Unlike expert interviews, elite interviews have long been established as a basic form of qualitative interviewing. This may be due to the long-standing tradition and eminent importance of elite research in sociology, whereat this research was and still is coined by diverse and sometimes contradictory theoretical paradigms and political standpoints (Bottomore, 1993). In its early days, elite research was driven by the idea that, since the power elite controls society to a great extent, empirical insights into the worldviews and interests of the elite are necessary in order to understand societal order and change. From its beginning, empirical elite research was primarily based on generating qualitative data (see Dexter, 1970). Applying standardised methods and restrictive designs was viewed as inappropriate for investigating the elite empirically, since elites are used to developing their ideas in open communication and are trained to ad-lib. Therefore, conducting elite interviews, in other words, was (and still is) tantamount to carrying out qualitative interviews (see Roulston and Choi, Chapter 15, this volume).
In contrast, with respect to expert interviews, the widely held view is that even though expert interviews are frequently conducted in the many contexts and fields of social science, they are only rarely thought-through and to a lesser extent methodologically reflected (Meuser and Nagel, 2009). After all, in recent years the debate about expert interviews has gradually become more concrete (see Bogner et al., 2009). The focus of this debate lies primarily on issues of what constitutes an expert, the differences between the various forms of expert interviews and their role in research design, as well as the specifics of interviewing and interaction in comparison to other qualitative interview forms.
Regarding elite interviews, Littig (2009) has argued that qualitative data collection, [Page 653]with both elites and experts, faces similar methodological challenges: First of all, the access may be difficult in particular with regard to elites and high-level experts because they often tend to present themselves as unavailable (Conti and O'Neil, 2007; Pfadenhauer, 2009). Second, interviewee's statements may be influenced by various, subject-related variables such as gender, age or – and even more than in other interview settings – the professional status of the interviewer which he or she cannot control. Third, interviewers should apply different strategies of interviewing, dependent on their specific aims and the significance of the respective interview in the context of the research project (Mikecz, 2012). Short and clear-cut questions may result in a survey-style communication focusing on facts and information; in contrast, inviting interviewees to engage in detailed and extensive narratives may be helpful to gain insight into their worldviews and patterns of thought.
Expert interviews – again, similar to elite interviews – are now frequently considered a standard qualitative research method. With respect to the methodological debate (Flick, 2009), the expert interview is situated in the qualitative paradigm – even though, in principle, expert interviews can also follow standardised communication patterns as applied in quantitative research (survey). Today, qualitative expert interviews are carried out in different fields of the political sciences and social research, such as international relations, organisational research, policy research, gender studies, and so forth. Especially in the exploratory phase of a project, interviewing experts is regarded as a more efficient and concentrated method of generating data than, for instance, participatory observation (see Wästerfors, Chapter 20, this volume) or systematic quantitative surveys. Conducting expert interviews can serve to shorten time-consuming data-producing processes, particularly if the experts are the key to practical insider knowledge and are interviewed as surrogates for a wider circle of players. Expert interviews also lend themselves to those kinds of situations in which it might prove difficult or impossible to gain access to a particular social field (as is the case, for instance, with taboo subjects).
Beyond these efficiency aspects, the expert interview has attracted attention within the qualitative paradigm because experts have become a central object of empirical research in the social sciences during the last few decades. This resulted from a fundamental change: The role and the influence of experts in nearly all spheres of modern society have increasingly become problematised from science as well as from civil society actors. With an increase in counter-experts and laypeople challenging the knowledge claims of scientific experts, questions arose such as: Who is legitimately considered to be an expert? How concise, how certain or reflexive is expert knowledge? How is expertise produced in expert panels characterised by interdisciplinarity and a variety of worldviews and approaches? Obviously, the superiority of expert knowledge is no longer taken for granted even though (or because) its importance for individual everyday life decisions or political decision-making can hardly be denied. As a result, the expert increasingly becomes subject to empirical sociological research. In contrast, the elite interview was established as a standard qualitative research method long ago. This has to do with the fine tradition of elite theory in sociology.
