Abstract
Keywords
Introduction
While longitudinal studies are often designed as large-scale cohort and panel studies (Holland et al., 2006; McLeod & Thomson, 2012; Neale et al., 2012), the number of qualitative longitudinal studies is still limited. Henwood and Lang stated in their report to the Economic and Social Research Council (ESRC; UK) that qualitative longitudinal research (QLR) has many advantages over purely quantitative longitudinal studies. QLR is more sensitive to contextual issues and can “make it possible to investigate how people’s everyday attitudes and actions are embedded in patterns of socio-cultural change” (Henwood & Lang, 2003, as cited in Holland et al., 2006, p. 2). Holland et al. (2006) stress that, accordingly, more QLR or mixed-methods research with a qualitative direction should be conducted. While QLR findings are published, its methodological reflections are less frequently documented. These methodological reflections describe methods of structuring data and analysis (e.g. Coltart & Henwood, 2012; Henderson et al., 2012; Kraus, 2000; Kühn & Witzel, 2000). However, the complexities of setting up a longitudinal study and the ethical implications of conducting longitudinal research with young people are seldom reflected upon in methodological journals. The small number of papers published on these topics are mostly related to the
In this paper, we wish to reflect upon and share our experience with implementing the first qualitative wave of a large mixed-methods longitudinal study that investigates the lives of young people in Vienna. After providing an overview of the topics and also describing the aim of the study and its complex research design in “The Project” section, we illustrate in “Managing a Large and Diverse Research Team” section how we realized this design by managing a large and diverse research team including student researchers and stakeholders in the field. In particular, we describe the tools we used to coordinate the research and also the benefits and challenges of blending research and teaching. In “Methodological Rigor and Quality of the Data” section we address the learnings from a pilot study, our measures to ensure a good quality of data by describing our methodological training, our materials and instructions for data collection, as well as our efforts to establish self-reflexivity as an important component in the research process. Finally, in “Ethical Issues” section, we describe ethical challenges encountered during research with non-privileged young people in the context of school.
The Project
The
Methods Used for Gathering Data
In the first qualitative wave, several qualitative research methods were selected for collecting data. This data gathering took place at five different
School walks
Following the project presentations in October 2016 at each of the five schools, a participatory research method called the “school walk” (Wöhrer, 2018) was employed to start the research process. School walks are an adaption of the “tour” used with very young children in the mosaic approach by Clark (2001). The pupils were asked to form small groups and to show their school to the researchers. The instruction was very open and the pupils could choose the route according to their own priorities. During the walk we asked questions like: What is this room? What do you usually do here? What do you like best at school? We tried to listen to every pupil in the group and explicitly ask those who were less communicative. Furthermore, we encouraged narrations about spaces and activities in school as well as their plans for when they leave school. The researchers involved wrote memos about the school walks and the pupils’ narrations (see Annex 2 in Supplemental Material).
Starting our fieldwork with school walks gave priority to the pupils’ perspectives. It signaled to the young people that they were the experts concerning school and their living world, while the researchers were the ones who listened and learned. This underscored the objective to work against school logic, which conceives adults as experts and pupils as learners (Feichter, 2015; Wöhrer et al., 2017). Furthermore, the memos revealed interesting information about group dynamics between pupils, the topics that pupils are interested in, and their future plans that could be contrasted with narrations in the interviews.
Participant observations
In the following weeks, many researchers spent several mornings in a school and observed school lessons and morning breaks. The focus of these observations was on interaction between teachers and pupils, interaction between pupils, and group dynamics between pupils. The researchers wrote field notes about their participant observations and drew maps of the school classes.
Our methodological approach to participant observation is mapped out by Delamont (2012, pp. 206–207), who describes participant observation as a mixture of observing and talking to people in their natural context in order to grasp how the world looks like to them. Hence, participant observation can but must not imply real participation. In our case, the researchers were situated at the back of the class and their primary role was to observe. However, both pupils and teachers often addressed the researchers in individual conversations during breaks or free-work periods. For example, pupils asked what school the researchers went to or if it is difficult to study at university. Teachers explained their didactical methods or their pedagogical interventions to the researchers.
