Focus Groups in Migration Research: A Forum for “Public Thinking”?

This chapter outlines how to use focus groups (FGs) in migration studies, considering this method a forum for “public thinking” and discussing controversial issues. Moreover, the use of FGs allows us to understand the process of creating consensus and dissent via interaction. The chapter is structured in five sections: the first one introduces what FGs are and why they are useful for migration research; the second focuses on how to build the groups and how to do comparative migration research with FGs; the third illustrates how to prepare and to facilitate group discussion, and how to ask questions and engage participants in collaborative migration research; the fourth introduces how to interpret discussions and how to analyse the everyday naturalization of nation, ethnicity and race; the final section discusses how to communicate FG results. Each section is devoted to a specific methodological issue and it includes at least one “box” with an example from European migration research.

means of participation and representation in mainstream society and politics" (see Zapata-Barrero and Yalaz, Chap. 1 Introduction), they need to put first some methodological options. Doing FGs in qualitative research 2 involves opting for low standardization (keeping in mind that the theme presented for discussion to the participants will be context-sensitive). In the research, it is also important to maintain low directivity (researchers restrict their power to define the contents of the discussion). Finally, regarding topic guidelines with questions for group discussion, a low level of structure (the outline of the topics to be discussed will be flexible and the questions asked to participants present low constraints) is preferable 3 While interviews excel at eliciting "private" accounts, FGs give researchers access to the narratives and arguments that participants present in group situations, whether these are peer groups or researcher-convened groups of strangers. The added value of FGs compared to individual interviews is that not only do they allow the researchers to listen in to accounts about everyday experiences, but researchers are also able to observe the interactional context in which these accounts are produced. If we follow conversation analysts' invitation not to listen naively to the voices of social actors while thinking one has "direct access" to their experience, it is very useful to observe what happens in the interaction and to study the actions through which the taken for granted world is reproduced.
FGs are used to understand the process of creating consensus and dissent via interaction (Frisina 2010). FGs allow researchers to investigate how hegemonic 4 representations are formed, how they are negotiated, and how they can change. Moreover, especially within collaborative or participatory research with marginalized social actors, FGs may offer a "safe space" for generating counter-hegemonic discourses.
2 FGs have most frequently been used within the context of quantitative studies for developing and refining research instruments. One of the most common uses of FGs is during the exploratory phase of a research project. In many cases, they are integrated in a multi-method design with other quantitative/qualitative methods. 3 FGs may have different levels of standardization, directivity and structure, like interviews (Bichi 2007). When the goal is answering a "how much…?" question, doing an extensive study, reaching a statistically representative sample, involving a high number of people and groups, it is preferable to choose high standardization/directivity/structure. However, in depth studies -which try to answer "how…?" and "what…?" questions, including a limited number of cases, people, groupsallow researchers to carefully study meaning-making and to understand diverse cultural frameworks of social actors. 4 According to Stuart Hall's reading of Antonio Gramsci (see Hall and Mellino 2007), hegemony is the combination of functional strategies in the maintenance of the status quo, thus, to the interest of those in power. Nevertheless, hegemony is the unstable because the construction of consensus is always imperfect and the common sense is inevitably shaken by tensions. So, by hegemonic representation I mean that one which reinforces the hierarchy of power (whether it is about gender or class or race, etc.). By counter-hegemonic representations, I am referring to discourses call into question the common sense and the social hierarchy that it produces. Focus groups have been appreciated for different reasons: • For generating high quality and interactive data, for offering the possibility of theoretical advances regarding the co-construction of meaning between people in specific social contexts, for addressing ethical concerns about power and the imposition of meaning (Wilkinson 1998) • For being contextual and less-hierarchical (reducing the influence of the researcher), for emphasizing the collective rather than the individual and fostering free expression of ideas, and for encouraging people to speak up (Madriz 2000, 838).