A groundbreaking book on elite interviewing was first published in 1970 by the political scientist J.L. Dexter. In his understanding, elite interviews target particular social groups ‘the influential, the prominent, the well-informed’ (2006, p. 19), representatives of the political and economic elites, which might be reluctant to reveal their views or [Page 654]perspectives and therefore require special treatment:
It is an interview with any interviewee – and stress should be placed on the word ‘any’ – who in terms of the current purposes of the interviewer is given special, non-standardized treatment. By special, non-standard treatment I mean: stressing the interviewee's definition of the situation, encouraging the interviewee to structure an account of the situation, letting the interviewee introduce to a considerable extent (an extent which will of course vary from project to project and interviewer to interviewer) his notion of what he regards as relevant, instead of relying upon the investigator's notions of relevance. (Dexter, 2006, p. 18)
What might sound like common knowledge for qualitative interviewing today was certainly innovative at a time dominated by quantitative standards. Following these standards Dexter drew attention to the fact that the purpose of interviewing the elite is not simply to gather objective facts and knowledge. He stressed that non-standardised interviews are always influenced by the social relationship between the interviewee and the interviewer, and thus guides the interaction. Dexter also gave many practical hints on how to conduct elite interviews like using open-ended questions, being flexible or listening carefully. Dexter's main characterisation of elite interviews is still shared by more recent literature. In fact, his vague definition of the elite has in essence remained constant in the methodological literature to date (e.g. Moyser and Wagstaffe, 1987; Seldon, 1996; Odendahl and Shaw, 2002, Harvey, 2011). Furthermore, interviews with the elite in Dexter's tradition are often not seen as a precise research tool. The sampling is not representative, the statements made by interviewees can be distorted by gaps in their memories, different interviewees can give different information on the same topic, etc. (e.g. Richards, 1996, p. 200f.).
In recent years, the term ‘elite interview’ has been used in the Anglo-American tradition to describe interviews with ‘informants (usually male) who occupy a senior or middle management position’ (Welch, et al. 2002, p. 613) or ‘those who occupy senior management and Board level positions within organizations’ (Harvey, 2011, p. 433). This functional definition of ‘elites’ comes close to the understanding of ‘experts’ as used in German-speaking countries to describe the counterpart to elite interviews, namely expert interviews. The latter have been regarded as a distinct interview form for some years (Bogner et al., 2002; Gläser and Laudel, 2004). The methodological debate on expert interviews started in 1991, when Meuser and Nagel published their article on a common research practice, expert interviews, which had not been methodologically reflected until then. In the following years a vivid debate led to the rise of a variety of approaches, which thus can no longer be referred to in singular (Bogner et al., 2009, 2014). Expert interviews differ in the notion of ‘experts’ (from a broad voluntarist to a narrow functionalist understanding), the purpose of the interviews (explorative, systematic, theory-generating) and the more interpretative-hermeneutic or positivist understanding of expert knowledge. However, with regard to methodology, the differences between the expert interview, which the German-language literature primarily refers to, and the elite interview the Anglophone world focuses on, are small.
As far as elites and experts are concerned, recent social science research trends have proved relatively stable. Elite theory can be traced back to the early times of sociology including seminal studies such as Robert Michels’ description of the ‘iron law of oligarchy’ (Michels, 1911), Thorstein Veblen's Theory of the Leisure Class (Veblen, 1899) or C. Wright Mill's critique of The Power Elite (Mills, 1956). In general, elite theory is guided by the assumption that a small minority, [Page 655]consisting of members of the economic elite and policy-planning networks, holds most power; pro-democratic theories consider this to be a fundamental threat to modern democracies. Regarding experts, the sociological debate sets in at a point in time when eminent scholars point to the ongoing specialisation and differentiation of modern society associated with the rise of technical experts and an ‘intellectual', that is, science-based, technology (Ellul, 1964; Bell, 1973). Soon, these diagnoses of an increasing significance of expertise and expert systems fuelled a debate on whether modern societies are about to come under the experts’ dominance (‘technocracy').
