Abstract
Capturing the voices of the ‘disadvantaged’ in society may often be difficult to achieve. In this article, we argue that the combination of data triangulation and a ‘slow’ interview technique may be a good method when the target group is vulnerable people who may find the interview challenging. Data triangulation can reveal a social phenomenon’s complexity by providing a fuller picture, while in-depth interviews and a ‘slow’ interview technique can enhance data quality. We present an alternative to what we perceive as being dominant and tacit support for use of voice recordings in interview-based studies, thus our research approach goes against the current flow in the research community. In the absence of a voice recording, the quality of the interview can be assured through cooperation between the researcher and the informant. We also discuss the importance of establishing trust when meeting the informant in order to ensure rich data.
From statistics to life stories
As social scientists, our mission is to contribute towards new or richer knowledge about a phenomenon. In this article, we will show how we have developed a data collection concept, which provides us with rich data about a phenomenon where statistics is plentiful, but where data behind the statistics is scarce. The phenomena we have studied are disability amongst young adults. In Norway, there are comprehensive statistics available about disability by virtue of national registers and epidemiological studies. The latter often include socio-economic variables, social variables and variables relating to the employment market. Some are life epidemiological studies based on data from various stages in life, whilst others also include multi-generational variables. These studies identify risk factors and calculate the probability of obtaining a disability pension later in life, based on predefined and standardized variables, but these do not provide any answers in terms of causality. Our research is a supplement to these quantitative studies. We go more in-depth with the aim to explore the life stories and processes behind the statistics.
Before being registered disabled there are a set of many events, experiences, choices and decisions that are not captured by these statistics. Insight into the various assessments from various actors as well as their actions will thus enrich the knowledge field about why some young adults are unable to get a foothold into the employment market.
This article is aimed mainly at researchers with an interest in disability issues, but the methodology may also be of interest to social workers and sociologists who interview vulnerable groups in their research. In this article, we argue that trust in the interview situation is important for obtaining rich data, and that trust can be built between researcher and informant during the interview by using a ‘slow’ interview technique combined with full transparency between researcher and informant. The researchers behind this article have extensive experience interviewing both articulate and busy professionals who master the power of words, and people who are far removed from the employment market and who are less assertive, confident and proficient with words. From our experience, one size does not fit all, making it all the more important to focus on how to interact with and record the stories of the least proficient.
As life stories and contexts are our primary field of interest, we start by using a case approach. According to Yin (2003: 13–14), a case study is an empirical inquiry that investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident. Further to this, the case study inquiry copes with the technically distinctive situation in which there will be many more variables of interest than data points, and as one result rely on multiple sources of evidence, with data needing to converge in a triangulating fashion, and as another result benefits from the prior development of theoretical propositions to guide data collection and analysis. Our approach has been an embedded multiple-case approach (Yin, 2003: 40). In each case, we used a data triangulation method where the interview itself played a central role. In several of the cases, we also read the informants’ case files prior to the interviews.
Before we take a look at what we call a ‘slow’ interview technique, we will first look at the contexts in which we use this technique; these are the data triangulation method and interviews. Finally, we discuss the importance of our data collection method for both the interviews as well as the quality of the data.
Triangulation
Triangulation is described in the literature as an approach where the researcher uses either multiple methods, several theories, different data sources (in time and space) or different independent researchers in order to strengthen the study’s credibility (Denzin, 1978). In other words, this means retrieving data from multiple sources and then comparing across the data sources (Miles and Huberman, 1984; Mathison, 1988; Ryen, 2012; Thagaard, 2013). According to Mathison (1988), researchers from this field of literature tend to agree that triangulation is a good thing and that research and evaluations will be strengthened by following such a practice. Using triangulation, researchers will have the opportunity of gaining a better and broader understanding of the phenomenon under investigation: …triangulation is supposed to support a finding by showing that independent measures of it agree with it or, at least, do not contradict it. (Miles and Huberman, 1984: 235)
Not only can triangulation be used to provide a fuller picture, Ryen (2012: 201) also points to other strengths this method has: Triangulation allows for asking different but complementary questions within the same study that the interpretation of one dataset may open up for understanding another dataset, thus allowing the researcher to get closer to the phenomenon in question and a greater or more nuanced understanding of the issue. An important disadvantage of triangulation is that it takes time, is costly, and can be difficult to repeat or make comparative studies afterwards. Additionally, in order to produce a more standardised report, researchers can be tempted to carry out data analyses where they try to compare data that is not comparable (Ryen, 2012).