Participant observations helped to gain insights into interaction at school and conversations about the transitory period after
Interviews with pupils
From February to March 2017, 107 pupils were interviewed using qualitative interviews with a biographically-oriented narrative opening about their lives and experiences to date (Schütze, 1983). Subsequently, the respondents were free to take their time to fully narrate experiences and events from their own life without the interviewers interrupting the narration with questions of their own. By adopting this approach, we wanted to open up a space for respondents to unfold their narration according to their own relevance systems, telling us what they felt and experienced as important in their lives. After the main narrative, during which interviewers listened carefully and took notes on the subjects referred to, interviewers first asked internal narrative questions, which are questions aimed at further developing topics and accounts brought up by the respondents (Rosenthal, 2012, p. 52). Only then were external narrative questions (Rosenthal, 2012, p. 52) posed by the interviewers, addressing topics that interested us and had not yet been mentioned. They centered around the themes of family, childhood, friends, hobbies and leisure time, school, measures of vocational guidance, future plans (concerning the near future, anticipated family and job plans), and role models. This breadth of topical areas enables a variety of research questions to be addressed by secondary researchers.
Each interview was complemented with a network chart in which pupils visualized their relevant family and friends and a short socio-demographic questionnaire including information about the socio-economic status of the pupil’s family, living situation, parents’ education and profession, religion, etc.
As pupils are not allowed to be in a room with a single, unknown adult at school, each interview had to be conducted by two persons. This obstacle was turned into an advantage for the project by having each interviewer accompanied by an observer and, accordingly, this provided two different memos and perspectives for each interview. The audio files were transcribed and a standardized document with keywords was completed for each interview.
The 107 interviews with young people form the core of the longitudinal project as they cover the individual life stories of our respondents and their subjective perspectives on their lives. Also, these will be repeated once a year over the course of the project, allowing us to study the unfolding of young people’s lives over time (Neale, 2015).
Interviews with teachers
In the autumn of 2017 17 teachers at the respective schools were interviewed with a semi-structured interview guide about their work life, experiences at school, classroom dynamics and their ideas about the transition process experienced by young people after
Altogether, the first wave of the qualitative panel was comprised of 22 school walks, 36 participatory observations at school, 107 interviews with pupils aged 13–16, and 17 interviews with teachers (Flecker, Jesser, & Wöhrer, 2017).
Triangulation of Different Qualitative Methods
When following the definitions by Denzin (2017, pp. 301), who distinguishes between data triangulation, investigator triangulation, theoretical triangulation, and methodological triangulation, we considered the second and the fourth type of triangulation. In the participant observations and for the interviews with pupils, we always had two different persons observing the same situation, and accordingly, two perspectives on and field notes about each situation. Regarding the field notes, the researchers often agreed on different foci beforehand, e.g. one observer focusing on the interaction between pupils themselves and the other one on the interaction between teacher and pupils. The idea here was to obtain a broader picture of the observed situations. In the interview memos, interviewer and observer used very similar, but not identical templates: The interviewers were asked to write more about the topical contents of the interview and their point of view as interviewers, the observers more about the dynamic of the narrations (discontinuation, what was left out, etc.).
The methodological triangulation, i.e. the use of different data-gathering methods at the schools, was chosen for two reasons. Firstly, it enabled collecting more diverse data about the participants (Darbyshire et al., 2005): The gathered data contrast with each other and create different perspectives (Meetoo & Temple, 2003). We followed Flick’s (2018, p. 450) suggestions for a “strong program of triangulation,” i.e. we used triangulation as a source of “extra knowledge” (not to simply confirm one approach with the other) and we deliberately chose these four methods for combining different perspectives. We want to illustrate this with the following example: In the interviews a number of pupils told us about xenophobic or racist remarks or actions by teachers which they had experienced during their years at school. The teachers on the other hand mostly stressed how they want to support their pupils, but some of them used a lot of generalizations about “Turkish girls” or “Muslim boys” in the interviews. In the participant observations we experienced teachers making derogatory remarks about the names or attitudes of pupils with a migrant background. Thus, the different types of data helped us to obtain a more nuanced picture of how pupils with a migrant background are discriminated against in school settings. The data are in part contradictory, but help understand how remarks that were often meant either as helpful or at least as non-pejorative by teachers non-sensitized to racist discrimination were generalizations operating with derogatory characteristics and accordingly seen as racist remarks by the pupils.
Secondly, using this set of methods enabled the researchers to gain familiarity with the institution of
Managing a Large and Diverse Research Team
Organizational Structure of the Project
One of the project’s important features is the integration of master’s students into the research. The University of Vienna’s Master of Sociology program includes a course that is called “Forschungslabor” (which translates as “research lab”), which takes place over two semesters, comprising three teaching units each of 45 minutes every week, and is affiliated to one of the department’s research projects. The first wave of the qualitative panel was planned and conducted within the context of such a research lab attended by 17 sociology master’s students under the supervision of three course instructors, all members of the steering committee, and with the support of the two research assistants.
