Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Education. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 05 March 2021

Qualitative Approaches to Studying Marginalized Communitiesfree

  • Dana MoreeDana MoreeCharles University


Marginalized populations are by definition composed of people who have fewer possibilities and options in their lives than those studying them. This fact has to be reflected before, during, and after the research itself. There are many facets of this basic assumption. One of them is, how are marginalized perceived by the researcher? Are they helpless victims, or people who are able to tell their own stories? Another relevant detail is the personality of the researcher. When the researcher comes from outside the marginalized group, the key question is, which methodology can be best applied to give a voice to those who are marginalized? On the other hand, when the researcher is a member of the group being studied, the key question is how to achieve the distance necessary for analysis. There could be many more such relevant facets, but the quality of the final research product is partially determined by any number of decisions that are made during the planning of the research and the conducting of the research.

All of these decisions have methodological consequences. There are a wide range of qualitative research approaches, such as participatory research, autoethnographic research, narrative and biographical research, or traditional qualitative research based on interviews with representatives of marginalized groups. In the early 21st century, there has been a shift away from a top-down, outsider perspective that sees the marginalized as helpless victims and toward more participatory research designs that promote and give a space to the marginalized voice.

The common denominator of all these decisions is whose voice is being heard—does it belong to the marginalized group or to the outside world? Is it possible to overcome the boundaries between these two worlds? And what role does methodology play in this story?

Researchers’ Attitudes

Contact between the research professional and members of marginalized groups is always very sensitive, and a potentially fraught situation. At the same time it is a many-faceted contact. The aim of this article is to analyze those facets of the interaction that can be influenced by the researcher, whether consciously or unconsciously.

There are two aspects of the research process linked to and influenced by the attitude of the researcher. First, research on marginalized groups is often a delicate undertaking. To generalize, marginalization is a consequence of many problems, which are visible at both the individual and the societal levels. Quite often, these problems are linked to traumatic experiences. All of these experiences have a strong impact which we feel as human beings and with which we have to cope. The more reflective we are, the better we can deal with the situation at hand, and the more cognizant of the inherent pitfalls we can be, going forward.

Second, researchers have to seek answers to questions that they hold implicit answers to for themselves, based on their own personal attitudes and value systems, in any kind of research. However, research among marginalized groups is especially sensitive because of the clear power differences at play; such differences are an inevitable part of every type of research, but they play a particularly crucial role in the research of marginalized groups. Ideas and suggestions for working with values and cultural presuppositions might bring more clarity on how this issue could be approached. Specifically, the concept of cultural antagonism as it relates to our basic orientation on relationships and rules might be worth bearing in mind (Trompenaars, 1989). “Antagonism” here means that, in the moment when a decision must be made, we consider the most extreme options available, and make a decision which is closer to one pole or another (Hofstede, 2002; Moree, 2015; Trompenaars, 1989). I would like to offer this antagonism as the basic frame by which we can reflect on particular decisions we have to make in a research situation. The main question in the research of marginalized groups is if we tend to focus more on relationships or, on rules and outcomes.

Knowledge from several disciplines might be helpful as we consider this issue. I offer a connection between traditional research disciplines such as anthropology, pedagogy, and other social sciences, but at the same time it is worth embracing all disciplines that look more carefully at the interpersonal level of interaction, such as social pedagogy and social psychology. Based on this interdisciplinary perspective, I will try to answer the question of how can we conduct qualitative research in a situation where marginalized individuals and groups are the focus. How can we conduct such research in an ethical way? This means not reproducing marginalization, but rather supporting the empowerment of the marginalized, or at least ensuring that the research will not make the situation of marginalized individuals or communities even worse.

Consequences of Marginalization

When we think about research among marginalized people, we can start by asking questions to define and clarify this kind of contact. How do we distinguish marginalized people, and how does their situation influence the research?

Answers to these questions seem to be relatively obvious on the surface. We can all imagine that there are plenty of groups around the world that could be defined as marginalized. Ethnic, religious, and sexual minority groups, women, people from disadvantaged areas, and many others who have limited access to resources and limited possibilities to change their situation all come to mind. In these cases the situation is quite clear. But at the same time there might also be other groups who feel disadvantaged, or that they suffer from a kind of marginalization which might be difficult to perceive from the outside—nevertheless it exists and plays a role in the research.

However, in the case where marginalization plays a role, a specific situation creates a very special relationship between the researcher and the population being researched. This happens first if the researcher represents the stronger or dominant part of the society (i.e., those who are empowered), and second, if this dominance is extant when the research is conducted, and third, if this aspect of life experience—that is, dominance versus marginalization—is a question being addressed by the research itself. In this very specific context we have to be aware that the problem in knowing is twofold. On the one hand, we have to take into account the human relationship between the researcher and the researched on the level of personal interaction. On the other hand, we have to think about the distance between them, which is necessary for the researcher to come up with any reasonably neutral portrayal of those being studied. Both of these levels should be carefully reflected upon during the whole research process, from the initial planning through the final interpretation of the data.

In this article I invite the reader to engage the whole research cycle while reflecting on these issues. I will look more closely at what marginalization means and the consequences it has for every single step of the research process.

Planning the Research

Representatives of marginalized groups are not usually those who initiate any research on themselves (Guillemin & Gillam, 2004). And, marginalized people are not usually the ones who benefit from the research results. However, we see more attempts to reconsider these notions in the early 21st century (Rammelt, 2014). One of the important questions at the beginning of research planning is who benefits from the research (Bhopal, 2010; Whitzmann, James, & Poweseu, 2013). This question is an ethical one. Researchers have, relatively speaking, a great deal of freedom to plan research, and at the same time have well-developed mechanisms which control the ethical dimension of the research. Guillemin and Gillam (2004) discuss procedural ethics and ethics in the practice of research. Procedural ethics is a process of negotiating planned research with ethical considerations; usually this happens via ethical commissions. Many institutions have agreed to their ethical codes, which they use in the process of decision-making about research proposals. There is debate, however, about how helpful these codes are for planning research (Guillemin & Gillam, 2004). Moreover, particular countries differ in how obligatory this formal part of research is, and which institutions require what recommendations regarding these considerations. Ethics play a key role in practical research decisions, which we as researchers must make throughout the entire research cycle.