In this context, the expert was considered an agent of truth and authority increasingly dominating political decision-making in modern societies (‘truth speaks to power'). In contrast, more recent contributions point out that scientific self-criticism and the rise of counter-experts – often backed by protest movements – have long since contributed to terminating this golden age of expertise (Turner, 2002). Obviously, with regard to controversial issues such as global warming or genetically modified organisms (GMO) every kind of (scientific) expertise can be fundamentally challenged – with the help of alternative expertise. Theorists of a ‘reflexive modernisation’ consider this development to be a moment of societal self-enlightenment (Beck et al., 1994). Following Giddens (1991), expert knowledge is part of the ‘institutional reflexivity’ that supposes all premises of individual and organisational activity will be routinely examined in light of new information. However, at the same time, this growth in relevance of expert knowledge is paradoxically accompanied by a crisis of recognition on the part of the experts. Today, most people are experts in challenging expertise by taking ‘alternative voices’ into account. In regard to sociology, experts and elites have moved into the centre of interest both from modernisation theory as well as from the sociology of scientific knowledge (cf. Jasanoff et al., 1995; Maasen and Weingart, 2005).
These different strands of sociological theory were conducive to the development of a broader and more profound notion of what characterises an expert. In general, experts are considered to be people with special knowledge or skills, most often equated with professionals from the fields of science, engineering and technology. In this perspective experts are primarily defined in contrast to their counterpart, that is, laity. With the emergence of autonomous fields of professional action that are responsible for innovation, namely research and technology, the expert-lay division became increasingly established in modern societies. However, in the context of qualitative research, experts are not primarily interviewed because of an interesting ‘solid’ or canonical knowledge as one can find in handbooks and encyclopaedias. In fact, qualitative methodology does not believe in ‘objective’ knowledge or ‘neutral’ facts. Rather, it is primarily interested in expert knowledge because it determines social practices and institutions to a certain extent (Bogner and Menz, 2009). In other words: the social relevance of experts in modern life and their ability to affect people's practices to a significant degree – this is why social scientists interview them.
Accordingly, experts can be understood as people who possess specific knowledge that relates to a clearly demarcated range of problems and plays an authoritative role in decision-making of different kinds. Due to this knowledge, their interpretations provide guidelines for social action and structure a particular field of social action in a meaningful way.
Obviously, in the context of qualitative interviewing, expert knowledge is not only of interest because it is characterised by a high degree of reflexivity, of coherence or certainty. Rather, our interviews are – at least implicitly – based on the assumption that experts are able to exercise power in a particular social context by applying special knowledge. To put it simply: Following a sociology of knowledge perspective, experts represent or personify a complex interdependence of knowledge and power. This does not mean that the experts’ power necessarily becomes manifest in political influence or economic wealth; rather, a good many times it may be limited. However, experts exert influence by determining the way people understand and interpret the world or particular problems; they exert influence by establishing a particular issue-framing, even if they are experts for the powerless or for neglected social problems. The particular significance of specific knowledge may be considered a major difference between experts and elites.
Elites are not primarily determined by special knowledge; rather, becoming a member of a certain elite group is mainly dependent on personal contacts and networks, family background, milieu, poise and habitus (Bottomore, 1993). Usually, elites – much more than experts – are characterised by an exceptionally high socio-economic status. They rely on inherited social privileges and merits. Experts, in contrast, are considered to have outstanding cognitive abilities and the societal acceptance of their authority is highly dependent on their performance. Thus, the notion of expert implies a meritocratic semantics. Certainly, there may be interferences between elites and high-level experts as sometimes experts are appointed a leadership position in research institutions, corporations or even politics. Thereby, these outstanding experts holding a superior position become part of a certain segment of the elite; this expert-elite hybrid is characterised by a combination of highly relevant, reflexive knowledge and a remarkable degree of power.
Despite these and similar interferences, the focus on knowledge and power may serve as a good starting point to come to a methodologically sensible definition of experts and elites. The following diagram (see Figure 41.1) illustrates our (simplifying) attempt to differentiate [Page 657]between experts and elites along the two dimensions of knowledge and power.