Organizing 22 school walks, 36 observations,107 interviews with pupils, and 17 interviews with teachers in five different schools was a challenging endeavor. In total, 73 people were involved in the research during the first qualitative wave: 14 members of the department, 22 master’s students as interviewers as well as 36 bachelor’s students and one external person as observers.
Using Online Tools for Coordination and Communication
The researchers in charge of the data-gathering process had to ensure that all researchers were in the right place at the right time with the right set of information and materials. Therefore, different online tools were used to assist with the data collection process: (1) Materials for school walks, observations, and interviews were made available on an e-learning platform hosted by the University. All researchers were registered in this information repository and could download the materials at any time. Moreover, the platform could be used for communication. It proved to be a versatile instrument for exchanging information and managing changes at short notice. (2) An online calendar was used to coordinate the interviews. The interviewers and observers could see which time slots were provided by the schools and insert their names. A second table within the online calendar was used as a tool for exchanging contact details. This allowed interviewers and observers to alert each other in case of unforeseeable changes on the day of the interview. An illustration of this calendar can be seen in Annex 3 in Supplemental Material.
Benefits and Challenges of Combining Research and Teaching
The small number of students in the research lab and the high student-teacher ratio proved beneficial to the project. Students could be truly involved in decision-making and solving problems as they emerged. This engendered a high level of commitment and motivation among these students, encouraging them to put their best effort into the project. Student researchers perceived themselves as part of the research team. Some of them volunteered to hold extra interviews or observations, which was remunerated by the project’s budget. Certain student researchers are currently using the project’s data as part of their master's theses.
Despite the many benefits of including student researchers, several limitations were inevitable. For example, the university course format was only suitable to a certain extent for implementing a large research project. The limited number of teaching units required much of the coordination and counseling to take place via e-mail or on teaching platforms. Even with three instructors, this project required a considerably higher amount of personal effort and time expenditure from the instructors than any other course on empirical research methods.
Furthermore, it was challenging to coordinate and synchronize the circular logic of qualitative research with the mostly linear logic of a university course. The project’s different structural and temporal levels added additional layers of complexity to this process. The complexity resulted in (1) different levels of decision-making, (2) different priorities, and (3) different timelines.
Different levels of decision-making
While overarching decisions regarding the research project were discussed and agreed upon by the project steering committee that met once a month, decisions concerning the data collection process often needed to be made ad hoc and were made either by the three researchers in charge of the research lab or by their students in tandem with the three researchers. After a few interviews were conducted and the experiences shared in class, we, for example, made the decision to include two questions concerning religion and religious practices in the socio-demographic questionnaire that complemented every qualitative interview. In order to remain flexible enough to implement changes as soon as possible, an intermediate decision body was created by the three instructors and two research assistants responsible for data collection.
Different priorities
Different priorities needed to be balanced within the project: Within a research logic, certain issues seemed particularly important, such as careful fieldwork, documentation of the data collection, or a circular process of data collection and analysis. However, these sometimes differed from the priorities within the logic of teaching, for example, where students should have more time to become acquainted with theoretical and methodological approaches.
Different timelines
The circular logic of qualitative research and the mostly linear logic of the university course produced different timelines. Accordingly, the research project timeline sometimes overruled the research lab timeline and produced loops or additional work for the students (e.g. as new discussions in the steering committee led to procedural changes). On the other hand, however, the teaching logic defined some temporal aspects for the research project. As the interviews took place at the schools, these institutions and their timetables strongly influenced the timelines of both the research project and the research lab. An example briefly illustrates the intersecting logics and timelines: The interviews—which were initially planned for January 2017—had to be postponed because preparations in the research lab took longer than anticipated (e.g. preparatory readings, interview training). Starting the interviews in February was actually better suited to the school logic, as it coincided with weeks of lower activity; however, this conflicted with the Austrian university schedule, where February is a month-long break. Accordingly, most interviews took place in March, which delayed the transcriptions and the beginning of analysis.