When we come back to the researcher, we can see two possible starting points in the first stage of research planning. First, the benefit for the marginalized group is often not integrated into the research as one of its main aims. In such a situation there is a specific responsibility of the researcher, which might be labeled as a “no damage approach.” Guillemin and Gillam (2004) declare that the main principle is respect and dignity for the researched population. No research should hurt anyone, and people should never be reduced to means or objects fulfilling the aim of someone else’s actions (the researcher, in our case). Guillemin and Gillam (2004) point out that this tension can be resolved when research aims are negotiated with their target population. When the aims are accepted by the researched community, this ethical tension might be less salient.

Second, the ambition at the beginning of the research is that it should improve the situation of the marginalized groups. Cox, Geisen, and Green (2008, p. 2) even ask the question, “what is the contribution of social research to social change?” Even if we plan our research with the overt aim to provoke social change, one danger in any kind of emancipatory research is that the benefit for the researched community might not be real and long-lasting (Levinson, 2017). Kurt Lewin (1946) suggests that there are many aspects to the research which should be included and negotiated so that the change or benefit is likely to be realized and permanent. A researcher has to make decisions about how representatives of marginalized groups will be integrated in the research design and planning, which stages they will be integrated into during the process of data collection, and how the researcher should ensure the validity of the research afterwards. The integration of the marginalized can be encouraged in many ways. They might define some plans for change together with the researcher, and based on the research results they might plan concrete steps which would help to bring the change into existence. But they can be also integrated as people who learn something during the research—for example, how to take photographs (Finney & Rishbeth, 2006)—and thus experience empowerment. Kurt Lewin (1946) uses the term “scientifically lower or higher research” to note when the researcher looks for a balance between scientific perfectionism on one side, and involvement of the marginalized on the other.

In both cases there is one danger which researchers can underestimate. Researchers are also people, with their own experiences and life stories. The fact that they work as researchers already means that they are representatives of those who have succeeded in the current structure. They have studied, they work at universities or other institutions, they have the resources necessary to conduct research, and they are able to ensure that their results will be published. This might be one reason why we may automatically have a tendency to see the underprivileged as helpless victims (Link & Phelan, 2001). But the reality of marginalized individuals and groups is multifaceted and complex. In some aspects of their lives they might be helpless and marginalized, but at the same time there might be other aspects of their lives in which they might be even stronger or more successful than the researcher. Levinska and Doubek (2017) offer the example of Alice, a woman who experienced a very difficult home environment in a poor and rough neighborhood in the Czech Republic, but succeeded in attending school without any support from her family, and got a job working for an NGO helping others to escape deprived situations. Their study ends with Alice in process of decision making to leave the NGO, which does not work in a sensitive manner, in her eyes. Such a decision can be appreciated as a very good one even by researchers themselves, who might be also very unsatisfied with their employer, but might find it difficult to admit that they are leaving for ethical reasons.

We can see that there is always potential for change, empowerment, and improvement, as many concrete examples from a wide range of research show (Boal, 1979, 1992, 1995; Conrad, 2004; Freire, 1974; Wrentschur, 2008). This potential should be kept in mind when thinking about discrete individuals or groups, because otherwise their marginalization will only be reproduced. What is interesting, when we look at research among marginalized groups on the international level, is that we find much more often a perspective which describes troubles, rather than stories about success or empowerment. This means that we know much more about the reasons underlying the difficult situations of the marginalized than we do about the strategies used to overcome those trials. In the early 21st century, however, research has tried to increasingly stress the perspective of empowerment. Gkofa (2017), for example, published research on Roma students at universities that offered descriptions of the steps which were necessary for them to overcome their disadvantages. She interviewed successful young people of Roma origin who were able to obtain university degrees despite the fact that their home environments were very deprived. In her research, she specifically asks them about the factors which made this success possible. Levinska and Doubek (2017) have been doing multiyear field research in a marginalized community in a small village in the Czech Republic, where they reveal the perspective of marginalized people from poor areas who went through the process of empowerment: in doing so, they were able to alter the structure of their community and improve their living conditions. Finally, they were even able to start changing the perceptions of those in positions of power—in this example, representatives of the NGO, who brought their expertise at the beginning of the emancipatory process, but then took a secondary role to the marginalized people themselves.

Decision-Making and Ethics

Working among marginalized groups means doing responsible research, where the researcher is deeply aware of the consequences of what he or she is doing. The researchers should reflect on their own positions, beliefs, and relationships, and should be aware of the influence of their backgrounds and actions on the research. This is what is meant by the researchers’ positionality (Milner, 2007), an awareness of which leads to both personal reflection and responsible-decision making.

In order to oversee the particular steps in the research, many decisions have to be made, and these decisions should also be morally responsible (Cox, Geisen, & Green, 2008). But what does this actually mean in practice? Fischer (2008) points out that we should distinguish between ethics and morality. Ethically responsible behavior has three dimensions (p. 1): it is recognized that one’s behavior influences the well-being of other human beings; behavior should be intentional and justifiable; and finally, behavior should be based on free, personal decision-making, rather than a reaction to outside pressures. Morality means, according to Fischer, that we should reflect on the accordance of our behavior with these principles.

But realistically, in the field, the researcher is not usually free from outside pressures. Researchers as well as researched communities are settled in their social contexts, which are not neutral or value free. Social environments always produce sets of expectations and frameworks of conditions. From the perspective of the researcher, there are many pressures linked to the system in which the current scientific debate takes place. There are systems of funding of research institutions, systems of funding for individual researchers, systems of reporting and evaluating research outcomes and, last but not least, the systematized pressure of the amount of publication expected from researchers (Linkova & Stockelova, 2012). Alongside this pressure generated by academia and research institutions, research among marginalized groups is affected by political contexts pressures and perhaps even pressure from those in positions of power (Stockelová & Grygar, 2008). All of these are outside pressures which do not have anything to do directly with the actions of the marginalized group. They are inherently linked to the social environment of specific researchers, and they create a set of dilemmas and decisions, with which the researcher and researched have to contend. Several aspects of this dilemma-influenced decision might be worth mentioning in this context.