In light of this diagram, it becomes obvious that interviews with the elite can, but do not necessarily have to, be expert interviews. This is because experts, defined by their special or professional knowledge and their influence on decision-making processes, can, but do not have to, be members of an elite group. In the end, this depends on their formal position, their influence on social practices and the extent to which they are able to determine what is considered to be relevant or not in a particular field of social action.
Generally speaking, interviews (not only those with experts and elites) are aimed at uncovering some kind of knowledge the interviewees possess. If there were no differences in knowledge between the interviewer and the interviewee, interviewing would not make any sense. But what kind of knowledge is it the researchers are aiming at in interviews with experts and elites? At what kind of knowledge are expert and elite interviews aimed? As noted above it is not always ‘correct’ or ‘better’ knowledge the researcher wants to assemble. What form of knowledge the object of desire is, depends on the epistemological framework of the study that the interview is part of, as well as on its research design. At first we differentiate between three forms of knowledge relevant to the interview (cf. Bogner and Menz, 2009). After that we describe four types of interviews with experts and elites.
- Technical knowledge comprises facts and information about operations and events governed by rules, application routines specific to a field, bureaucratic competences, and so on. This ‘technical’ knowledge remains most closely related to the understanding of expertise as a specific advantage, where an expert's knowledge can be distinguished from everyday knowledge because it is more systematic in its content, better thought-out and may be more reliable (for example, academic knowledge or specialised knowledge about a specific social area like an organisation, cf. Schütz, 1964). In this context, an expert's knowledge provides a specific kind of advantage because it is more systematic, better thought-through and likely more reliable than what can be learned from other sources. Similarly, its prominent position in society grants the elite privileged access to a certain kind of information that the researcher does not have. From a rigorous methodological point of view, the strength of the expert and elite interviews lies not primarily in this field of knowledge (even if interviews, especially expert interviews, are quite often used for this purpose). The expert or member of the elite could be mistaken or hold very subjective views – in other words: he or she is a potential cause of error. Consequently, if other sources for the facts and information need – for example, documents, statistics, textbooks – are available, they should be used instead. However, conducting interviews for the purpose of collecting ‘objective’ data and information is inevitable if these alternative sources of information are not accessible.
- Process knowledge refers to the acquisition of information about sequences of actions, interaction routines, organisational constellations, and past or current events, in which the interviewee is directly involved or which at least are closely related to his or her field of action. Unlike technical knowledge, this is not specialised knowledge in a narrow sense (something one can acquire through educational qualifications), but more a matter of knowledge based on practical experience acquired through one's own context of action. It is therefore strongly connected to the expert or elite as a subject and not easily transferable. This form of knowledge is experience-based but unlike ‘tacit knowledge’ (Polanyi, 1966) it can be verbally expressed in an interview situation.
- Interpretative knowledge (Bogner and Menz, 2009) entails subjective orientations, rules, points of view and interpretations, and thus renders expert end elite knowledge a heterogeneous [Page 658]conglomeration. Interpretative knowledge does not only comprise (subjective) perceptions and descriptions of reality but also normative dispositions. As the researcher reconstructs this interpretative knowledge, he/she enters, to put it in old-fashioned terms, into the sphere of ideas and ideologies, of fragmentary, inconsistent configurations of meaning and patterns of explanation. It is not a homogeneous body of knowledge but a compilation of related but not necessarily logically structured interpretations. Interpretative knowledge does not presume that the expert or the elite has a ‘better’ access to reality (as is the case when the focus is on technical knowledge) than the interviewer. But he or she has a specific subjective perspective related to the research topic. This means the interviewee ‘is always right', interpretative knowledge is always true; it is a social fact on its own.
It is almost impossible to tell, on the basis of something said in an interview, whether a statement should be considered ‘technical knowledge’ and as such in no need of further interpretation, or ‘interpretative knowledge', that is the expression of a subjective construction of meaning on the part of the interviewee. The differentiation between the three kinds of knowledge is not based on any characteristics of the knowledge itself, but is primarily a construction of the social scientist interpreting it. It is always the result of an act of abstraction and systematisation performed by the researcher, an ‘analytic construction'.
