to reinforce the notion that the relationship between interviewer and respondent is reciprocal. This reflects Oakley's (1981) contention that “the goal of finding out about people through interviewing is best achieved when the relationship of interviewer and interviewee is non-hierarchical” (p. 41). Here, the emphasis is on Burgess' notion of conversation, where a response to the question ‘Why?’ might also be expected of the interviewer. Again, Oakley clarifies this point when, in referring to Goode and Hatt (1952), she notes:
an interview is not simply a conversation. It is rather, a pseudo-conversation. In order to be successful, it must have all the warmth and personality exchange of a conversation with the clarity and guidelines of scientific searching’ (p. 33).
Equally, however, an interviewer who asks ‘What are the reasons for this?’ or ‘Can you tell me more’, might be using these turns of phrase as a ‘distancing strategy’. This aids the interviewer in maintaining impartiality, objectivity and the scientific searching that Oakley advocates (see Bargiela et al., 2002). Here, then, the emphasis is on Burgess' notion of ‘purpose’. Cross-national research teams have to find a way to resolve this dilemma, if only because the differences in the quality of responses generated may have an impact on the research findings.
The debate surrounding asking ‘Why?’ or ‘Tell me more’ cannot be divorced from that which considers the role of children and young people in society and, in particular, the nature of the relationship that exists between young people and adults (see Archard [1993] and Shamgar-Handelman [1994] noted above). Specifically, asking ‘Why?’ may appear in some nation-states as a common feature of the dialogue that takes place between wider society, adults (parents, teachers, social workers, and so on) and young people, where young people are constantly required to justify a standpoint, decision, or point of view. Here then we need to acknowledge that the nature of the questions that researchers ask young people may be phrased in a manner that reflects their status in wider society. It is, therefore, perhaps unsurprising to find that the Swedish researchers, based in a culture that is far more respectful to both youth in general and young people per se (as indicated, for example, by a plethora of social and legal rights that are denied to young people in other countries), argued that it was more appropriate to use a neutral and more polite form of enquiry.
In an attempt to resolve the dilemma created by whether the wider research team should ask ‘Why?’ or use a more neutral alternative, it was decided that it was better to accommodate the needs of the Swedish partners. However, in addressing their needs, the team could be accused of committing a further methodological misdemeanour. In particular, the compromise that was adopted raises a question as to the legitimacy of a methodological practice that has been informed by Swedish research ethics, when applied to research contexts located outside of Sweden and, therefore, outside of the cultural context in which this methodology developed. Clearly, cross-national research teams may have to consider adopting a methodological approach that permits variation in the application of methodological tools, which will also minimise the dominance of one national-based methodological approach and philosophy above another.
This cross-national team encountered other, similar, methodological dilemmas. Again, these reflect on the cultural assumptions that remain implicit within qualitative research methodology. However, these issues were not obvious until well after the data-gathering process had been completed. Thus, it was only when the UK team began to analyse the transcripts that it became obvious that we might have fallen, inadvertently, into the ethnocentric trap, traditionally levelled at the ‘safari’ approaches to cross-national research. This was because the analysis of our data was almost wholly undertaken by the British-based team members, an issue that finds a reflection in many other cross-national studies where projects are lead by one particular national-based partner in a cross-national team. At its most simplistic, the UK team members stand accused of being voyeurs, imposing a British perspective and a British understanding to data that had been collected in other cultural contexts. Fortunately, the other research partners had been involved in establishing the NUD*ist nodes and sub-nodes, generated from the original questions that we had collectively identified and agreed on. These nodes formed the basis of the analysis, and therefore the involvement of other members of the team was, to some extent, implicit in the process. However, in making sense of the information collected, the British team realized that this did not stop them from being caught-up in a classic ‘whispers’ type scenario. This was because British team members were the last in a potentially long line of other researchers involved in making sense of each interview transcript. In explaining this, it is worth highlighting the stages from interview to transcription and analysis:
The interview with the respondent. Transcribing by the interviewer or another person in the same country in which the interview. Translation into English by the interviewer, transcriber or another person in the same country in which the interview. Entry into NUD*ist and categorisation by a member of the UK team. Analysis by a member of the UK team.
Thus, when the original interviewer is included, those interviews completed in Sweden, Portugal, or Greece could have been processed by a total of five different people, and in two different cultures and languages, during the transcribing/ translation/ analysis stages. Such a scenario leads to a concern of whether the British researchers, when exploring data gathered in another EU Member State, heard the authentic and original voice of the child respondent or one filtered through another adult researcher. Here there is an additional danger surrounding whether the cultural context of meaning, as it is embedded in language, becomes lost, misinterpreted, or perhaps understood in another cultural context. Such a situation did not take place with those interviews conducted in the UK, which though often having two or more researchers making sense of the data, always remained in the cultural context in which the information had been collected.
Toward Cross-National Research Principles
The debates that surround childhood, cross-national research, and methodology are closely related and they demonstrate a number of tensions that researchers are increasingly required to address. There is a sense of irony that while cross-national research has accelerated and deepened our understanding of the cultural specificity of childhood, it appears as having done little to address the cultural specificity of methodology and reconcile the tensions that exist. Though Hantrais and Mangen (1998) and Room (1986) have provided a lead, this article has highlighted the point that cross-national research, whether qualitative or quantitative, has to grapple with some major problems. Here, the presence of supra-national institutions, such as the European Union, appears as deepening, rather than alleviating these issues. Given the challenges presented to the integrity of quantitative data, these institutions offer additional contexts, which need noting. These new contexts do little to minimize the methodological challenges that exist when conducting research across European Union Member States.
Qualitative methodology, when adopted by cross-national research teams, also has to balance a number of tensions. Many of these revolve around the question of ethics. If, as has been argued in this article, research methodology is shaped by the socio-legal and cultural context of the country in which it has developed, then ethics and ethical codes must also be regarded in the same manner. In this sense, it is possible to argue that there are no universal ethical ‘truths’ or universally ethical ‘ways of conducting’ research that are applicable to all national/cultural settings. This applies within the context of the cross-national studies within the EU and elsewhere. Instead, the most we can expect cross-national teams to agree on are shared principles, which may vary in their application and meaning when interpreted as methodological tools. After all, should researchers conducting a study in their country follow rules and guidelines adopted in another national context? Are cross-national teams able to make judgements concerning the efficacy of one methodological approach above another approach developed in a different context?
In concluding, the temptation is to paraphrase Franklin's (1995) contribution noted above, and apply it to the research context. Here it is possible to argue that definitions of methodology, as well as the varied methodologies which researchers' experience are social constructs formed by a range of social, historical and cultural factors.
Cross-national research teams may only begin to solve the questions and dilemmas that have been raised in this article when they address the division of work and responsibilities within projects. The notion of a cross-national team suggests that work and responsibilities will be divided equally. However, though the work involved in collecting the data is often one that is shared, data analysis is regularly conducted by team members in one national context. Costs (financial and time) normally prohibit a more equitable division of this aspect of the work. Thus, it is one thing for research to be conducted in cross-national settings, it is another for findings to be evaluated within cross-national contexts. The Sixth Framework, with its emphasis on supporting larger teams of researchers drawn from a wider number of European States, appears unlikely to resolve this issue.