We all make many decisions on a regular basis; the question is how clearly we as individuals are able to perceive our real motivations for these decisions. A decision can be motivated from the outside (heteronomous motivation), or from within (autonomous motivation) (Anzenbacher, 1994). This assumes an ability to distinguish what is right and what is wrong, as well as an assumption that everybody knows that doing good is desirable (Anzenbacher, 1994). But this basic distinction has one peculiar gap. The individual’s perception is always influenced by the outside world, which creates feelings of right and wrong, which are produced again by the shared feelings of many individuals. Individuals are influenced by general feelings about what is right and what is wrong, and at the same time the researcher recreates this perception through his or her actions (Anzenbacher, 1994). This means that decision-making about right and wrong in a particular research project must be done carefully, and aspects of autonomy or purposefulness should be integrated into this decision-making process.

If we apply this understanding to our environments, we face many possible questions. How does the structure of funding influence our research? Are we able to finish research in marginalized communities where we have personal contacts among individuals even if we do not receive funding for later stages of the research? Is it important which financial sources we use for our research? Does something like “dirty money,” so to speak, exist in the context of the researched community? If so, how do we deal with this? How much distance do we need from the political environment and from particular actors in this environment? How much do we give voice to the researched population, even if it makes the research more difficult for the researchers, funders, or other powerful groups? How open are we to challenging our preconceptions? There may be many other such questions, most of which we may not be able to answer. But our awareness opens a space for responsible decision-making.

The process of decision-making has another dimension. In order to be able to take responsibility for particular steps of our work, we also need space for a healthy process of decision-making. Decisions should not be made under pressure, and we should take as much time as we need to make them. Ideally, we also have the option to make partial decisions, which we can reconfirm or renegotiate later. Lewin (1946) calls this a process of permanent reflection reconnaissance.

Power Differences

As noted already, power differences are an integral aspect of any research, but this issue is especially relevant in research among marginalized groups. Power differences might be based on many aspects of our lives, and they are integral aspect, of discrimination and prejudices, which very often influence the lives of marginalized groups. These processes are a consequence of how stigma works. There are some descriptors of our lives which are perceived by more powerful parts of societies as more welcome than the others. Researchers might also have some experiences with stigma themselves, or they may be a member of a group that might be perceived in the research situation as stigmatized persons or less powerful (e.g., white women doing research among dark-skinned, poor, male workers, or female, childless researchers among rich mothers of autistic children). The existence of stigma might influence our research; therefore, it is worth examining how stigma works.

Societies are not equal; each society has developed its own way to distinguish among its people, because some desire advantage over others (Anthias, 2011; Moree, 2015). The processes of labeling or stigmatization are universal, and all societies or groups practice them. What differs is the nature of these labels and stigmas. The larger probability of stigmatization is linked to groups or individuals who bear visible difference (Girard, 1989). But stigmas are based on a variety of factors. Link, Bruce, and Phelan (2001) list some typical aspects of stigma creation: stigma is based on differences, not similarities; and its function is to divide people into two basic categories—us and them. Stigma is very often the reason for status loss, and its consequence is limited access to social, economic, and political power. In the research situation, the question is if this “us-them” perspective exists between the researcher and the researched. If, for example, we conduct research in the Roma community, our research will be affected in different ways if we ourselves also belong to the Roma community, or if we are perceived as outsiders, or if we represent the powerful majority population in the eyes of the researched community.

Stigmatized persons or groups as well as those who stigmatize can be characterized by some descriptors. What we know about the stigmatized person is that stigmatization is very often linked to a particular person’s self-understanding, developed throughout her or his life and as a result of socialization. The important thing is how others react to the person (Crocker & Major, 1989). Stigma does not always have the same power; the closer the people are who label, the worse the consequences of the stigma are (Crocker & Major, 1989). Stigmatized persons very often internalize part of the stigma; it becomes part of their integral world, their personality (David & Derthick, 2016; Pyke, 2010), and they develop a wide range of strategies to cope with it in daily life (Allport, 1954; Crocker & Major, 1989; David & Derthick, 2016; Pyke, 2010).

Oppression based on stigma is multilayered. It can be perceived on an interpersonal, structural, or cultural level (Link, Bruce, & Phelan, 2001). People who stigmatize others also bear some typical character traits: stigmatization is, in fact, a transmission of our own frustrations and uncertainties onto others (Allport, 1954; Jung, 1998; Perls, Hefferline, & Goodmann, 2004). It also has a very important function for representatives of those who have more power. Stigma enhances the feeling of solidarity among in-group representatives (Douglasova, 2014). In the research setting, stigma can be produced by the researcher toward the researched community—or vice versa. To give an example, when the researcher conducts research among Roma people and he or she is in contact with this group for a long time, the social barriers weaken and people are not that careful about what they say and how they say it. It might happen that they start to make ugly jokes about the majority population, represented in such a moment by the researcher, who in turn might be accused of discrimination against Roma.

When a researcher conducts research among marginalized people, he or she becomes a player in the power game, and should be aware of these processes. Lewin (1946) suggests that we should automatically try to discover who the people with power are during research among marginalized groups—including the contracting authority for the research. Power differences are thus present on many levels. We should try to discover and be aware of all of them. We might also analyze how stigma works in a community, who is stigmatized, who labels, and who is labeled. Furthermore, it is helpful to discern what real consequences marginalization has, and who profits from it (Moree, Vavrova, & Felcmanova, 2017).

Rammelt (2014) adds one more element to this broad understanding of how stigma works. There are not only the power differences that we can analyze at the beginning of the research; there might also be new ones which occur during the research. These new power imbalances might be caused, for example, by the fact that research also changes the social environment of marginalized people, and this environment might react to these changes during the research. Specific researchers differ in the extent to which they integrate the study of power differences into the research design. For example, feminist researchers are quite sensitive about these issues (Vaňková, 2010). However, power differences also affect the researcher, and the only way to cope with them is by reflecting on them and on the real power positions and dynamics of the researcher and of the researched communities.

Research Methods

We have noted the basic antagonism between orientation to rules and orientation to relationships as a meta-position which might be helpful for researching marginalized groups. Many decisions on the scale between these two positions are demonstrated in the research methods that we use. This is what we really do in the field. We might imagine that this basic decision-making includes several aspects which might be also placed on scales. And decision-making means that we move on these scales between their poles. We might think about the aspects of the chosen methodology shown in Table 1.