For example (Bogner et al., 2014, p. 17ff.), if a manager states that ‘the activities of trade unions are harmful to the economic development of the firm’ the researcher can classify this as correct ‘objective’ information or as the result of personal experience, for example, in a survey about the economic impact of trade unions’ behaviour. In a more qualitative-oriented organisational case study the researcher would take this statement as ‘interpretative knowledge’ which is important not because of its truthfulness but because of the practical effects of it as an action orientation. If the manager's behaviour is guided by this ‘knowledge’ it will have an effect, for example, on patterns of industrial relations.
The analytic differentiation between the forms of knowledge in the interview makes it possible to describe the epistemological interest of different forms of the expert and elite interview more precisely. However, to distinguish between specific forms of expert and elite interviews, another differentiation is necessary: Interviews can take quite different positions within the research design. Especially expert interviews are quite often not the only method of data collection within a specific study. They are combined with other forms of interview techniques or with other qualitative methods, such as documentary analysis or observations. Furthermore, they are often part of a triangulation between qualitative and quantitative methods (Flick, 2011; Menz and Nies, 2017).
The expert interview owes its prominence in empirical social research partly to its use as an exploratory tool, which precedes the main methods of data collection. In both quantitative and qualitative research projects, expert interviews can serve to establish an initial orientation in a field that is either substantively new or poorly defined, as a way of helping the researcher to develop a clearer idea of the problem or as a preliminary move in the construction of a final interview guide. In this sense, exploratory interviews help to structure the area under investigation and to generate hypotheses. The experts interviewed may themselves be part of the group of interest to a study, but in many cases experts are deliberately used as a complementary source of information about the group of interest, the actual subject. In the latter case, the expert's role is that of someone who possesses ‘contextual knowledge'.1 Here, the main focus of the interview lies on the ‘technical’ and ‘process knowledge'. Its function is to gather initial – not systematic, but nevertheless ‘objective’ – information about the context of the research topic, which is afterwards investigated in more detail with other methods.[Page 659]
Expert and elite interviews can also have an explorative function if they focus on ‘interpretative knowledge'. Orientations, interpretations and evaluations are explored to get an impression of the field, for example, in order to formulate the first hypothesis, which can guide the further research. In this case the subsequent main study uses similar methods.
If the expert or elite interview is the (or one of the) main source(s) of data collection they can be called ‘grounding interviews'. There are two forms of grounding interviews: The systematising interview is oriented towards gaining access to exclusive knowledge – both ‘technical’ and ‘process knowledge’ – possessed by the expert or the elite. This kind of interview is an attempt to obtain systematic and complete information. The interviewee enlightens the researcher on ‘objective’ matters. The main focus, though, is not on the interpretative character of knowledge but rather on its capacity to provide researchers with facts concerning the research question. Interviewees are a source of information with regard to the reconstruction of sequences of events and social situations: ‘Experts are people who have special knowledge about social facts, and expert interviews are a way of gaining access to this knowledge’ (Gläser and Laudel, 2004, p. 10). From this methodological perspective it is not the experts themselves that are the object of investigation; rather they function as informants, providing information about the actual object of investigation.
The second form of the grounding interview is the theory-generating interview (for more detail, see Bogner and Menz, 2009; Meuser and Nagel, 2009). In this case the interviewee no longer serves as the catalyst of the research process, or, put differently, as a means by which the researcher obtains useful information and elucidation of the issue under investigation. In essence, the goal of the theory-generating interview is to communicatively open up and analytically reconstruct the subjective dimension of knowledge. Here, the action orientations and implicit decision-making maxims of experts within a particular specialist field, or the elites, are the starting point for the formulation of theory. The researcher seeks to formulate a theoretically rich conceptualisation of (often implicit, yet reconstructible) knowledge, conceptions of the world and routines, which the experts and elites develop in their activities and which are constitutive for the functioning of social systems. In ideal terms, this procedure seeks to generate theory via the interpretative generalisation of a typology – in contrast to the representative statistical conclusions that result from standardised methods. Following Glaser and Strauss (1967), qualitative theory is here drawn up via theoretical sampling (see Schreier, Chapter 6, this volume) and comparative analysis (see Sørensen et al., Chapter 10, this volume). This constitutes a process of inductive theory formulation, at the conclusion of which the researcher will ideally have a ‘formal’ theory. It follows that the theory-generating interview must be classified as part of the methodological canon, oriented along the fundamental principles of interpretative sociology.