Methodological Rigor and Quality of the Data
Methodological rigor and quality of the collected data were primary concerns throughout the project since we were to start building a database for a longitudinal study, with the data being explored by many researchers in the future. Quality issues were also central because of the large and heterogenous team of researchers in the field who had different levels of training and experience with qualitative fieldwork in general and the applied methods in particular. On a methodological level, we therefore implemented several strategies for quality assurance: (1) We drew on the experiences in a pilot study conducted before starting with the qualitative longitudinal study, (2) we provided extensive methodological training for researchers, (3) we put together materials to assist researchers in the field, and (4) we reserved a large amount of time and space for interviewer reflexivity.
Learnings From the Pilot Study
In spring 2016, one year before the main study started, we conducted a pilot study (Pohn-Lauggas, 2016) in one of the schools that later participated in the main project. Students from a research course on narrative interviewing techniques at the department interviewed 21 pupils in one graduating class. The objective of the pilot study was to learn more about interviewing pupils aged 13–16 with a narrative-biographical approach, and about performing research with students as researchers.
The pilot study was extremely helpful on a methodological level. The interviews were substantive with regard to content and benefit the preparation of fieldwork. (1) The pupils’ narrations as displayed in the interview transcripts enabled us to obtain a feeling for the breadth of topics relevant to the young people and were used to phrase external narrative questions in the interview guide. (2) The interview memos from the pilot study revealed that information is documented differently depending on the student in question and sometimes very little written information is included in their documentation. As a response, we decided to standardize as much documentation as possible and elaborate reflective questions that facilitated the documentation process and enhanced methodological rigor of the data (see “Materials and Instructions” section). (3) During the interview training, transcripts and memos from the pilot study aided in sensitizing ourselves to interviewing as it became clear that we had to be prepared to face potentially difficult topics during the interviews and carefully handle intense emotional situations (see “Enhancing Researcher Self-Reflexivity” section). We also used examples from the pilot interviews for one of the interviewing exercises (see “Methodological Training for Researchers” section). Finally, (4) the interviewers’ experiences made us aware of ethical challenges regarding the conducting of research in schools as an institution shaped by hierarchical power relations (see “Finding a Position in the Field” section). This awareness was important for planning our own entry to the field and designing a suitable process for gaining informed consent (see “Protection of Privacy and Anonymity” section). It also resulted in the decision to triangulate different methods of data collection and to include methods that allowed us to gain more information on the context of school.
With regard to the student researchers, the pilot study encouraged our approach to combine research and teaching. Students taking part in the pilot study showed a high level of motivation and felt that they profited from taking part in an actual research project. Three of them used the data collected in the pilot study for their master’s theses (Chan, 2017; Kompek, in progress; Müller-Keplinger, 2018).
Methodological Training for Researchers
Prior to data collection, our efforts started with providing our researchers with comprehensive methodological training. The training consisted of practical exercises alongside theoretical input and was given in the course of the research lab. Nonetheless, all participating researchers were invited to join the sessions. Training was essential especially for student researchers, as students had different knowledge of and practical experience with the research methods being applied. They had all already collected some form of empirical data before, but none of them had conducted school walks and only a few had used observations and narrative interview techniques (Schütze, 1983). We addressed all data collection methods in the research lab, explaining methodological procedures and discussing relevant literature, but since in the overall project design the interviews were conceptualized as the core of the qualitative longitudinal study, being repeated over the course of 5 years, extensive preparations and practical training prior to data collection focused largely on interviewing. While school walks and participatory observations were conducted with little practical training before entering the field, they were both supervised closely during data collection, with one of the instructors being present in the respective school most of the time. After the school walks and observations, different settings for reflection, feedback and exchange—within the peer group of fellow student researchers and together with the instructors—enabled the researchers to sensitize themselves to methodological procedures and become more aware of the dynamics in the field (see “Materials and Instructions” section).
Regarding interview training, we started with a recapitulation of the methodological procedures of narrative interviewing (Rosenthal, 2012; Schütze, 1983). Discussing the work of Rosenthal and Köttig (2006), who raise the question of whether socially disadvantaged young people are capable of telling their life stories, we explored our own reservations and sensitized ourselves to the situation of our respondents. We also familiarized ourselves with strategies for stimulating the pupils to elaborate narrations of their own and to overcome challenges that might appear during the interview. We discussed, for example, how active and empathetic listening and the verbalization of observed emotional contents of narrated experience (Finke, 2000, p. 753) demonstrate the interviewer’s willingness to engage with sensitive or difficult topics and can encourage respondents to talk about their experiences in more depth.