Table 1. Scale of Decision Making

no change -------------------------------------------------------------------------------------------------------- change

no participation of respondents -------------------------------------------------------------- a lot of participation

short amount of time spend in the field --------------------------------------------- continual work in the field

external concerns ------------------------------------------------------------------------------------- internal concerns

In other words, there are many ways to conduct research. What we know about research among marginalized communities is that if we want to integrate them, it is better to choose research designs that enable closer and longer contact with the researched population and that also include their perspective when planning and conducting the research. These scales might help in thinking about various methodological aspects. Let us go through several examples of methods which are often used for this kind of research.

When we stay on the left side of the scales, we usually make choices for traditional qualitative research based on interviews. These might be shorter or longer interviews, or even interviews based on oral history, in which we ask prepared sets of questions and respondents are those giving answers. Contact is limited in time and respondents are not integrated into the process of preparing research questions or interpreting their results.

The more we move toward the right side of the scale, the more we have to integrate respondents into the research, and the more we have to look for appropriate research methods to accomplish this. Our concerns are more situated among the researched people than concentrated on outside interests or pressures (Freire, 1974; Rammelt, 2014).

In addition to traditional qualitative research methods, we can choose from among a wide range of other qualitative research designs. When we try to use the scale, we can point out that there are researchers who perceive narrative research methods as more sensitive toward marginalized people then traditional interviews. Henry (2015) identifies researchers like Etter-Lewis and Foster (1996), Okihiro (1984), or Reinharz (1992) as some who employ narrative research methods. The main difference between traditional interviews and narrative research is that in the former, the process of inscribing or assigning meaning is in the hands of the researcher, but in the latter, participants are given a voice, and bring their own understanding and interpretations of their experiences into the interviews and into the research as a whole (Trahar, 2009). In narrative research, the focus is on interactions with and the actions of respondents, which are seen as a kind of story which the respondents tell. The members of the marginalized community in question are those giving meaning to their experiences and interpreting their perception of reality (Cortazzi & Lixian, 2009). This kind of research seems to be especially useful in comparative settings, where different meanings can play crucial roles, and also in settings where people act communally and these actions can be perceived as stories which they tell, for example in schools (Trahar, 2009; Cortazzi & Lixian, 2009).

Ethnographic research is another research design that enables more intensive contact between the researcher and the researched. This kind of research is suitable in especially complex environments where there are many interactions, like schools, refugee camps, or neighborhoods (Gobbo, 2011, 2017). Sometimes this kind of research is used because it gives the researcher an opportunity to spend much more time with the community. Or sometimes another research design is chosen, but the on-the-ground circumstances dictate that ethnographic research must be conducted. Researching, for example, refugees in Czech refugee camps is almost impossible by means of interviews. From 2000 to 2002, when the country experienced the highest amount of refugees in contemporary history, the rate of accepted applications from refugees was approximately 1%, the consequence of which was that any kind of interview was impossible. This is because interviews were perceived by refugees as an influential tool, the consequence of which, in their eyes, would very probably be expulsion from the country. In these circumstances, ethnography is one of few possibilities to describe their situation (Moree, 2008).

Participatory action research (PAR) is a valuable methodology when it comes to working with marginalized communities. PAR escapes many of the disadvantages of traditional research designs. It is a way to integrate the researched population, to give them a voice and to do research collaboratively. Lewin, who is perceived as a founder of this approach, stresses change, which is one of the main aims of such research. He points out that constant reconceptualization and reflection throughout all stages of the research are necessary. He suggests several questions that the researcher should continually ask: what is the present situation? What are the dangers? What shall we do? (Lewin, 1946, p. 34). However, there are other authors who argue that participatory action research design is not a method that involves the researched population in an appropriate or adequate way. The crucial question is how the research is conducted and how the actual participants feel. If they have a feeling of being integrated, empowered for change, or even of changing the situation themselves, then this is a sufficient level of participation (Levinson, 2017; Rammelt, 2014; Whitzman, James, & Poweseu, 2013). Levinson (2017) points out that participatory action research can be successful only if its ideals are shared with members of the community. The process of negotiating research goals is continual and difficulties may occur not only between the researcher and researched community, but also within the research team.

In autoethnography, researchers study their own participation in lived reality. In this type of research, the researchers are studying their own experiences, but at the same time they are representatives of the researched group’s wider experience, even though their own personal experience might be quite different theirs. This happened, for example, in Czechoslovakia after the political changes of 1989, when a group of sociologists started a research project based on their ethnography, where they tried to analyze and understand the profound and rapid societal changes after the fall of communism (Konopasek, 2000). Another example might be Henry (2015) with her analyses of being a high school teacher and researcher, as well as a representative of a visible minority, at the same time.

Then there are the cases when nothing works. Sometimes a researcher can face a situation when reality is so complicated that no known research design seems feasible, and a new approach dictated by the research environment has to be developed. In these situations, the researcher can combine a wide range of methods (Finney & Rishbeth, 2006), or he or she can create his or her own research design.

This section closes with an example of how this can happen which is drawn from personal experience. I am a member of a group in the Czech Republic that uses the theater of the oppressed as a method for working with a wide range of people from a broad spectrum of marginalized groups. Action research design is very often used in the field of theater of the oppressed, as we know (Conrad, 2004; Wrentschur, 2008). However, when we tried to apply this design in a traditional way, we were never successful, for several reasons. First, participants were not sure what they would like to solve by the use of the theater of the oppressed at the beginning of our cooperation. The topic of their performance grew slowly based on group processes, and facilitating the group process appeared to be much more effective than negotiating common goals, which took a long time at the beginning. Second, the first night of a new performance when a theater piece was performed in front of the audience was always a kind of breaking point in the whole process of the group work. Participants experienced empowerment and they started to interpret their story in a different way afterwards. Third, the audience and the solution they offered also changed the participants’ perceptions of their original story. The whole process was so multifaceted that it was almost impossible to capture all its aspects by means of a combination of field notes and interviews.