To sum up, we can distinguish between two main forms of interviews with regard to their function within the research design (see Gobo, Chapter 5, this volume):
- interviews with an exploratory function and
- interviews used for systematic data collection in order to ground comprehensive empirical descriptions and theoretical concepts.
Within each type we can differentiate between an informational and an interpretative focus of the interview. Informational interviews aim primarily at technical and process knowledge whereas interpretative interviews – situated in the ‘interpretative paradigm’ (Wilson, 1970) – focus on orientations and evaluations as subjective (but not necessarily individual) perspectives of experts and elites (see Table 41.1).
|Table 41.1 Forms of expert and elite interviews in relation to their function in the research design and epistemological background|
|Exploratory interviews||Grounding interviews|
|Informational Interviews||exploratory data collection||systematising interview|
|Interpretive Interviews||exploration of interpretations||theory-generating interview|
Compared to ordinary, in-depth interviews the expert and elite interviews are characterised by a particular interaction structure. Usually, interviews are seen as having a certain kind of unbalanced power relations (there have been some efforts, for example from feminist research approaches, to reduce this asymmetry, e.g. Oakley, 1981). There often is a situation of ‘studying down’ (Plesner, 2011). The standard situation in interviews is an asymmetrical one in which the interviewer defines the setting and the topics. And it is seen as one of the main concerns of a good interviewing strategy ‘to make somebody talk', to give the interviewee complete expression for unfolding his subjective position and attitudes.
Conversations with experts and elites are different. In this case, the researcher communicates with people who are usually well aware of their expertise and their social position and who are used to being ‘in charge’ and listened to by others. Expert and elite interviews which can be described as ‘studying up’ (Plesner, 2011) are at risk of the interviewee taking over the structuring of the course (Gillham, 2000, p. 82). In some cases he or she even displays a patronising attitude towards the interviewer, attempts to show how well disposed he or she is, and to dictate the content of the conversation to the (seemingly) inexperienced or inferior interviewer – often with a gender-specific bias if a young female researcher interviews an older male expert or member of the elite (Abels and Behrens, 2009).2
To avoid such adverse asymmetrical communication some authors suggest that the interviewer should deliberately demonstrate his or her own expertise in order to gain the recognition of the interviewee.3 Trinczek, for example, argues that interviewers who want to conduct successful expert interviews with managers must, as an indispensable precondition, have expert status themselves or, as a minimum requirement, appear reasonably comparable to and an ‘equal’ of the interviewee in respect of age and qualifications.
The interviewer is indeed required to be an expert himself: the more an interviewer demonstrates knowledgeability during the interview by giving competent assessments, stating reasons, and raising counterarguments, the more managers in turn will be willing to offer their own knowledge and take a stance on issues, thus disclosing their subjective structures of relevance and patterns of orientation in absence of strategic considerations. (Trinczek, 2009, p. 211)
There are many striking arguments for this position. If the expert interview is seen primarily as something that will produce ‘useful information’ and elucidation of ‘facts’ (as is the case with the informational interviews), the high level of specialist interaction between co-experts will have a productive effect and the interview will be of value for the detailed analysis of the issue at stake. If, on the other hand, the goal of the investigation is the reconstruction of interpretative knowledge (as is the case with theory-generating expert interviews), the ‘technicist element’ becomes problematic, since the implicit normative and practical premises of expert opinion will be presupposed as a shared basis of the conversation between expert and co-expert, and it will be difficult to gain access to them for the purposes of analysis.[Page 661]
That is why some authors (Abels and Behrens, 2009; Bogner and Menz, 2009) stress that an adverse asymmetrical interaction situation where the interviewer is seen as inferior or naïve is not generally problematic. Instead of reacting to these paternalistic behaviours by displaying resentment for not being perceived or taken seriously as an expert in the desired way, interviewers would be better advised to turn this discriminatory paternalism to their strategic advantage, as a way of making the collection of data more productive. Naïve questions stand a good chance of producing the most interesting and productive answers – especially in the framework of a research design that seeks to generate theory. The interviewers have the freedom to do whatever they want, and can ask questions that under other circumstances would have endangered the stabilised scheme of expectation. This can make it possible to gain access to information that might not otherwise be revealed, particularly because a naïve interviewer is seen as especially trustworthy (Abels and Behrens, 2009). On the other hand, the disadvantages of this interaction structure are obvious: interviewees sometimes bore researchers with interminable monologues about trivia or things they already know, they plod through the contents of textbooks, or retreat to common places. There is hardly any likelihood that difficult specialist issues can be clarified, since it is easier to ignore supplementary questions.