Why Are Focus Groups Useful for Migration Research?
If a crucial feature of qualitative migration research is to offer a critical assessment of social and political reality, FGs can be considered forums for "public thinking", where controversial issues (border/mobility, diversity/citizenship, integration/ participation and discrimination/racialization) are discussed. As we will see in this chapter, research with FGs does not include only migrants. FGs are very useful in studying the interactions of migrants (and their descendants) with the host society (i.e., public and private actors). FGs may provide a space for questioning taken for granted points of view and experiences, they may generate processes of "consciousness raising" and transform personal troubles into public issues (Mills 1959). Moreover, if all migration research is comparative (Bloemraad 2013), using FGs is a particularly suitable method for making comparisons (see next paragraph on how to construct groups).
Finally, if we are especially interested in European migration research, considering Europe not only as our empirical field, but also reflecting on its specific historical cultural legacy 5 , FGs can be a useful method for trying to decolonize research practice (Tuhiwai 2012) and not to reproduce (racial) oppressions in the way we approach our research subjects (Bonilla-Silva and Zuberi 2008). Particularly for migrants arriving in countries "polluted" by racial power relations with a long colonial history 6 , FGs can be a "safe space" for exchanging ideas and sharing experiences.
How to construct the groups? How to do comparative migration research with FG (Sect. 11.3)? How to prepare and to facilitate group discussion? How to ask questions and engage participants in collaborative migration research (Sect. 11.4)? How to interpret discussions? How to analyse the everyday naturalization of nation, ethnicity and race (Sect. 11.5)? How to communicate FG research results and to whom (Sect. 11.6)?
Each section is devoted to a specific methodological issue and shows its relevance for migration studies. Section 11.3 is on research design/sampling and on the place of FGs in comparative migration research. Section 11.4 is on preparing a creative questioning route for discussion and on facilitating FGs within collaborative/participatory migration research. Section 11.5 is on FG analysis and on the de-naturalization of nation, ethnicity and race. Section 11.6 is on the communication of FG results and on the importance of engaging civil society for a more "public" migration research.
Finally, each section includes at least one "box" with a European migration research example: Box 11.1, which focuses on the relations between citizens living in the French-speaking part of Belgium and asylum seekers, is a useful example for learning how to build groups and how to make territorial comparisons. Box 11.2, which highlights the relations between Muslims and non-Muslims in Italy after 9/11, is useful for understanding how to use the narratives of "stories to be completed" to facilitate group discussion. Box 11.3, which concentrates on the public self-representations of young people with and without a migrant background in the North East of Italy, is useful for understanding how to use photo-elicitation in FGs within participatory action research. Box 11.4, which centres on the discursive construction of national identity of Austrians, is useful for appreciating Critical Discourse Analysis approaches for interpreting FG discussions and learning the critique of the naturalization of nations. Box 11.5, which features young people with and without a migrant background and religious education in Italian public schools, is useful for understanding the importance of including the processes of agreement-disagreement and identification-differentiation among group participants in FG analysis. Box 11.6, which presents an example of a back-talk FG after an ethnography with young Muslims of Italy, is useful in showing the importance of discussing results with research participants and in considering the back-talk FG as a follow-up tool in migration research. Finally, Box 11.7, which highlights reproductive health and access to local welfare for migrant women in Padua, Italy, is useful for reflecting on scientific knowledge for whom?/for what? and for learning how to use FGs with participatory video to communicate results to wider audiences and to engage civil society in a public discussion on the crisis of the Italian welfare system.