After several attempts with different groups it was clear that being able to go deeper, to tackle theories produced by the participants themselves and then confront them with the responses offered by the audience during performances, needed a new kind of methodological perspective. After many unsuccessful attempts we finally found a method that enabled us to integrate all these layers. We divided this kind of participatory research process into two stages which were communicated with participants. Group work leading to the formulation of a common story and preparation of a performance was, from the research perspective, perceived as a process of formulating a research question. This stage was based on ethnography. The theater performance was then perceived as a research question that was being asked by the marginalized community of the audience. Interventions of the audience during performances were perceived as the answering of the research questions. Performances and audience interventions can be taped and coded by traditional open coding methods (Moree, Vavrova, & Felcmanova, 2017). Working with marginalized groups can require creative thinking about research design, and a shifting understanding of who the researchers are, who the research participants are, and who the audiences are.

Contact and Identity—We Are Both Human Beings

One of the specific features of research among marginalized groups is the fact that the researcher and the research participants are in a very sensitive situation, and would probably not meet each other if the research was not being carried out. Research defines the situation. But at the same time, when the researcher makes the decision to spend a certain amount of time in the field and meets with research participants over a period of time, he or she may make a decision to not be outside ,but rather be inside, the situation of the marginalized (Bhopal, 2010). The question is what the researcher does with this situation, and which aspects of the situation influence the outcome.

Identity Issues and Trust Building

Henry (2015), Milner (2007), and many others suggest that the identity of the researcher influences the relationship between the researcher and the participant to a great degree. The question is how much the fact that the researcher bears some marginalized features (e.g., skin color in research among ethnic minorities) influences the outcomes of the research. This important issue can be approached by reflecting on the identity of both the researcher and the researched.

Bhopal (2010) suggests that there are many layers of identity which may play a role in this situation. This goes hand in hand with the concept of intersectionality (Anthias, 2011; Gobbo, 2004). Identity is a very complex topic. However, we can look at some interpretations which help us to understand what happens in contact with people where differences play a role. Howard (2000) provides an overview of understandings of identity in social sciences from a historical perspective. Her point is that ethnic identity was perceived as the main layer of identity for a long time. This perception is called culturalism (Leeman & Ledoux, 2003; Moree, 2008), or the cultural standard model (Bittl & Moree, 2007). However, there has been increased interest in multiple-identity approaches, or intersectionality, in the early 21st century, where ethnicity is perceived as only one part of identity, which is accompanied by many other factors including social status, religion, sexual orientation, experience of marginalization, and gender (Gordon, 2005; Howard, 2000; Moree, 2015; Samovar et al., 2013). Bauman (2004) notes, moreover, the fluid character of identity, which may change across a lifetime. The situational aspect of identity means that in every single situation, other layers of identity are important, and these are very much based on the specific situation and roles that are at play in a given moment (Moree, 2015).

Applying this understanding of identity in a research situation, it is evident that the contact between researcher and participant can be viewed from many perspectives. On one hand there are two people, where the researcher is very often representative of a power position or institution, and the participant is the marginalized one. On the other hand, they both may have shared sets of experiences—like the experience of racism or sexism (Bhopal, 2010). Shared experience might be an important part of the contact during the research and thus might affect the research process as well as outcomes. Exploring shared experiences can help in the process of trust building, which is an important part of the relationship between researcher and participant.

Finney and Rishbeth (2006) suggest that in research where respondents can benefit from the research (e.g., learning to take good photos and then organizing an exhibition), the entire experience of contact between researcher and participant is divided into several stages. Encouragement is followed by trust building, and only then might discussion also be fruitful. Trustful and open communication are necessary conditions in this process. Finney and Rishbeth (2006) suggest one more perspective, which brings us back to the question of autonomous and heteronomous motivation. This is a question of who benefits from the research, and what happens when the expectations of the researcher and the researched differ. Trusting and open communication might also mean a loss of some data, which is the risk in research based on a high quality of trust between researcher and participant. This can happen when the participants in the research mention some details which they do not want to be published but which, at the same time, may be important for understanding the whole research context. In such a situation the researcher must address the dilemma of how she or he will balance the needs of the research with concern for the relationship with the researched.

When marginalization itself is a topic of research, there are additional important issues of identity and trust. There might be a general agreement that marginalization is an issue in a particular situation, but there can be significant differences in how the reasons behind that marginalization are interpreted. In research among Roma (or other marginalized groups), the fact of being Roma is very often perceived by the researcher as a reason explaining behavior, actions, or habits, but it is not necessarily recognized as such by the marginalized community members themselves (Setti, 2017). For example, is the fact that mothers are fearful for and protective of their children necessarily a consequence of living in disadvantaged areas, or is it a feature of all mothers, Roma and non-Roma alike? Interpretation of such data is very dependent on the flexibility of the researcher and his or her openness to a wider range of interpretations (Hammer, Bennett, & Wiseman, 2003; Van der Zee & Van Oudenhoven, 2000, 2001).

Negotiations About Needs

Any kind of contact requires continual renegotiations of how the contact should look, so that both sides are satisfied. The more uncertainty there is, the greater the number of difficulties that accompany this process of negotiation. The same can be said for the research situation.

When Vaňková (2010) started her ethnographic research among the Aromanians of Bulgaria, she wanted to be very open and participatory. However, she experienced several incidents which led her to question how open her position actually was. She reflects on the fact that power balances shift during the research process. At the beginning of the research, the researcher is in a position of power, but as she grows closer to the participants and spends more time in the field, she is drawn further into the net of the expectations and assumptions of those she is studying. She points out several of these, such as the expectation that she would record Aromanian history for future generations, mediate outside interests in this small community, manage their representation at the international level in the research community, and bring books in their language into the community. They asked a lot of the researcher, and led her to question how legitimate avoidance of these requirements by the researcher is.

Working with marginalized groups and power imbalances always mean an increased necessity to negotiate the needs of both sides. Such a negotiation is not only important at the outset, but during the whole research process, especially when the situation unexpectedly changes. The question is: what do we know about our needs, and how can we negotiate them?