Practical exercises followed the literature review and reflections. Ideas for interviewing exercises were inspired by suggestions from Helfferich (2011) and by methods for narrative interview training introduced by Rosenthal in the 1980s, which are now imparted by her students at the Center of Methods in Social Sciences at the University in Göttingen. In teams of three, every student had the possibility of starting a biographically-oriented narrative interview with a fellow student; the third student observed. The students were encouraged to reflect on their experiences in the different roles and to consider how the way in which questions were asked produced different kinds of accounts, and how interview settings as well as verbal and non-verbal reactions of the interviewer influenced the course of narrations (Helfferich, 2011, pp. 48–51). Two other exercises focused explicitly on posing internal narrative questions. In one exercise we used quotations from interviews in the pilot study and discussed possible ways for the interviewer to elaborate on what was being presented. This exercise illustrated the breadth of possible directions for follow-up questions and was therefore helpful in reducing worries about following “the right” leads. In the other exercise, students interviewed each other in turn about recent travel experiences, in front of the whole class. After a short narration, the instructor interrupted and the interviewing student was asked to pose a first narrative follow-up question, which was further developing one of the issues addressed so far. After a short narration, the instructor interrupted again and the student had to pose another narrative question, going into still greater detail, ideally producing a narration in which the respondent immersed themselves in the memory of a situation or an event and gave a vivid narration using direct speech. This exercise was meant to train students in posing narrative questions and in encouraging interview partners to elaborate on themes and topics that were covered only superficially in the initial narration. Conducting the exercise in front of the whole class was time-intensive, but proved valuable since students could learn from one another and get a better feeling for and familiarity with possible ways of carefully picking up what was told and reflect it back to the respondent in a question. It turned out that students had reservations about addressing a topic that was already mentioned in the interview, feeling rude to ask again or to repeat the phrase “can you tell me more about…” over and over again. Learning that the interviewed students experienced the situation as encouraging and the interviewer as appreciative and interested, contributed a great deal to reassuring the students in their role as interviewers.
In addition to the training, we held two meetings with staff members that provided instructions for carrying out the interviews and the opportunity to ask questions about the data-gathering process.
Materials and Instructions
In the phase of data collection, practical instructions assisted researchers every step of the way. Every field contact was accompanied by a description of how to do it. The interview guidelines, for example, contained detailed instructions on how to introduce the project to ensure that the opening of the interviews would not differ greatly, and also contained suggestions on how to word internal and external narrative questions during the interview to prevent closed questions and facilitate maximum openness to narrations from the pupils. Furthermore, a checklist with the necessary interview materials was created to prevent poorly-conducted interviews arising from missing documents or equipment.
Besides the collection of data, the documentation of data collection was also accompanied by instructions and templates. In order to enhance the quality of documentation and thus the analytical value of the data, guidelines and templates were developed for all written documentation. Templates for interview memos, for example, included a series of questions to be addressed in the course of the memo. We included reflective questions, concerning expectations and emotions regarding the interview and the interview partner, emotions, thoughts and impressions during and after the interview, etc. An extra topical area was reserved solely for reflection, focusing on the processing of the interview and how the interviewer experienced their role as an interviewer (see Annex 1 in Supplemental Material for a translation of the whole template). Similarly, other templates were created for memos of school walks (see Annex 2 in Supplemental Material), field notes, and consent forms.
Students found both the assisting materials for data collection as well as the templates for data documentation to be helpful. The former gave reassurance before and during data collection as they provided something to “hold on to”. The latter stimulated an inner dialog regarding their experiences during fieldwork and made them aware of aspects they had not yet considered in their reflection. Only the interview guide became rather lengthy because of instructions and concrete phrasings of external narrative questions and some students mentioned that they felt worried about not being able to address all the relevant topics during the interview. As we were aware that it is difficult to phrase narrative questions spontaneously, we decided to use the guide despite its length. However, reflecting on these feelings in class enabled us to put most worries at ease by offering to arrange a second meeting to continue and finish an interview if one meeting was not enough.
Enhancing Researcher Self-Reflexivity
Throughout the project, researcher self-reflexivity was of utmost importance to the core research team. It soon became clear that self-reflexivity was a key measure for addressing both methodological and ethical issues, especially (but not exclusively) when involving young participants who are in some cases marginalized. Thus, the researchers aimed to integrate and partly formalize measures promoting interviewer self-reflexivity within the team at different stages of the research process (Pezalla et al., 2012).