How to Build the Groups? How to Do Comparative Migration Research with FGs?
Sampling is the keystone of good qualitative research design. FG participants are selected 7 through a purposive sampling strategy, which aims at reflecting a diversity of cases within the population under study. The question of how many focus groups to hold 8 is determined by the comparisons that the researcher wishes to make, keeping in mind that comparative migration research entails the systematic analysis of a relatively small number of cases (Bloemraad 2013).
Since FGs have the group as the main unit of analysis, it makes sense to ensure that group members share at least one important characteristic and the classical composition for FG research is made up of "homogeneous" groups to facilitate comparison. For migration researchers, this means being self-reflexive and very careful about how the groups are built and how to match moderator/facilitator and group 9 without reproducing the processes of ethnicization/racialization of migrants and their children. The important characteristic shared by the group (and the moderator/facilitator) is not necessarily linked to migrants' national belonging or ethnicity. With FGs European comparative migration researchers can learn from the limits of the North American tradition of comparative migration studies, which very often contrast different migrant groups in the same geographical location, assuming that national origin, ethnicity, and race fundamentally matters (Bloemraad 2013). Rather, according to Amelina and Faist (2012), migration scholars are invited to avoid "naturalizing views" of ethnicity and nation and thus not select ethnicity or nation as dominant categories relevant for setting up the research organization. "Methodological transnationalism" (Ibidem) encourages us to step out of cultural traits based on belonging to one nation-state/ethnicity/race/religion. Rather, the shared characteristic for building groups could be gender or class/position in labor market, or patterns of mobility, condition of exit, legal status, age at migration, or length of stay, depending on the research question. 7 Recruitment strategies can take two routes. They can either be "top down", using lists of names provided by local organizations or by resorting to public announcements in newspapers and social media, or "bottom up", through informal social networks, gatekeepers or direct knowledge with some preliminary fieldwork. In either case, the motivation of the participants remains key to generating interesting data. 8 Moving in the path traced by the theorists of Grounded Theory (GT), one cannot answer the question on the number of focus groups needed a priori. 9 As Encarnacion Gutierrez-Rodriguez pointed out (Gutierrez-Rodriguez 2010), in her focus group experience with domestic workers from Latin America, she told them about her family background -her mother was a domestic worker from Latin America in Germany -but she realized the fragility of an assumed commonality. The Black feminist perspective of intersectionality (Crenshaw 1989) helps us to recognize the complexity of "matching a facilitator/moderator and a group" by acknowledging the multiple positionalities emerging during the research process.
Another classical composition of FGs is made up of groups of people who do not know each other in order to facilitate self-disclosure and reduce the risk of ethical complications related to the right to privacy of each participant. In fact, the facilitation of groups made up of existing groups involves some difficulties for the researchers (i.e., avoiding allusions in FGs which after are difficult to analyze), but working with "natural groups" can be very useful to respond adequately to specific research questions. Thus, variations and adaptations are always possible and there are many ways of making comparisons in migration studies. However, contrasting groups is always crucial, because by making comparisons between cases "we can decenter what is taken for granted in a particular time or place" (Bloemraad 2013: 29).
The next example (Box 11.1) shows a migration research: • Which is based on "territorial comparisons" • Which constructed "heterogeneous groups" • Where people sometimes knew each other/met outside the research, because they lived in the same local context. 10 In Wallonia and in Brussels. 11 The interviews serve to explore the personal views of different actors involved in local and federal policies on asylum. The focus groups allow us to understand the influence of the local environment. 12 A reflective summary of this innovative research helps us to grasp what may be difficulties of this type of sampling design. The asylum seekers struggled to get their voices heard in the group interactions and their criticism of the management of reception could hardly be expressed in a semipublic context like focus groups, where they were few in number. Often, there were two asylum seekers, in some cases one, out of between 7 and 9 people.

Box 11.1: The Welcoming of Asylum Seekers in Belgium (Gsir et al. 2004)
The research question is about social relations between citizens living in the French-speaking part of Belgium 10 and asylum seekers. The investigation explores the ways in which everyday interactions are built and pays attention to representations of asylum seekers. The study uses in-depth interviews and FGs 11 , it is comparative and six locations includes in the sample: municipalities with reception centres for asylum seekers (Fraipont, Brussels/Petit-Château, Rixensart) and places where there are no reception centres (Sainte-Marguerite headquarters in Liège, Bockstael quarter of Brussels, Ottignies). The purpose is to analyse the influence of the presence or absence of the centres on the representations of applicants for asylum, including the daily interactions between the inhabitants of these places and asylum seekers. The use of focus groups is motivated by the desire to investigate how the representations of asylum seekers are formed and how they change. The FGs are used to observe interactions between institutional actors (mayors, policemen, directors of shelters ...), associations (non-governmental organizations, sports associations, cultural ...), ordinary citizens and asylum seekers 12 . (continued)