The discussion around triangulation as a research strategy is based on two assumptions (Mathison, 1988). First, there are inherent biases in all data sources, made evident when researchers apply different methods or data sources and make these biases visible. Second, one can obtain a convergence towards the ‘truth’ of a social phenomenon when using triangulation. In terms of the latter, Mathison (1988) states that sometimes one can find that the data do converge, but often the data are inconsistent and sometimes contradictory. In such cases, the responsibility to ensure validity of data is placed on the researcher, who must find probable explanations for the phenomenon being studied. Triangulation will in practice provide a richer more complex picture of some social phenomenon, but rarely a clear answer regarding what is happening. This is supported by Patton (1980), who writes: There is no magic in triangulation. (…) ‘The point of triangulation is to study and understand when and why there are differences’. (Patton, 1980: 330–331)
The interview
Interviews are widely used as a data collection method in the social sciences, where the purpose is to reveal other people’s views, descriptions and perspectives on the themes that are addressed. Thagaard (2013: 95) writes that the interview provides a particularly sound basis for gaining insight into people’s experiences, thoughts and feelings, while Kvale (2001) emphasizes that personal interviews provide a unique access to the world of the informants’ as well as providing an understanding of how they perceive the world as the basis for their actions.
Briggs (1986: 1) refers to one estimate indicating that 90% of social science researchers use interview data. This is reason enough to have a heightened awareness of the strengths and weaknesses of the qualitative interview and explore new ways to obtain better and rich data.
What interview data really says something about is described differently throughout the literature. From a positivist perspective, the researcher is an objective or neutral recipient of the informant’s experiences, while from a constructivist perspective, interview data are considered a result of a social interaction between the researcher and the informant (Thagaard, 2013: 95; Rapley, 2007: 16). According to Briggs (1986), it will not be possible to create a situation where the researcher does not in any way influence the interview. Briggs is, for example, concerned that the interviews are formal and almost always quite different situations from those in which people usually act. He argues that it is important to use a wider range of techniques in order to provoke and obtain a greater range of more controlled responses. Briggs continues by emphasizing that such research requires great insight into the qualitative research interview, that it is not something one creates hastily, but that it demands a primary focus in the research. Kvale and Brinkmann (2009) support such a perspective, commenting that in interviews with, what they call elite persons, the asymmetric power relationship between researcher and informant may be more balanced due to the elite person’s power. It is debatable whether, for example, a primary school teacher sees herself or himself as such an elite person, as Kvale and Brinkmann (2009: 99) point out. Järvinen and Mik-Meyer (2005) discuss how elite persons are likely to play a dominant role in an interview, in contrast to marginalized or vulnerable persons.
With regard to vulnerable groups, it can be both difficult to obtain consent for the interview as well as to get the informants to share information, especially if the interview topic addresses sensitive personal information. The focus in this article is on the latter: How researchers can meet the informant in such a way that the informant is willing to open up and share relevant information to the research topic.
Philosopher Martin Buber (1923) is concerned with the meeting between people, i.e. how one person can meet and ‘see’ the other person. He writes about the importance of being willing to give of ourselves and to facilitate for truly ‘seeing’ and ‘listening’ to the informant in front of you: Nothing is constituent of an experience; nothing opens up other than through direct meetings that use the power of reciprocity. (Buber, 1923: 26) [unofficial translation] Every relationship in the world is exclusive – it excludes something else. Others come in and avenge their exclusion. (Buber, 1923: 91) [unofficial translation]
Our approach
In our approach, we utilize data triangulation along with a ‘slow’ interview technique. Data triangulation broadens our analysis by shedding light on the topic from different perspectives, while the ‘slow’ interview technique provides depth through reciprocity and the establishment of trust between interviewer and informant. It is the ‘slow’ interview technique which we regard as the most important element in this approach, and therefore it is given most space in this article. It is nevertheless important to emphasize that it is the combination of the two approaches that ensures rich data.
Data triangulation
Triangulation is composed of various data sources, which can be used to shed light on the same phenomenon, and in our approach, different perspectives are provided by actors who have held various positions in relation to a theme or a decision-making process. This approach strengthens the studies internal validity, especially in cases where actors from different positions provide similar responses. As researchers, we interpret this as a key finding or theme in our data towards approaching a fundamental understanding of the theme or topic. If, on the other hand, the responses differ, we broaden our empirical understanding of the theme. Hoel (2005) argues that triangulation is a necessity in some subjects, such as in the social sciences or professional subjects, because it can be a tool, which expands our understanding within a field characterized by close relations with the public, moral dilemmas as well as tacit knowledge.