Open reflection
Within the research lab, we allocated the first 15–30 minutes of every teaching unit to open reflection on issues relating to the ongoing data collection process (we organized separate meetings for staff interviewers). Students were free to talk about everything they experienced in the field. Sometimes it was necessary for the instructors to comment on these experiences and define criteria on how to deal with situations (e.g. when respondents talked about minor criminal acts they had committed, we gave clear instructions to keep these accounts confidential and not report them to the authorities). At other times, a student’s account provoked the disclosure of similar experiences by other students, and this facilitated mutual sensitization to methodological and ethical difficulties of the research (e.g. when respondents would narrate very little and need a lot of encouragement and reassurance from the interviewer).
Guided reflection
Furthermore, every step of the data collection process was reflected upon separately in different groups and formats. In general, the instructors stimulated a reflection process by posing questions. Reflection phases before data collection centered around researchers’ expectations and preconceptions, while reflections accompanying fieldwork focused on interview experiences and difficulties arising during and after fieldwork, feelings and emotions evoked by being in the field, potential ways to address difficult interview situations, the handling of research instruments, and the discussion of ethical questions. Most of the time, students reflected in small groups first, followed by a discussion with the whole class. This was important, since frequently the discussions resulted in a refinement of research instruments, an adaption of the interviewing process, or a revision of templates for documentation and anonymization. In one such session we reflected upon the students’ field notes. It turned out that even though there were field notes by four different people of lessons with one and the same teacher, none of the researchers had documented that this teacher was wearing a headscarf. The instructors were surprised that nobody had noted this important description. In the discussion, the students addressed their insecurity about this issue. They felt that it was racist to document the headscarf as, for them, it was a symbol that should mean no difference at all. Eventually, we agreed to document everything that can be seen—and, if relevant, add a note of self-reflection about feeling insecure or discriminatory.
Feedback on written documentation
Another measure aiming at researcher reflexivity, among other things, was feedback on all kinds of written documentation (e.g. field notes) given by the course instructors (in written form, in front of the class, and in supervision meetings with only three or four students at a time), by fellow students in class and, on several occasions, also by the steering committee who e.g. commented on the interview guidelines.
Reflection meeting with professional supervisor
Finally, we organized a reflection meeting held for the student researchers with the support of a professional supervisor and psychotherapist for children and young people. This supervisor was external to the project as well as to the different schools. The session was aimed specifically at reflecting upon and helping students process difficult interview situations. These generally occurred when delicate and serious topics emerged during the interviews. As some of the interviewed pupils experience(d) grave life situations—such as violence in their family, forced migration, or drug abuse—these were presented in some of the narrations. Especially the less-experienced researchers profited from learning about strategies to better understand their emergent feelings of sadness, anger, or responsibility, and to regain a restrained yet empathetic stance toward respondents. In reaction to the emergence of difficult topics, we equipped interviewers with a series of folders from psychosocial service centers (e g. violence prevention, drug abuse, sexuality). They could be handed out after the interview, in case the researcher felt any of these topics were or might be an issue.
Ethical Issues
Conducting this type of research revealed a variety of ethical issues. While some are related to what Guillemin and Gillam (2004, p. 264) call “procedural ethics,” most of them are aspects of “ethics in practice” (Guillemin and Gillam (2004, p. 264)), i.e. “day-to-day” ethical issues and decisions that arise in all phases of research. First, the decentralized and longitudinal research design raised concerns regarding the protection of the participants’ privacy and anonymity. Second, conducting research within the institutional context of schools implied difficulties in establishing informed consent with respondents. Third, it was necessary to address power relations in the field and think about establishing a suitable role as researchers.
Protection of Privacy and Anonymity
Finding a consistent way to anonymize participants’ personal data, which could form the basis of upcoming waves of inquiry, was a key challenge for the first wave of data collection. A set of pseudonyms for all pupils and teachers of the participating classes—which met the quality standards of qualitative research without divulging the critical information retained in the key identifiers (e.g. ethnicity) (Flick et al., 2009; Reinders, 2005)—was centrally generated by two team members and provided to the researchers involved. Additionally, every interviewer and the respective person transcribing an interview was provided with a table of existing pseudonyms for school pupils, teachers and interviewers. This was necessary in order to co-ordinate the names across (1) memos and transcripts, since both interviewer and observer wrote an interview memo and an additional person worked on the respective interview transcript; (2) different waves; (3) different interviews with classmates or friends who mentioned the same colleagues or teachers. Accordingly, the table of pseudonyms was updated frequently and shared between the relevant researchers. Here, the help of the students was essential as they often reminded us of names we had to include in the list, like former classmates who dropped out of school, but were mentioned by several pupils.