How to Prepare and to Facilitate a Group Discussion? How to Ask Questions in Collaborative Migration Research?
The researchers have to prepare topic guidelines with care and have to reflect seriously on which communicative moves are the most appropriate to support the interactions between participants and to allow dissent to emerge. So how should they prepare the discussion questions?
The secret to concise guidance, through which we can focus on the main topics to be discussed, lies in their construction. First, the keywords which summarize the main points to be discussed have to be identified. To avoid dispersion, it is fundamental to limit the number of keywords. If the issues are particularly complex, the option to hold discussion groups "in stages" in which connected issues are addressed with the same group in several meetings (usually two or three) should be kept in mind. Second, it is useful to rank the points in order of importance and then study the chronology of their presentation. A consolidated strategy is funnelling: the most important topics should be placed in the centre of the discussion session, then tentatively addressed "mid-meeting" (which will normally last an hour and a half 13 ), For the choice of sites with reception centres, the researchers created the following classification: organized opposition (places where there is opposition from municipal authorities, and/or segments of the population through demonstrations and petitions, etc.); open (locations where no protests occurred and where there have been public demonstrations of solidarity, in the form of gifts and other voluntary initiatives); indifferent (no clear public stance, either for or against the reception centre).
In each of the six identified local contexts (with or without a shelter), two 2-h FGs were carried out a week apart. The first identified interactions, positive and negative, towards asylum seekers, and the second, going back to the topics discussed in the first focus group, elicited possible solutions to problems regarding the reception of asylum seekers. Each time a larger number of participants than expected were invited, allowing for no-shows (which always occurred for various reasons).
The research showed that social relations are shaped by the presence/ absence of reception centres for asylum seekers. Where there were centres, the expression of suspicion towards asylum seekers became more palpable and the inhabitants felt "downgraded", afraid that their homes would lose value. In addition, it was found that factors that promote positive interaction between asylum seekers and the local population included the existence of a voluntary sector, the good will of local political authorities, good management of the centres, the presence of children of asylum seekers in public schools.
Box 11.1 (continued) after the participants have become acquainted with each other and they begin to explore the topic. Preparing an outline of questions allows us to ponder how we should address the participants, so as not to impose our ideas or use inappropriate terms; to give rhythm to the group work and to use their time profitably.
The fundamental task of researchers is not to monitor the group, but to support it and guide it. The terms used for this task of discussion management are three: leader (maximum directivity), moderator (average directivity), and facilitator (minimum directivity). The preferred option is facilitation and, according to Putcha and Potter (2004), the practices to be implemented to facilitate group discussion are learned and are essentially three: to generate informality; to generate participation; and to generate a variety of viewpoints. Generating informality means keeping the space frame sufficiently large (Goffman 1981), in which they direct the focus group participants, so they are not obsessed with the "keeping of face" (that is, of their good reputation), especially if it comes to focused interaction among "outsiders". Even the physical space plays its part, so it is better, for example, that it looks like "a living room rather than an office" (Putcha and Potter 2004: 39). To promote the participation of the research subjects, it is important to create the group discursively, addressing and putting questions to a plural subject. To create a variety of opinions, it is fundamental to hook into the participants' body language, to recognize any emergent dissent which, before being said and made explicit, is often communicated more "softly", indirectly, through the body. Communicative activities are in fact multimodal: in addition to the verbal exchanges, it is necessary to observe/listen to interactions beyond the words, opening up, so to speak, all our sensory channels. In my opinion, the bodies of the participants should be considered as an active part in interactions and it is interesting to observe what they do, how they contribute to building and negotiating the meaning of group discussion. A second researcher is needed to take notes systematically, listening and observing the bodies in the group discussion. The observer can help the facilitator to note the hinted at dissension and to recognize the conflict in the discussion, which is often expressed obliquely through body language. It is said that silence is consent, but silences in the group discussion should not be trivialized and silent-dissents must be recognized and explained. If we consider culture as a battlefield (Hall and Mellino 2007), in which different meanings and different versions of the world are competing, then through focus groups we can study that consent is never given once and for all, and the common sense that daily life is made of is challenged by subjectivity that arises actively against what is given.
Let us now look at two research examples on how to do creative questioning (using stories or images as questions) and engage participants in a more collaborative research process. Box 11.2 shows how to use the narratives of "stories to be completed" to facilitate group discussion, and Box 11.3 shows how to use photoelicitation in FGs within participatory action research (PAR) 14 .

Box 11.2: The Relationship Between Muslims and Italian Society After 9/11 (Allievi 2009)
The research question concerned the relationship between Muslims and non-Muslims and the impact of Islamophobia after 9/11. The larger study included focus groups in the UK, France, the Netherlands, Belgium and Germany. The Italian research team was composed of Stefano Allievi, Annalisa Frisina and Luca Trappolin. We created nine FGs in three different cities (Rome, Padua and Milan). In each group, a theme was developed though three meetings -in an arc of 15 days -in the same discussion groups. In the case of Milan, for example, the theme was "places and actors of social transformation" (Frisina 2009). The profiles of the participants included actors who had personally lived the changes associated with migration (a daughter of Jordanian-Palestinian immigrants born in Italy; an immigrant parent of the Muslim faith; a partner of an Italian-Somali couple; an activist in community centres from Ecuador…) and actors who work in places of marked cultural and religious diversity (teachers in secondary schools with a significant number of students of foreign origin; a director of an Afro-Italian theatre company; cultural mediators in health care and education; trade union activists…).

Exercises of daily imagination
In the discussion, we introduced some "exercises of imagination to think about cases that had not occurred, but could have occurred" (this was my way -as facilitator -to invite participants to discuss).