In our studies, the informants belong to two main groups: People who are recipients of social security support payments and people who are employed to exercise a profession. In four of our studies, both sets of informants are represented as groups: Two of the studies encompassed recipients of a temporary or permanent disability pension (Jentoft and Olsen, 2009), the third encompassed recipients of a rehabilitation allowance (Olsen et al., 2006), while the fourth encompassed young adults who are recipients of a disability pension (Olsen et al., 2009). In all of these studies, case officers from the former National Insurance Service (NIS) and/or Public Employment Service were interviewed. In the fourth study, the NIS’ advisory medical consultants were also interviewed. In all of the four studies, the unit of analysis was the recipient of the social security support system.
In addition to the above, we conducted a study of early intervention in child neglect, where the ‘school’ was the unit of analysis (Olsen and Jentoft, 2013). In this study, only public sector employees were interviewed and data triangulation was addressed by interviewing various professions represented, including teachers, head teachers, school nurses and employees of the child welfare authorities, as well as the Norwegian Labour and Welfare Administration (NAV). This latter study was based on the findings from the previous study (Olsen et al., 2009) where young recipients of disability pensions connected childhood experiences with disability in early adulthood.
File access
For two of the studies, we obtained access to case files prior to the interviews. In the first study, our unit of analysis were recipients of rehabilitation allowance (Olsen et al., 2006), while in the second study, the unit of analysis was young adults who receive a disability pension (Olsen et al., 2009). Both of these studies were registered with the Data Protection Official for Research at the Norwegian Social Science Data Services (NSD), and access to the files was made conditional on the recipients’ informed consent. The case files in question were the recipients’ case files from the NIS and the Public Employment Service. In the first study, the files were extensive and provided a wealth of information, making it possible to reconstruct much of the interaction between the main actors and gain insight into the decision-making process. The case files contained the formal written documents from caseworkers, statements from medical specialists and other individuals who had recorded something in writing about the case. Several files also contained informal handwritten notes from phone calls and meetings. The information from the files gave us a good vantage point for the interviews to follow. By providing us with a basic understanding of the formal elements of the decision-making process, less time needed to be spent on these issues in the interview. On the basis of this experience, we adopted the same method when designing the study where young disabled people were the target.
When it came to the study with the young disabled people, access to the case files not only allowed us to gain insight into the case prior to the interview, but was also a key – a door-opener – into the informant’s world. By allowing us to read their case files, the informants unlocked the door to their world. When we met them for the interview, we already knew part of their story, also parts that otherwise would be difficult to reach. This made it easier to access difficult or sensitive questions in the actual interview. Further to this, having read the case file prior to the interview also meant that we could spend less time on talking about ‘hard facts’ that already were documented in the file, and more time on the why questions and what the informants believed to be important explanations for events in their lives. The case files thus constituted an important foundation from which we could build on in the actual interviews.
A ‘slow’ interview technique
A ‘slow’ interview technique takes us further into the phenomena we want to study. It is important to emphasize that a ‘slow’ interview technique can function without access to case files, but in the two studies mentioned above, access to case files was a strong contributor towards gaining a deeper understanding of the informants. In another study where ‘school’ is the unit of analysis, the case file access process was not adopted. In the latter study, data triangulation was carried out through using interviews with various professions. Here, the ‘slow’ interview technique was also applied.
There are two main elements in the practical implementation of interviews that give rise to the name ‘slow’ interview technique: We do not use voice recordings, but instead use handwritten notes, which take more time to write. We additionally use a dialogue-based quality assurance process. These elements will be described below, along with a discussion of the importance of this technique.
Handwritten notes
Using the ‘slow’ interview technique involves de-selecting audio recording and writing interview notes by hand. This is a conscious choice made by the researcher for at least two reasons: First, it is a visible reminder to the informants that they are being interviewed, as opposed to a casual conversation. Second, it takes time to write – time that the informants have at their disposal to further reflect on the questions and what to answer. We have experienced that the more time the informant has available to reflect over the questions, the more the informant supplements and enriches the data throughout the interview (Olsen et al., 2009).
The interviews were semi-structured face to face interviews. The interviews covered the same themes and were structured in a similar manner, but at the same time allowing for the individual stories. Using this ‘slow’ approach, we persecuted the tracks and did not break up the informant’s story. While we made the notes, the informants stories were sorted in meaningful categories. The semi-structured format made it possible to a certain degree categorize along similar as well as cross-sectional themes. Although the researchers asked the questions, it was the informants who at all times controlled how much additional information they wished to share with the researchers. If questions that we were interested in were not covered, we would follow up with questions. The informants were informed from the outset that they need not answer a question they felt uncomfortable with.