Supplementary pseudonyms for people mentioned in the interviews, such as relatives or peers beyond school, were generated by the researchers working with the respective cases and written into “pseudonymization protocols” that were created for every case. These files will be supplemented and disseminated from wave to wave. All personal data were saved on a secured server and only the steering committee of the project has access to them. All other researchers using the data only receive pseudonymized data.
A significant problem in longitudinal research when working at case level is how to protect the participants’ privacy over time (Thomson, 2007, p. 579). As the research project proceeds and the amount of specific information pertaining to respondents expands, their anonymity could become difficult to maintain. For example, if there is only one girl in the sample that gave birth at the age of 15, it is hard to present findings about her without revealing her identity. Even though we can quite easily guarantee anonymity toward an international academic audience, for example, by removing more detailed information about the pupils and the institutions involved, this is of course more difficult when sharing results with the participating schools or the respondents themselves. As both the co-operating schools and the participating young people were interested in our findings, we decided not to present analyses about any individual cases, but only typologies and other findings on a more general level in these contexts.
Informed Consent
Like ethics in general, informed consent is a continuous and repeated undertaking (Marzano, 2012). Our informed consent procedures were embedded in institutional frameworks. As the fieldwork was conducted in the participating schools, their institutional logics as well as the specific cultures of the five schools had an impact on researchers’ work. This required a concerted effort in establishing a process for negotiating voluntary and informed consent with young people, which had to be tailored to the context of school (Morrow & Richards, 1996; Sin, 2005). In the case of this project, there seemed to be a noticeable incongruity between the premise of voluntary participation in qualitative research and the institutional logics of school, where in the latter, participating in classroom activities is not a matter of choice for the pupils. From the first time the researchers entered the school classes, this disproportion was noticeable. The teachers’ support of the study was experienced as ambivalent by the researchers: The teachers acted as our gatekeepers and their support was essential to get access to the young people and organizational help for conducting the interviews. At the same time the teachers generally encouraged the pupils to participate in the study. Even though there was no direct pressure and no consequences for pupils who did not agree to participate in an interview, the teachers’ obvious interest in the project might have swayed the pupils’ decisions.
Our approach to this situation involved establishing consent in several stages in order to be sure the pupils really participated voluntarily. We started our fieldwork with meetings and phone calls with principals and teachers of graduating classes in order to obtain institutional as well as individual teachers’ agreements for conducting our research in the schools and in certain classes. When this was settled, the teachers handed out parent letters with information about the project and an opt-out form. The decision in favor of opt-out forms was much debated among the team, but was motivated by indications about opt-in methods reinforcing the social selection of pupils participating: The failure for parents to opt their children into research is strongly linked to a low socio-economic status and does not explicitly express refusal (Jones et al., 2013). As a matter of fact, none of the parents opted their child out. In a next step, one staff researcher and five master’s students introduced the project to the pupils at each of the five schools. One of our intentions was to illustrate that the project differs from other school activities. Accordingly, we encouraged contributions and questions by the pupils and we repeatedly emphasized that participating in the interviews was not a school-related obligation, but a voluntary decision (i.e. an opportunity for the young people to “have their voices heard”).
Following our presentation, we introduced the idea of school walks and asked who wanted to participate in the school walks and who wanted to stay in class. We were surprised that all the pupils wanted to participate in every class. After returning to class, the researchers asked pupils interested in participating in the project to write their names on a list called the “Let’s take part list.” Only those who had written their name on the list were considered as potential interview partners. When the researchers came to school for interviews, the pupils had to agree again that they wanted to be interviewed rather than participate in the lessons. At the beginning of each interview, the young people were informed about the project, how their data would be used by the interviewers, and that there was a possibility to opt out at every stage of the project. They also received a leaflet with information about the project. After the interview, the pupils and the interviewers signed a mutual informed consent form.
With these multiple opportunities to opt out, we tried to counteract institutional and structural pressures to participate and give the young people the greatest possible “degree of control over their participation” (Conolly, 2008, p. 206). This process of establishing voluntary, informed consent goes beyond the recommendations given in school research (e.g. Drinck, 2013).