The situation of cooperation with a new person accompanied by power imbalance always causes uncertainties, and on both sides. Many things have to be clarified, and this process should lead to decreasing fears and uncertainties, and increasing trust and fulfillment of our needs. However, this process might be sometimes relatively difficult. Let us conclude with what we know about the link between fears and needs. There are many needs described in the literature, such as the needs for certainty, belonging, transcendence, and self-confidence (Maslow, 1943), creativity (Foucault & Chomsky, 2005), meaningfulness (Frankl, 1997), love, acknowledgment, and autonomy (Bittl & Moree, 2007). The relationship between two people is trustful, safe, and fruitful when the needs of both people are fulfilled. However, first contacts between people very often lead to an increase in fears and uncertainties on both sides. The problem might be that when we act based on our fears, we usually adopt at least one of three strategies: escape (the researcher is not clear about the research purpose, s/he does not communicate clearly his/her vision, etc.), fight (researcher stays higher up on the scale of power imbalance and is not successful in trust building), or accommodation (researcher tries to fulfill the wishes of the participants). The reason it is so difficult to avoid these strategies is that very often we feel our fears as much stronger than our needs in a stressful situation. That is why a strategy was developed which might help to clarify the link between fears and needs and might help to manage these initial, difficult situations. If we look carefully at both fears and needs, we can see that they create an interacting set. When we are unable to fulfill our need for feeling oriented and safe, a fear of the unknown appears. For another example, lack of acknowledgment might bring a fear of judgement. Bittl and Moree (2007) suggest coupling fears and needs into pairs, which should increase orientation in the stressful situation as in Table 2.

Table 2. Pairs of Fears and Needs

Fear of

Need of









Emptiness/death/loss of meaning


This approach suggests how we might cope with uncertain situations not only at the beginning of the research, but also during it. Such an approach requires an ability to reflect on the part of the researcher, who takes care not only of the research and collecting data, but also of the process of negotiation of particular research steps. The fact that we know how fears and needs are coupled can help when we feel our fears more strongly and are not able to feel what we need (Moree, 2015). This approach is very strongly influenced by intercultural communication strategies, which might be helpful in research stages where communication is crucial for conducting research and for outcomes.

Implementation of this approach into the research situation might mean several steps. Negotiation about needs and fears requires a kind of permanent ability to reflect on what happens during contact. This is the case not only at the outset of the research, but also during the whole period of the research. We can permanently negotiate about fulfilling our needs and ask about the needs of research participants. This model has one further aspect which might be helpful in specific situations. The needs outlined here can be fulfilled only together with other people. As soon as we start to have feelings of dissatisfaction during the research, we can ask about and renegotiate the needs of both sides. Such an approach gives us more possibilities to react to unexpected developments during the research.

Interpretation of Data and Outcomes of the Research

The final part of the research process evokes the same questions as the initial stage: how strongly is the voice of the marginalized people represented, and how can research results influence the situation of the marginalized group? Last but not least, we have to ask about the reliability and relative interpersonal distance of the researcher.

There are two perspectives which we might entertain during this stage. The end brings us back to the beginning. In other words, what we negotiated at the beginning should be done at the end as well. We can use our scales of no change vs. change; no participation of respondents vs. a lot of participation; short amount of time spent in the field vs. continual work in the field; and external concerns vs. internal concerns. The voice of respondents should be integrated to the extent we negotiated with them at the beginning of our contact. When the expectation and agreements went in the direction of a “no-damage approach,” participants would probably not be integrated into the process of interpretation of data at all. The more participation, the more the voice of the marginalized needs to be heard in the final stages of the research.

Research among marginalized people very often means powerful stories and strong emotions on the side of the participants as well as the researchers. It might make interpretation of collected data more difficult, because it might be more difficult to have some distance from it and from the particular situations, which might be specific to this kind of research.

To resolve this problem we can distinguish between two levels of research activity: contact with participants, and the mental activity of the researcher afterwards. Individual contact is very strongly influenced by the processes of negotiation, trust building, and sensitive and ethical working with every single human being participating in the research. By means of these interactions the researcher gathers information, data, field notes, and other sources of information, which he or she uses later to finalize research outcomes. We can imagine this process through the example of a class in a school. First of all we picture particular children—we have to be in contact with them, carefully look at them, and try to capture their features and their stories. We are in deep human contact with these people and we let them explain to us their perspectives, and share their stories. In the final stage we need to draw the whole background, the whole picture, and to interpret our findings. We can do this not because we are better or more intelligent than the respondents are, but simply because we talked to all of them and so we have much more data than they would have. Distinguishing between these two levels is very important for being able to get some needed distance from the gathered data.

Rammelt (2014) suggests that we should also be mindful of what strategies might be implemented as a consequence of our research. These can be brought in from the outside, or they can grow out of the research data. In his view, interventions that result in research outcomes should be defined in dialogue with the local people themselves.

Finally, we must come back to the question of the link between research and the societal environment of the researcher (and researched). Lewin (1946) points out that research outcomes should not be automatically perceived as knowledge that is transferable into political decisions or changes. The conduct of the research as well as further planning of its impact should be carefully negotiated with those who are involved in the topic that was the main issue of the research. These negotiations should occur not only at the level of the researcher and the researched community, but in the case of changes on the level of politics, more information, more factors, and more actors should be included. Every single situation is different, but at the beginning and at the end of the research, the researcher might also think about negotiations with powerful and influential people who might be able to use the research results.


In 2009 a group of neo-Nazis attacked the house of a Roma family in a small town in the Czech Republic in the middle of night. They threw a Molotov cocktail into the house, which caught fire. The family escaped, but their two-year-old daughter, Natalka, suffered burns over more than 80% of her body. She almost died and still suffers the severe consequences of these events. There was a lot of support for the family in the Czech population, but at the same time the family experienced a lot of hate crimes. Some years later the situation calmed down but Natalka and her family still suffer and still need help. In 2017 the Slovak politician, sociologist, and former dissident Fedor Gal decided to use her story as the basis for a theatrical performance. He was involved in the story of Natalka from 2009. When he reflected on his experience on the website, he tried to summarize what he had learned:

After several years of participation (in the story of Natálka) I understood one key issue: even the word ‘to help’ is perverted. Either you find a way of relating to people like these, they become your friends or partners, we do it all together, or it is just ‘top-down’ . . . It took me years before I could understand that the so-called ‘solving of the Roma problem’ is nonsense. I already know that that there is a limit to where you can intervene. If you intervene without relationships and without knowledge of the issue, you make invalids of these people. That is one borderline. The other is that you have to understand that they are not your clients in a psychiatric clinic. They become your friends and they are equal or you leave it. That’s the way things are.