Exercise 1: "Youth and Sport"
In an amateur volleyball team, there is a 15-year-old Italo-Egyptian girl; she is the best player and the captain of the team. The period of the tournaments comes round, but her parents do not want her to travel with the team and play in public. In your opinion, what will the girl do? What do you think the others will do, that is, her parents, coach, teammates, the parents of her teammates?
With the first stories finished, some elements proposed by participants 15 were used to go on with the discussion. empowerment by calling attention to racism, discrimination and other social structures responsible for the conditions they face. PAR with migrants (Rodriguez 2013) explores the causes that make migrants vulnerable and restores the legitimacy of immigrants as civic actors with the right to demand better conditions. 15 In particular, an Italo-Palestinian practicing Muslim girl was taken as the starting point and how the clothing issue had been resolved in her case was narrated -through the flexibility of her volleyball coach -who did not require her to wear tight short shorts -and the solidarity of her teammates, who were playing away when they put on white bandannas. The girl indeed habitually wore a hijab, the Islamic headscarf, but in volleyball matches wearing this new -more inclusive -uniform of her team of white bandanna and baggy knee-length shorts. (continued)

Box 11.3: FGs in PAR and the Use of Photo-Elicitation: Unpacking Photovoice (Frisina 2011a)
Recently, the boundaries between focus groups and workshops in PAR have faded and those methods overlap when the making of data arises from the usage of photo/video-elicitation, or art-based/activity-based processes (Carretta and Vacchelli 2015). PAR, and with it the techniques that it employs as workshops, have been criticized for being excessively empiricist and lacking theorization. The process of data generation and the power dynamics existing between researchers and participants need to be more (continued)

Return to Exercise 1
Imagine that, despite her teammates being very supportive of the Italo-Egyptian girl, their parents enter forcefully into the discussion and rebel against the common front of the girls. Some parents say things like "we are not Taliban, we cannot dress our girls like them!" What can be done? How can these conflicts be prevented or managed?
From the FG several discourses emerged that represented a part of Italian society in crisis, because of its growing cultural and religious diversity. Therefore, it seemed appropriate to present the following narrative to the group:

We are in a primary school on the outskirts of Milan, and a second grade teacher, who has two Arab children in her class, raises the problem of how to celebrate Christmas. She decides, this year, not to make nativity scenes, despite the fact that in the school there is a tradition of a contest in which a prize is given to the most beautiful nativity scene. This decision is explained to children as a sign of respect for their companions of other religions and the kids talk about it with their parents. What do you think will happen? We are not asking you to say what you would do, but imagine what would happen, based on your own experience.
In this second case as well, we took the elements proposed in the daily stories of the participants to propose this follow up.

Imagine the case in which different groups ask for a space to celebrate their religious festivals: what would happen? In our system, may schools decide the days off? What do you think would happen if the school calendars were changed for intercultural reasons?
Box 11.2 (continued) problematized 16 . When I investigated how groups of young people with and without a migrant background 17 construct visual self-representations to present in the public sphere in Veneto 18 , I preferred to replace workshops with focus groups and to unpack photovoice 19 into three different research methods, so as to facilitate the complex process of analysing the data generated and to promote greater reflexivity.
First, I started with image-making, calling research participants to make pictures on three themes ("Self-portraits/Portraits of a new generation"; "People/places in the town I live in that make me feel (in)secure"; "Being/ feeling a citizen, being/feeling a foreigner") and to write a brief comment on each photo.
Second, I used photo-elicitation within serial FGs. Photo-elicitation helped reduce the directivity of the researcher, replacing verbal questions with "visual questions" (the images produced by participants were presented to the group without their corresponding verbal comments). FGs allowed me to investigate the negotiation of meaning of images and to grasp how seeing differently emerges contextually. Reconstructing the photovoice process through repeated FGs 20 was useful in developing a more reflexive research practice: how should the researcher interact with participants, what role should (s)he play in the group? Wang's (2006) (2015), on a Minga biographic workshop with highly skilled migrant women. She proposed to acknowledge both commonality and difference in the process of negotiating research partnerships and she argued for a more complex understanding of privilege and power. 17 I adopted a daily multiculturalism perspective to study young people (Harris 2009) to overcome the idea that diversity is something (problematic) which (white, autochthonous) adults must manage and challenge the assimilationist assumptions which see the nation as pre-existent, a well-defined reality in which these young people should "integrate". It is rather a question of bringing to light the daily micro-practices of producing differences and contesting dominant representations of "us". 18 As I am interested in the use of diversity as a political resource -to legitimate relations of power and exploitation, or to carry out social critique and demand change -I believed that Veneto, with its strong presence of the Northern League political party and high percentage of children of migrants in public schools, was an appropriate empirical context. 19 It is a visual PAR which promises to enable people to define for themselves and others, including policy makers, what is worth remembering and what needs to be changed Wang (1999). 20 I developed six FG sessions (for each of the five groups of young people) and each lasted about 3 h. (continued)