Dialogue-based quality assurance
Before the interviews began, the informants were asked if they would be willing to read through the notes from the interview, and comment and correct them if needed. The reason for this was twofold. First of all, it underpins a quality assurance system. Second, it enables the building of trust between the interviewer and interviewee. The notes were sent to the informants via post or email once they had been transferred from a handwritten state to an electronic document. The informant was invited to amend, delete or add text, and then return the document to the researcher. It is then the revised or ‘quality checked’ version, which forms the basis for further analysis.
This approach is selected because we want to ensure that the data recorded from the interviews is in line with what the informants want to convey, i.e. it is the informants’ opinions and perspectives we want to record, not the interviewer’s perception of what the informant may have meant. In the interviews we focus on what the informants believe to be important relationships and causalities in their life stories: What were they affected by, why they acted as they did, how they explain outcomes, etc. The time it takes to quality check the data, also extends the time for reflection. Since it may take several days or weeks before the informant receives the interview notes for quality check, the informants have added time to ponder over the questions and enrich the data. For us, it has been more important to collect data which gives a fuller picture of the informant’s story, than in which context the story was told in (in the interview situation or after the interview). We do not see the interview as concluded before the interview notes have been returned to the researcher after the quality check. This is part of a ‘slow’ interview technique that we call a dialogue-based quality assurance technique.
By including the informants in ‘checking’ the interview data, we not only use a participatory method to assure the quality of the data, in addition we demystify the interview notes. Since the informant knew at the offset of the interview that he/she would have full access to read and amend the interview notes, they could be at more ease during the interview. Nothing would be recorded which they would not be in agreement with or which would be secret to them. Through this openness, the informants were invited into the world of the researcher: What did the researcher write down, what did the researcher omit, how did the researcher understand what I tried to convey. By sharing the notes and inviting the informants to read and amend them, the researcher gave something back to the informant, and thus reinforced the trust relationship.
In our research, we have been keen to facilitate a good meeting between interviewer and interviewee. Several informants have since told us that this was the first time someone was willing to listen to what they had to say. We did not rush the interviews and space was given for the informants to be seen – to become ‘visible’. As researchers we, on the other hand, experienced many emotional and tough interview situations, which definitely have left their mark. As Buber (1923) states, when you want to meet other people on their own terms, there is always a victim and a risk in reciprocity.
Against the flow
Our data collection method differs from what is standard practice today, at least if we take what is recommended in the literature as a starting point. We believe that our method has made it possible to attain rich and qualitatively good data from informant-groups which have been difficult to reach through more current standard practices.
Textbooks in methods are concerned with how to record an interview via sound recording, video recording, note taking and use of memory. According to Kvale and Brinkmann (2009), the most commonly used methods are voice or sound recordings, because then one can secure a verbatim reproduction of the interview, as well as concentrating on the interview subject and the dynamics of the interview. However, these authors do not problematize the use of sound recordings, neither with regard to how it can have an alienating effect on the informant, nor that sound recordings can tend to expediate the interview and not leave enough time for reflection. Conversely, they are more critical of the use of memory and note taking by hand. Kvale and Brinkmann (2009) believe it can be disruptive to take extensive notes because it can break the conversational flow, and as researcher, you get less time to register the behaviour of the informants during the interview process. The authors also make a point of how memory may have certain limitations, but that it may also have its advantages. The latter being that it can act as a selective filter potentially to retain precisely those meanings that are significant for the interview theme and content (Kvale and Brinkmann, 2009). Other authors also strongly recommend using sound recording when preparing qualitative interviews. Repstad (2009) summarizes this literature succinctly when he writes that the interview situation is in itself artificial, therefore the sound tape will not make much of a difference.
In some of the literature, however, the disadvantages of using sound recordings are discussed (Ryen, 2012; Repstad, 2009). Ryen (2012) notes how many researchers have found that interviewees keep talking after the tape recording is turned off. Many feel that the informants distort or alter the theme at this point in time. Ryen believes that it is reasonable to interpret such behaviour as a result of the sound recorder being turned off and that the informants therefore feel freer to open up. If this is the case, then it can be very problematic. It indicates that the use of sound recordings actually affect the quality of data, at least in as much as important information is held back and/or the informants’ stories are incomplete in essential areas (Ryen, 2012: 113). Repstad (2009) also discusses how some informants are almost allergic towards voice recordings. The author provides some practical guidelines on how to minimally disturb the interviewee by the presence of the voice recorder. Further to this, he advises the researcher to write down what is discussed after the sound recording is turned off. If the informant opposes the use of this material, then it can be used as background material in connection to the case instead (Repstad, 2009). These authors do not, however, discuss the possibility that the informants may be holding back important information throughout the whole interview. Neither do they reflect on the importance of tempo and sufficient time to reflect during the interview.