There are many differences between creating a theater performance and doing research, but in both cases we try to capture our social reality so as to be able to understand it better. To understand how marginalization works and what we can do to overcome it seems impossible without the building of relationships. When we come back to the antagonism between an orientation to rules or relationships, we always make the decision about which side we want to stand on. My conclusion is that research among marginalized groups should always incline to the side of relationships—relationships which are negotiated and reflected, but which mean real and meaningful relation to the other and to the self. To be able to be in such a relationship, a reflection on one’s own purposes, ideas, and attitudes might be helpful.

Contact between the researcher and the marginalized people being studied is a very unusual situation. However, when the whole research process is divided into stages, and when the researcher consults with their own conscience, the decision-making process is easier and more fulfilling. This is true especially for the first and the final stages of the research. In these stages, the researcher analyzes his or her own position, the possibilities for the research, the way he or she would like to work, and the consequences of the work.

The middle part of the research process is about the contact between the researcher and the researched community. In this stage the main focus is on the continued contact, the trust-building process, and the ability to communicate and work with those who tell their stories.

The common denominator of all of these stages is the ability to be reflective and truthful: to oneself, to those who we perceive as marginalized; and to the wider social context, which usually is governed by those having power. I conclude by citing Barbara Santoz, a sociologist working through the theater of the oppressed, who says, “It can sometimes happen that we wander in the fog, but if we know where we want to go, we are on the right path. Direction is what defines uncertainty. If the aim is clear, mistakes made during the journey and momentary uncertainties do not endanger the certainty of the journey. If the direction is not clear, no mastering of techniques will ensure a good result” (Santoz, 2015, p. 37).


This publication was supported by the The Ministry of Education, Youth and Sports - Institutional Support for Longterm Development of Research Organizations - Charles University, Faculty of Humanities (201X), e.g. (2018)

Further Reading

  • Cox, P., Geisen, T., & Green, R. (Eds.). (2008). Qualitative research and social change. New York, NY: Palgrave Macmillan.
  • Finney, N., & Rishbeth, C. (2006). Engaging with marginalised groups in public open space research: The potential of collaboration and comined methods. Planning Theory & Practice, 7(1), 27–46.
  • Guillemin, M., & Gillam, L. (2004). Ethics, reflexivity, and “ethically important moments” in research. Qualitative Inquiry, 10, (2), 261–280.
  • Henry, A. (2015). We especially welcome applications from members of visible minority groups: Reflections on race, gender and life at three universities. Race Ethnicity and Education, 18, (5), 589–610.
  • Lewin, K. (1946). Action research and minority problems. Journal of Social Issues, 34–46.
  • Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363–385.
  • Milner, H. R. (2007). Race, culture, and researcher positionality: Working through dangers seen, unseen, and unforeseen. Educational Researcher, 36, (7), 388–400.
  • Rammelt, C., F. (2014). Participatory action research in marginalised communities: Safe drinking water in rural Bangladesh. Systemic Practice and Action Research, 27, 195–210.