How to Interpret Discussions? How to Analyse the Everyday Naturalization of Nation, Ethnicity and Race?
The researchers' memories and notes are not enough; each focus group should be recorded 21 and transcribed 22 in order to produce a systematic coding of the empirical documentation produced 23 . Making sense of focus group data through developing a coding framework is a complex process, which involves generating a provisional 21 If we have taken care of our social relationships with participants by building trust and negotiating informed consent in the early stages of the research, the presence of a video camera will be accepted during the discussions. 22 The perfect transcription does not exist, because it's still a selective process, even when, hopefully, it is reported back word for word of both the participants and the facilitator (it is "verbatim"). 23 The corpus to be analysed is formed substantially by group discussion transcripts, but also includes the notes of the observer on body language taken during the focus groups; exchanges in the debriefing between research subjects and researchers (feedback on the context of the focus groups) and between the facilitator and observer (feedback on facilitation/on the data generation process); furthermore, if there are any, any communication that took place with the participants outside of the focus groups (for example, in e-mails) and any memos (notes of researchers in all phases of work in the field). the discursive construction of an "us" (militant) -which questions the structural causes of daily inequality/discrimination to become subject of changecan be traced to Freire's pedagogy of liberation, which the researcher recognizes as her source of inspiration. I preferred simply asking "What do you see?" for each photo and I performed my role as facilitator getting participants to share their different points of view. Third, I did participant observation during the cultural events when (a selection of) pictures and narratives made by participants went public (to understand whether and how photovoice empowered the participants).
Starting from this research experience, I argue that FGs with photo-elicitation within PAR can become reflexive forums on sensitive issues of public interest, where citizens with and without a migrant background feel the tensions in the dominant representations of "Us" and "Them", reflect collectively on how Otherness can be negotiated (Frisina 2014).

Box 11.3 (continued)
coding frame 24 , revising it, modelling coding frameworks, arranging codes in a hierarchical order and showing the links between subcategories. To do this, there are several software programs 25 which can be used as a complement, not a substitute for researchers' skills. Rigour is achieved through a systematic and iterative process, whereby coding categories are continuously subjected to review in light of disconfirming examples to identified concepts and patterns. It is important not to rely on the topic guide to furnish coding categories and to include "in-vivo" codes (more descriptive, which use participants' words) as well as "a-priori" codes (more interpretative, rooted in our theories). But what constitutes the encoding process? As Boeije says (Boeije 2010: 75-76), these are mainly two types of activities: • Segmenting the texts (in this case, primarily the transcripts of discussion groups), annotating these segments with codes, and then creating categories (and subcategories) • Reassembling the segments so that they acquire meaning from a certain theoretical perspective and make it possible to adequately understand the phenomenon studied.
One issue that remains controversial in the analysis of FGs is how to account for the interaction and not only perform a thematic analysis of "what" the participants discussed.
Since the 1990s, critical discourse analysis 26 (CDA) has been developed to investigate how social inequalities are reproduced or challenged through discourse. CDA can be very useful for interpreting FGs in migration research (see van Dijk, Chap. 13 in this book). This analytic approach has proved particularly suitable for studying the daily construction of nationalism, its exclusionary effects on certain social groups or the rise of populism in Europe and the discriminations against migrants.
Since a recurring theme is the construction of us-them, through discursive strategies that usually characterize the "we" positively and the "others" negatively, in analysing FGs we should ask: how are people named? To which category are they retraced? What qualities and characteristics are attributed to them? What rhetorical and argumentative tools are used to justify and legitimize the inclusion or exclusion of some "others"?
The following research examples show how CDA can be used for interpreting FG discussions and unmasking everyday naturalization of nations and racialization of migrants (Box 11.4) and how to include interaction in FG analysis, taking into consideration the processes of agreement-disagreement and identificationdifferentiation among group participants (Box 11.5). 24 Any section of text can be assigned as many codes as you think are appropriate. I suggest reading Kathy Charmaz on Constructivist Ground Theory (Charmaz 2006) to learn how to work on "emerging" categories. 25 www.atlasti.com, www.qsrinternational.com for NVivo, www.maxqda.com, www.transana.org 26 For a great introduction, see Wodak and Meyer (2016). (continued)