We have not found writing notes by hand to be a problem, like Kvale and Brinkmann (2009) purport it to be. In Olsen et al. (2009), 63 interviews were noted for hand, while in Olsen and Jentoft (2013), 100 interviews were handwritten. Some of these interviews took over 2 h to complete. All the interviews have since been transferred to electronic documents. We can add that when the tempo during the interview is slower, we manage to note down most of what is said, often in verbatim. This, combined with dialogue-based quality assurance, makes it possible to use direct citations from the interviews. It is a demanding job, but it can be done.
The proponents of voice recording also argue that it is difficult to establish and retain contact with the interviewee when writing notes by hand. We have not found this as a notable problem. Contact with the interviewee is established when we greet and through small talk in the initial phase of the meeting. Further to this, we have frequent eye contact throughout the interview, along with the active use of body language like nodding, smiling and facial expressions. All of this signals to the interviewee that we are paying close attention to what they are saying. Both through active listening and relevant follow-up questions, contact with the interviewee is retained. Neither does handwriting the notes seem to hinder the free flow of the conversation. Instead, it seems to offer natural pockets of time which the interviewees may use to reflect over the theme in question. If the interviewee seems uncomfortable or upset, we stop up and respond to the best of our ability.
After many years of interviewing, we have found that recording the interview by use of a tape recorder, or equivalent, often entails a quicker pace. The sound recorder makes the interview more time efficient. But to gain time efficiency, something else may go lost on the way: Time for reflection. In a modern world where technological innovations penetrate our lives in so many ways, we pose the question whether we need to stop and reconsider. Just because technology is available, does not mean we have to use it or that it is superior. Maybe we should take time back in the interview. At time of writing, we are not aware of any ongoing current discourse regarding sound recordings impact on the pace of the interview and data quality.
We have made a conscious choice to return to an ‘old’ way of working, namely to note down interviews by hand. We do this because we believe it enriches the data by providing more time for reflection. Not all technological solutions should be seen as a step forward. Sound recording captures ‘the word’ there and then, it becomes one moment’s conversation, but the question is whether it captures the story the informant wishes to convey. By allowing more time for each question during the interview, and allowing the informants to read, comment and emend the interview notes in their own time, more space and time for reflection is created.
By opting out of modern technology like sound recorders, we go against the flow. Our ‘slow’ interview technique provides adequate time to listen and ‘see’ the informants. This in turn provides us with an abundance of high quality data.
Time for reciprocity
Throughout our studies, we have been primarily concerned with content analysis. We are interested in mapping how the informants understand the world, and how this influences their decisions and actions. We have been less concerned with how sentences are worded, and more concerned with threads that weave a life story.
There seems to be a large degree of consensus that triangulation is a good thing, and when used correctly how this can strengthen research (Mathison, 1988). By connecting the perspectives of various actors who have a good knowledge of a phenomenon, a broader understanding of the theme or topic can be gained and richer data may be collected. This is particularly true in studies of high complexity. In our studies, we have interviewed actors who have different positions relative to the recipient of social security benefits. This helped to shed light on the phenomenon from multiple perspectives and gave us a more complete picture of the phenomenon. The validity of the findings also increased in those cases where the informants confirmed each other’s stories and perspectives. In this perspective, data triangulation can also be seen as part of the quality assurance of the data.
The interviews were designed in such a way that the interviewer introduced the theme initially, followed by the same opening questions. However, it was the informants themselves who determined the time and direction of the remaining interview. The informant could speak freely, often covering the questions we were interested in. In cases where this was not the case, we would follow up with questions, thus ensuring comparable data. In the interviews, we also used a ‘search and uncover’ method. As a consequence, we may to some extent have influenced the narrative that emerged. We have not made a major issue of this in our analysis, in contrast to what interactional inspired researchers do. However, we account in detail for the research processes and make the reader aware of the distortions that may have occurred in the data due to our method (Olsen and Jentoft, 2013).