However, there is no ‘best practice’ concerning the interaction structure in interviews with experts and elites. Different forms of knowledge and different functions of the interview within the research design make different interaction situations more preferable than others. Furthermore, the interaction structures can only partly be influenced by the interviewer. However, choosing the adequate strategy of asking will have a positive influence on productive data collection in conversations with experts and elites. To give just some hints:
Interviews with experts and elites are usually semi-structured interviews, conducted with at least a rough topic guide,4 which contains the central dimensions of the planned conversation. Exploratory interviews with experts and elites should be conducted as openly as possible, in order to make it possible to gather unexpected information and interpretations, which could have not been imagined when constructing the topic guide. The focus is on archiving initial breadth. Systematising interviews, in contrast, are based on a quite detailed topic guide, which entails a comprehensive catalogue of questions about facts. Characteristic for this interview is a permanent revision of the topic guide in the course of the research, according to the advancing state of information. Interviews aiming at interpretative knowledge, that is, theory-generating interviews, are using (thematically focused or problem-centred) narratives more extensively (see Witzel and Reiter, 2012), to give the interviewee space for presenting his or her orientations and beliefs.
The expert and elite interview has always had a certain thematic focus. The purpose of the interview is not to capture the interviewee as a ‘whole person’ (like some biographical interviews do) including as many facets as possible of the individual personality. Consequently, the topic guide usually includes topical or specialist questions and only some general question, for example, about the personal background. On the other hand, especially if the researchers are interested in substantively rich investigation of ‘interpretative knowledge', they should not cut off statements of the interviewees, which on first sight seem to be private and of no immediate interest for our research topic. It is only in the phase of evaluating data that it becomes clear whether the relevance structures and patterns of orientations used by the expert can be reconstructed exclusively by using his or her explanations given from within the professional context, or whether it is also necessary to incorporate comments made from the personal sphere. It is [Page 662]frequently the case that the very interview passages in which common places and pithy sayings from everyday life are mobilised, or arguments relying on metaphors from the ‘private sphere’ are put forward, prove to be of particular interest. In practice, one can hardly distinguish between the interviewee as holder of a social role or position and the interviewee as a ‘private person', and it makes no methodological sense to attempt to do so.
Multilingualism is a neglected problem in qualitative social research, including expert and elite interviews (see Resch and Enzenhofer, Chapter 9, this volume). This lack of attention is all the more surprising if we consider the accelerating pace of internationalisation in research, and comparative international research in particular, as well as the growing importance of international migration and multiculturalism in the social sciences. Researchers are thus increasingly likely to face methodological challenges arising from the use of different languages. The methodological consequences of multilingualism in social research affect all stages of the research process, from project design, to data collection and analysis to the presentation of results. This has major implications for quality assurance in the research process and for the quality of the findings. Conducting research in and with other languages ultimately confronts the researcher with fundamental hermeneutic and translatological questions, such as the possibility of understanding across languages, and the translatability of culture, that is, the possibility of understanding foreign cultures at all (Temple and Edwards, 2002; Inhetveen, 2012; Littig and Pöchhacker, 2014).