Box 11.5: The Making of Religious Pluralism from a New Generational Perspective (Frisina 2011b)
Italian society continues to be seen as homogeneous in religious terms and the teaching of the Catholic religion in state schools is a pillar of the historical and cultural heritage of the Italian population. However, profound changes have been underway, with migrant families settling in the country and their Italian-born offspring attending Italian state schools. How do they feel about religious education at school? How do they view the Italian model of secular-

Box 11.4: Discursive Construction of National Identity of the Austrians (Wodak et al. 2009)
The main sampling criterion was linked to political and territorial differences. So, six groups were constructed: the formally non-Austrian Viennese group; the Viennese group of a proletarian district (Simmering); the group from western Austria (Vorarlberg); the group from southern Austria (Carinthia); the Eastern group (Burgenland) and the rural Austrian group (Styria). From the focus groups, two main discursive strategies emerged, influenced by the political leanings of the group and the different regional contexts in which the meetings were held: • Conservative strategies of the construction of national identity: assume and emphasize the historical continuity with the nation's founding fathers; proceed to black and white pictures, which prefigure disastrous scenarios if citizens become "accomplices" to the change; pride as a result of a rhetorical defence of "our superior values to be preserved"; • Transformative strategies in the construction of national identity: emphasize the differences between "then and now" and between "now and tomorrow"; using the rhetoric of "changed circumstances", "the history teacher who teaches us ..."; pride as a product of the discourses of what Austria can become (an "example" for Eastern Europe or for the whole of Europe).
An interesting aspect that emerged from the analysis is the group telling of everyday stories (beginning with anecdotes about things that happened to other people) as a way to revive prejudices towards migrants and citizens of foreign origin. Instead of the usual disclaimer "I'm not a racist, but ... (racist speech)", a story with a moral racist was used (in a nutshell: "they" are inferior and therefore exploitation is justifiable), putting the tellers at a distance for what only was "reported" without any direct implication of responsibility.
ism and religious pluralism in Italy? What do they see as Italianness? I organized and facilitated seven focus-groups 27 with young people with and without a migrant background in secondary schools in a northern Italian town 28 . From the discussions, students' demands for change from a generational point of view emerged (seeing beyond education into religion, creating new horizons for religious pluralism in Italy). In the analysis of the focus group transcripts, I considered how the contents of the discussion were related to the way the participants interacted with one another. The discussion groups reproduced the "Catholic norm" (the assumption that Catholicism is the glue of Italians, a "standard" considered part of a "we"), but also highlighted the tensions in the dominant representation of Italianness based on a common religion. I reproduce below one coded sequence. In group discussions, as in everyday life, social actors use different forms of socially shared knowledge and bring it into play in the conversation. But the focus group is a particular form of dialogue, it is an institutionalized space for dissent and for change (the facilitator introduces the discussion encouraging divergent views to be expressed and supports the discussion by asking for examples and clarification for oblique dissent, stimulating a more direct comparison). The focus group, therefore, appears to be not only a valuable method for investigating how the social order is maintained throughout, but also to study the cracks, tension, ambivalence created by the discursive practices of daily resistance against various sources of normativity. In migration research, this method can be useful to understand how multiple belongings (ethnic, national or religious) are constantly negotiated and to explore the daily confrontation between the nationalistic binary logic "either/or" vs the "both/and" transnational logic of multiple memberships (Amelina and Faist 2012, p. 7).