Finding a Position in the Field
Beyond the process of gaining informed consent, researchers needed to reflect upon research within a social environment that is shaped by hierarchical power relations, especially along the lines of class, ethnicity and age. For the adult researchers, this first meant the challenge of creating a new, independent position within this field (Jones et al., 2013, p. 331). Second, researchers had to account for the structural differences in class and age between researchers and respondents—evoking a similarity to the teacher-pupil constellation—that were likely to reproduce hierarchical power relations. This might have had an impact on the respondents’ narrations and self-presentations (Arztmann et al., 2016; Manderson et al., 2006; Russel, 2005). As Lumsden (2012) argues, these issues are typical for research engaging with marginalized young people and necessitates a path between deceptive notions of neutrality and simply siding with the “underdogs”. While we could not solve these tensions, we agreed upon a careful reflection on what type of attitudes to adopt when approaching the young people and making this explicit in the field notes and interview memos (see “Materials and Instructions” section).
Conclusion
Holland et al. (2006, p. 22) describe a vision of “scaling up from small-scale qualitative enquiries” without losing the qualities of small-scale research. They suggest doing this by either connecting QLR with quantitative data sets or by collaboration between several teams and institutions to pool their data. The former is performed, for example, by Schlimbach (2015) at the German Youth Institute, the latter by the
In this paper we have described the solutions developed by us to meet these challenges. In
Ensuring adequate data quality turned out to be a challenge that we met with intense training, detailed instructions for fieldwork, formalization of procedures and standardized templates for data documentation. Engaging in longitudinal research with young people in the context of school also confronted us with ethical challenges, especially concerning anonymity, informed consent and our approach toward structural hierarchies. Developing an elaborated system for pseudonymization and establishing a multi-step informed consent process were important measures when facing the first two challenges, enabling us to protect the pupils’ identities and make consent as informed and voluntary as possible. Furthermore, we made self-reflexivity one of the main issues in meetings and data documentation, allowing us to reflect upon our position in the field.
Even though the above described measures proved to be vital for the implementation of the project’s first wave of data collection, we experienced most impressively the importance of retaining a certain degree of flexibility—regarding the project’s organizational structure, the placing of responsibilities, and naturally the data collection. Ultimately, not everything went according to plan: Sometimes interviews were canceled or conducted poorly, institutional environments did not always offer the best conditions for creating an atmosphere of trust for an interview, and in some instances, organizational structures impeded quick decision-making. It proved helpful, though, to reserve as much time and space as possible for reflection of ethical and methodological challenges and to establish a good communicative basis within the team of researchers, whether it consisted of students or more experienced researchers. It was sometimes astounding how much effort individual researchers dedicated to the team, often unexpectedly, whether time resources when an interviewer suddenly fell ill on the day of an interview, or technical expertise—setting up a communication forum on an e-learning platform on short notice, or the psychological expertise necessary to help someone process a difficult interview experience. These resources could be drawn upon due to a respectful and appreciative atmosphere within the whole, large project team, with senior researchers working alongside student researchers all the time, taking everybody’s ideas and experiences seriously regardless of their academic position.
Supplemental Material
Supplemental Material, Annex_1_Template_Interviewmemo - Insights From Setting up the First Wave of a Qualitative Longitudinal Study
Supplemental Material, Annex_1_Template_Interviewmemo for Insights From Setting up the First Wave of a Qualitative Longitudinal Study by Mag. Dr. Veronika Wöhrer, MMag. Dr. Andrea Jesser, Barbara Mataloni and Andre Schmidt in International Journal of Qualitative Methods
Supplemental Material
Supplemental Material, Annex_2_Template_School_walk - Insights From Setting up the First Wave of a Qualitative Longitudinal Study
Supplemental Material, Annex_2_Template_School_walk for Insights From Setting up the First Wave of a Qualitative Longitudinal Study by Mag. Dr. Veronika Wöhrer, MMag. Dr. Andrea Jesser, Barbara Mataloni and Andre Schmidt in International Journal of Qualitative Methods
Supplemental Material
Supplemental Material, Annex_3_Online_Interview_Plan_and_Contacts - Insights From Setting up the First Wave of a Qualitative Longitudinal Study
Supplemental Material, Annex_3_Online_Interview_Plan_and_Contacts for Insights From Setting up the First Wave of a Qualitative Longitudinal Study by Mag. Dr. Veronika Wöhrer, MMag. Dr. Andrea Jesser, Barbara Mataloni and Andre Schmidt in International Journal of Qualitative Methods
Footnotes
Acknowledgments
Declaration of Conflicting Interests
Funding
ORCID iD
Supplemental Material
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