We view the access to the informants’ files and the consent to be interviewed as an act of generosity and an expression of trust. As researchers we reciprocated by including the informants into our world. We did this through full transparency with regard to our field notes and by letting the informants have the final say in how the interview notes were worded. This they did by reading, checking, and, where necessary, amending the notes.
The relevance of a ‘slow’ and ‘non-slow’ interview technique for the interview situation and the dataset.
During the interview, it is important to achieve good contact and interaction with the informant. The focal point must be the informant and his or her story. Being present for the informant, acknowledging them, and actively ‘seeing’ them, is therefore important to achieve a good dialogue. Everything surrounding will be perceived as noise and may distract from the good dialogue, and may even lead to disconnection or alienation in the interview situation (Buber, 1923). In a sense, we can say that we are not interested in ‘smoking’ out our informant’s views, but rather in giving them room to answer reflectively. The first words to pass your lips are not always the most reflected. Given time in the interview situation, the informant may wish to supplement the initial answer, explain what he or she meant, illustrate the answer with an example, or amend the answer after further reflection. A spontaneous answer may not always be the most ‘complete’ answer: There may be more to the story when the informant is allowed to give it some additional thought. Audio recordings can obscure meetings because the use of technology, consciously or unconsciously, may affect many people’s behaviour. We also believe that the pace increases when the interview is recorded by aide of technology. With audio recorders, there is no need for the pause which would be used for writing. The pauses therefore take on a new character: Moving away from being a room for reflection, and becoming uncomfortable silences and hence increasing the desire to push on to the next question. Using sound recordings, we therefore risk an encounter between two estranged people in Buber’s sense.
Further to this, since we describe how we intend to carry out the dialogue based quality assurance method at the outset of the interview, the informants are likely to feel more relaxed and at ease during interview. Knowing that you will be able to read and amend the interview notes, helps to demystify the note taking and ensures that the informant retains ‘ownership’ and control over the data. By using this quality assurance method, we are clearly stating to the informants that you will own your own words and I will not violate this commitment to you. This is an invitation to establish trust, which along with the ‘slow’ interview technique, promotes richer data.
Conclusion
Although interviews with employees from different professions or positions, and access to case files broaden our understanding of the issue at hand, it is the interviews with the young disabled informants, which gave us the most in-depth understanding of the social phenomenon we were studying. Whilst the different professional groups and government employees meet the young adults in different roles, at different times, and at different phases in life, it is the young adults themselves who are best positioned to tell us the most coherent and complementary case history. The young adults are also best positioned to take the researcher into their world in an attempt to understand what underlies their actions.
We believe we are best able to establish trust through using the ‘slow’ interview technique. The technique creates time and space for the informant to relax and feel at ease during the interview. The interview questions are open and the interview quickly develops into a conversation. Several informants clearly expressed that this was the first time anyone had ‘seen’ them as a whole person and listened to their story. Our approach also provides informants with the opportunity to take ownership of the data, by allowing them access to read and amend the case notes. The informant’s self-monitoring of data helped to create greater openness in the dialogue, diffusing potentially difficult situations and contributing towards mutual trust and understanding to be built between the researcher and the informant.
No method fits all purposes, but it is important to be conscious of the alternatives open to the researcher when selecting a relevant method. Sound recordings are naturally the best if the aim is to capture the instant verbal reaction of the informant. But if the aim is to capture the reflective ‘interview conversation’, then our method may be more appropriate. Studies differ in terms of how important it is to spend time with the informant during the interview. In some studies, the aim is to obtain in-depth reflections and time may be of essence, while in other studies, the aim is simply to gain answers to predefined questions, often as quickly as possible. The question we raise is therefore: Although sound recordings may capture the word, does it capture the message?
Table 1 illustrates some differences between the ‘slow’ interview and use of other techniques, here simply called the ‘non-slow’ interview. It is quite possible to imagine a mix, such as the use of sound recording but where one introduces a dialogue based quality assurance technique, for example.
In summary, we have found that this method works well when interviewing vulnerable groups who are not accustomed to being in an interview situation and who need to feel safe in the role of informant. We would therefore recommend this method to other researchers who have vulnerable groups as their target group. The method provided us with rich data and opened up for ‘discovery’ that lay beyond what we had imagined. The open and exploratory approach based on data triangulation and in-depth interviews with the vulnerable target group, gave us a new insight which meant that we had to discard several of our working hypotheses. This we take as a good sign. Further to this, our research findings have been well received nationally, largely due to our ability to bring the vulnerable young adults voice into the public debate.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.