How to Communicate FG Results and to Whom? Public Sociology and Migration Studies
A distinctive feature of qualitative research is the construction of texts presenting the results. In these texts, the perspective of those who have done the research is supported by the voices of the participants. In the case of FGs, there will be extracts of discussions, so it is important not to make a list of points of view (as if they were separate individuals, like in interviews). Researchers should report at least three exchanges from the discussion (as was done, for example, in the extracts cited in Box 11.4) to be able to reconstruct the interactive context of the group. Following the proposal of a more reflexive sociology (Melucci 1998, pp. 22-31), it is desirable to practice writing up the results in various ways for different audiences, thus, not merely writing for the scientific community.
As qualitative migration researchers (De Tona et al. 2010, pp. 3-4), firstly, we are called to be reflexive (positioning ourselves in the research process and being responsible for the power imbalance in the relation researcher-researched) and to recognize the reflexivity of research participants (they are also able to reflect on and question the research process). Secondly, to respond to ethical and political challenges concerning contemporary migration, we need to be attentive to open dialogues with civil society.
If it is true that the neoliberal turn of the European Union "produces more and more deaths among the migrants who attempt to arrive, less naturalized and more precariousness for regular migrants" and that the fears of the "other" are constantly stoked by a media-political racism and increases the victimization of migrants (Palidda 2015, p. 97), it becomes urgent to reflect on scientific knowledge "for whom?". Box 11.6: "Back Talk Focus Groups" as a "Follow Up Tool" in Migration Studies (Frisina 2006) After 3 years of participant observation in the organization Young Muslims of Italy 29 , 50 non-directive interviews with children of immigrants who were born or raised in Italy and three FGs among young Muslims (both militant and not), I decided to present the main results of my research: identification with Islam emerges through interaction and is only one of many forms of identification for these youths. In their everyday life, these youths resist dominant securitarian and islamophobic frameworks through a range of tactics and strategies. Some of these youth used the current discourses on Islam as a resource for participation, sometimes with the risk of transforming diversity into a "profession". I drew together the most motivated and active of the research participants to question my interpretations in a FG. It was an opportunity to include them in the reflexive knowledge making, which cannot be an individual job, but is a dialogic practice to be exercised in a phase of "meta-research" (research on research). I distinguished three different types of opinions that emerged from the FGs: dissent, agreement, and suggestions. Drawing on this experience, I think Back-talk FGs can be useful for three main reasons: • They stimulate the reflection of the researchers by allowing them to generate new data • They empower participants by giving them a greater role in the research process • They ensure responsible dissemination of potentially sensitive issues to a potentially diverse and highly politicized audience. 29 For more see, www.giovanimusulmani.it e https://www.facebook.com/ GiovaniMusulmanidItaliaGMI/ Michael Burawoy (2005) called for a "public sociology", defending civil society against the domination of market and state and having civil society actors as its first audience. I think that this could also mean restoring the legitimacy of migrants (and their children) as civic actors and creating "safe spaces" to listen to their right to demand better life conditions.
In the concluding part of this chapter, therefore, two examples on how to involve civil society actors more extensively will be described to show: • The importance of discussing results with research participants and to consider back-talk FGs as a follow-up tool in migration research (Box 11.6) • How to use FGs and participatory video in order to communicate results to wider audiences (Box 11.7).
This chapter introduced what FGs are, illustrating how to build groups, how to prepare, facilitate and then interpret discussions, and finally how to communicate research results. It offered arguments in favour of using FGs in migration studies, because they offer a forum for "public thinking" and discussing controversial issues. This approach fits well with comparative migration research, it engages participants in collaborative research, it helps to question the everyday naturalization of nation, ethnicity and race. Finally, the chapter focused on Europe, not only as an empirical field (it included several European migration research examples), but also as a specific cultural legacy (i.e. considering how to decolonize research practice and reflecting on scientific knowledge "for whom").

Box 11.7: "As Human Beings and Citizens": Discussing Research Findings Through a Video (Frisina 2017)
The research was part of the European project Welfare and values. Migration, gender and religions and the Italian case study focused on reproductive health and on access to local welfare for migrant women in Padua. A video was produced to communicate research results to a wider audience. We worked together with intercultural mediators and welfare workers to organize FGs with migrant women from Nigeria and Romania. As the title of the video "As Human Beings and Citizens" suggests, it is a double invitation from migrant women. That is asking welfare workers to treat them fairly "as human beings", but also calling on other women in similar social conditions to act as people with rights -"as citizens". They refuse to be viewed through the dominant and disempowering charity framework as above all needy and destitute people.
We organized four FG sessions. In the first, we discussed how to use the camera and we preferred to opt for a participatory video (Milne et al. 2012). The goal was to listen to women talk about their everyday experiences, and thus generate reflexive knowledge during the process of communication, rather than focus on the final product (the video). We visited a Romanian association and a Nigerian beauty shop to recruit women, both regular and irregular migrants, to take part in the FGs and video. The three sessions, which were filmed, consisted of a discussion of the Padua case study research findings that the participants considered the most relevant to themselves and their experiences. The video includes a number of discussions on the limits of universalism and the gender conservatism of the local welfare system in Padua, as illustrated in everyday and personal stories about reproductive health.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.