  • Allport, G. (1954). The nature of prejudice. Garden City, NY: Doubleday Anchor Books.
  • Anthias, F. (2011). Intersections and translocations: New paradigms for thinking about cultural diversity and social identities. European Educational Research Journal, 10, 204–216.
  • Anzenbacher, A. (1994). Úvod do etiky. [Einführung in die Ethik]. Düsseldorf, Germany: Patmos Verlag.
  • Bauman, Z. (2004). Identity. Cambridge, U.K.: Polity Press.
  • Bhopal, K. (2010). Gender, identity and experience: Researching marginalised groups. Women´s Studies International Forum, 33, 188–195.
  • Bittl, K-H., & Moree, D. (2007). Abenteuer Kultur; Transkulturelles Lernen in der deutsch-tschechischen Jugendarbeit. Plzeň, Czech Republic: Koordinierungszentrum Deutsch-Tschechischer Jugendaustausch Tandem.
  • Boal, A. (1979). The theatre of the oppressed. New York, NY: Urizen Books.
  • Boal, A. (1992). Games for actors and non-actors. New York, NY: Routledge.
  • Boal, A. (1995). The rainbow of desire. London, U.K.: Routledge.
  • Conrad, D. (2004). Exploring risky youth experiences: Popular theatre as a participatory, performative research method. International Journal of Qualitative Methods.
  • Cortazzi, M., & Lixian, J. (2009). Asking questions, sharing stories and identity construction: Sociocultural issues in narrative research. In S. Trahar (Ed.) (pp. 27–47). Oxford, U.K.: Symposium Books.
  • Cox, P., Geisen, T., & Green, R. (Eds.). (2008). Qualitative research and social change. New York, NY: Palgrave Macmillan.
  • Crocker, J., & Major, B. (1989). Social stigma and self-esteem: The self-protective properties of stigma. Psychological Review, 96(4), 608–630.
  • David, E. J. R., & Derthick, A. O. (2016). What is internalized oppression, and so what? In E. J. R. David (Ed.), Internalized oppression: The psychology of marginalized groups (pp. 1–30). Berlin, Germany: Springer.
  • Douglasova, M. (2014). Cistota a nebezpeci: Analyza konceptu znecisteni a tabu. [Purity and danger: An analysis of concepts of pollutions and taboo]. Praha: Malvern.
  • Etter-Lewis, G., & Foster, M. (1996). Unrelated kin: Race and gender in women’s personal narratives. New York: Routledge.
  • Finney, N., & Rishbeth, C. (2006). Engaging with marginalised groups in public open space research: The potential of collaboration and combined methods. Planning Theory & Practice, 7(1), 27–46.
  • Fischer, O. (Ed.). (2008). Etika a lidska prava v socialni praci. [Ethics and human rights in social work]. Praha, Czech Republic: Centrum Socialnich Sluzeb.
  • Foucault, M., & Chomsky, N. (2005). Clovek, moc a spravedlnost. Praha, Czech Republic: INTU.
  • Frankl, V. E. (1997). Vule ke smyslu. Brno, Czech Republic: Cesta.
  • Freire, P. (1974). Education: The practice of freedom. Santiago, Chile: Institute for Agricultural Reform.
  • Girard, R. (1989). The scapegoat. Baltimore, MD: Johns Hopkins University Press.
  • Gkofa, P. (2017). Promoting social justice and enhancing educational success: Suggestions from twenty educationally successful Roma in Greece. Urban Review, 49, 443–463.
  • Glaser, G. B., & Strauss, L. A. (1969). The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine.
  • Gobbo, F. (2004). Cultural intersections: The life of a Roma cultural mediator. European Educational Research Journal, 3, 626–641.
  • Gobbo, F. (2017). Educational justice and occupational mobility: The role of civic commitment and imagination. Urban Review, 49, 420–443.
  • Gobbo, F. (2011). Ethnographic research in multicultural educational contexts as a contribution to intercultural dialogue. Policy Futures Education, 9, 35–42.
  • Gordon, P. S. (2005). Not a Mormon: Confessions of a dangerous nomo. Cultural Studies, 19, 423–429.
  • Guillemin, M., & Gillam, L. (2004). Ethics, reflexivity, and “ethically important moments” in research. Qualitative Inquiry, 10(2), 261–280.
  • Hammer, M. R., Bennett, M. J., & Wiseman, R. (2003). Measuring intercultural sensitivity: The intercultural development inventory. International Journal of Intercultural Relations, 27, 421–443.
  • Henry, A. (2015). We especially welcome applications from members of visible minority groups: Reflections on race, gender and life at three universities. Race Ethnicity and Education, 18(5), 589–610.
  • Hofstede, G. (2002). Allemaal Andersdenkenden. Amsterdam, The Netherlands: Uitgeverij Contact.
  • Howard, J. A. (2000). Social psychology of identities. Annual Review of Sociology, 26, 367–393.
  • Jung, C. G. (1998). Archetypy a nevědomí. Brno, Czech Republic: Nakladatelství Tomase Janecka.
  • Kondakci, Y. (2017). Differentiated function of school in socio-culturally disadvantaged context: A constructivist grounded theory study from Turkey. Urban Review, 49, 400–420.
  • Konopasek, Z. (Ed.). (2000). Our lives as database; Doing a sociology of ourselves: Czech social transitions in autobiographical research dialogues. Praha, Czech Republic: Nakladatelstvi Karolinum.
  • Leeman, Y., & Ledoux, G. (2003). Intercultural education in Dutch schools. Ontario Institute for Studies in Education of the University of Toronto. Curriculum Inquiry, 33, 4.
  • Levinska, M., & Doubek, D. (2017). Alice: Overcoming the border. Urban Review, 49, 478–498.
  • Levinson, M. (2017). When participants don’t wish to participate in participatory action research, and when others participate on their behalf: The representation of communities by real and faux participants. Urban Review, 49, 382–400.
  • Lewin, K. (1946). Action research and minority problems. Journal of Social Issues, 2(4), 34–46.
  • Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363–385.
  • Linkova, M., & Stockelova, T. (2012). Public accountability and politicization of science: The peculiar journey of Czech research assessment. Science and Public Policy, 39, 618–629.
  • Maslow, A. (1943). A theory of human motivation. Floyd, VA: Wiklers.
  • Milner, H. R. (2007). Race, culture, and researcher positionality: Working through dangers seen, unseen, and unforeseen. Educational Researcher, 36(7), 388–400.
  • Moree, D. (2008). How teachers cope with social and educational transformation: Struggling with multicultural education in the Czech classroom. Benešov, Czech Republic: Eman.
  • Moree, D. (2013). Učitelé na vlnách transformace; kultura školy před rokem 1989 a po něm. Praha, Czech Republic: Karolinum.
  • Moree, D. (2015). Zaklady interkulturniho souziti. Praha, Czech Republic: Portal.
  • Moree, D., Vavrova, T., & Felcmanova, A. (2017). Blue or red, why do categories attract? Urban Review, 49, 497–509.
  • Okihiro, G. (1984). The writing of ethnic history. In D. Dunaway & W. K. Baum (Eds.), Oral history and interdisciplinary anthology (pp. 199–214). Nashville, TN: American Association for State and Local History in cooperation with the Oral History Association.
  • Perls, F., Hefferline, R. F., & Goodmann, P. (2004). Gestalt terapie. Praha, Czech Republic: Triton.
  • Pyke, K. D. (2010). What is internalized racial oppression and why don't we study it? Acknowledging racism’s hidden injuries. Sociological Perspectives, 53(4), 551–572.
  • Rammelt, C. F. (2014). Participatory action research in marginalised communities: Safe drinking water in rural Bangladesh. Systemic Practice and Action Research, 27, 195–210.
  • Reinharz, S. (1992). Feminist methods in social science research. New York, NY: Oxford University Press.
  • Samovar, L. A., Porter, R. E., McDaniel, E. R., & Roy, C. S. (2013). Communication between cultures. Wadsworth, OH: Cengage Learning.
  • Santoz, B. (2015). Koreny a kridla; Strom divadla utlacovanych (pp. 35–38). In Antikompex: Divadlem ke změně. Praha, Czech Republic: Antikomplex.
  • Stockelová, T., & Grygar, J. (2008). Hrátky s čerty? Politické souvislosti a přesahy sociálně-vědního výzkumu. [Playing with devils? Political contexts and overlaps of the research in social sciences]. Biograf, 46, 43.
  • Trahar, S. (Ed.). (2009). Narrative research on learning: Comparative and international perspective. Oxford, U.K.: Symposium Books.
  • Trompenaars, F. (1989). Riding the waves of culture. Fairfield.
  • Van der Zee, K. I., & Van Oudenhoven, J. P. (2000). The multicultural personality questionnaire: A multidimensional instrument of multicultural effectiveness. European Journal of Personality, 14, 291–309.
  • Van der Zee, K. I., & Van Oudenhoven, J. P. (2001). The multicultural personality questionnaire: Reliability and validity of self- and other ratings of multicultural effectiveness. Journal of Research in Personality, 35, 278–288.
  • Vaňková, M. (2010). Výzkumník lapený: Úvahy k (a)symetrii vztahů mezi výzkumníkem a zkoumaným. Biograf, 46, 52–53.
  • Whitzmann, C., James, K., & Poweseu, I. (2013). Traveling together: Participatory research methods for disability inclusive road development in Papua New Guinea. Journal of Transport Geography, 26, 65–71.
  • Wrentschur, M. (2008). Forum theatre as a participatory tool for social research and development: A reflection on “Nobody is perfect”—a project with homeless people. In P. Cox, T. Geisen, & R. Green (Eds.), Qualitative research and social change (pp. 94–112). Basingstoke, U.K.: Palgrave Macmillan.