Beyond these fundamental methodological concerns, working with speakers of other languages also raises a number of practical questions like planning, budgeting, training etc. Consequently, if participants have different native languages, researchers often use English as a lingua franca. The use of international English is particularly common in political and business organisations. Accordingly both the respondent and the social scientist asking the questions are assumed to be proficient enough in English to conduct an interview. If this is not the case, the social researcher will usually make ad hoc arrangements for translation assistance, whether in the process of data collection or for transcription and analysis. Either option – ‘English only’ and ad hoc translation – is not without problems. Most crucially, there is a lack of established criteria by which one's own and other participants’ linguistic and communicative competences might be assessed, especially when it comes to determining whether language skills are sufficient to grasp finer points of communication, such as irony, loaded words and connotations.
Considering the increasing internationalisation of expert interviews and their frequent use in comparative research designs, Littig and Pöchhacker (2014) draw attention to this blind spot in the methodological literature. They suggested ‘socio-translational collaboration’ as a way of coming to grips with linguistic challenges in expert interviewing. A collaborative approach bringing together social researchers and professionals with translational competence rests on two basic premises, that is, a substantial degree of mutual knowledge of the respective conceptual frameworks and working methods, and consistent consideration of linguistic and cultural issues in all stages of the research process. Conducting expert interviews in an international context means that both the social researcher and the translator (in the wider sense) need to have expertise in either domain: the translator must also understand the basics of qualitative interviewing in [Page 663]general, and the expert interview setting, in particular; the social researcher must also be familiar with cross-cultural language issues and ways of resolving them with various techniques of translation and interpreting. The awareness of language issues among social researchers is still very limited, and the topic of multilingualism rarely features in courses on qualitative methods. Whereas the methodological literature to date reflects a pragmatic attitude towards problems of language and translation which manifests itself in the ad hoc recruitment of ‘native speakers’ and a narrow focus on the interview itself, the socio-translational collaborative approach calls for a proactive engagement with translational and cross-cultural issues from the very start of the research process and throughout all of its stages (for strategies see Enzenhofer and Resch 2011; and Resch and Enzenhofer, Chapter 9, this volume).
An important issue related to new technical means is interviewing via Skype or similar technical means. This has just become an option recently and thus not yet been reflected thoroughly in methodological terms (Deakin and Wakefield, 2014). However, reports on experiences with Skype are available (Salmons, 2010; Hanna, 2012; Weller, 2015). The gold standard of the interview situation certainly is the face-to-face interview. This has been stressed in the literature discussing telephone interviews with experts (Christmann, 2009). Compared to face-to-face interviews, telephone interviews allow for less control of the interview situation. The interviewer cannot know whether the interview partner is fully concentrated on the interview or distracted by other activities (answering emails or playing Internet games). The lack of non-verbal elements like eye contact is a disadvantage for the communicative situation: additional information, for example, about the engagement of the interviewee gets lost, the commitment to the interview might be reduced etc. To our experience interviews on the telephone tend to be shorter than face-to-face interviews. As a consequence this entails a loss of information. According to Christmann (2009) at least some of the disadvantages of telephone interviewing might be compensated by a clearly structured interview strategy (for detailed strategies, see Stephens, 2007). Especially for first informative or explorative interviews, the shortcomings of telephone interviews might be less severe. However, what might seem efficient (regarding time, travelling, budget) at first glance, could prove to be less effective with regard to the richness of information gained.
Interviews via Skype seem to be a better solution than telephone interviews if face-to-face interviews cannot be conducted due to time restrictions or budgetary reasons. As there is a virtual visual presence of the interview partners, the interview situation can be better controlled (on both sides). But side information about the interviewee's professional environment still gets lost. Given the target groups of expert and elite interviews it can be assumed that many of them are familiar with using recent technical devices; the younger generation more than the older. Thus Skype interviews will likely become more important in the future. For the time being, however, the most preferable interview situation is still face to face, though its advantages might be heightened (Weller, 2015).
In our times of the ongoing process of digitisation, it can be difficult to keep the identity of experts hidden. Consequently, the researcher should always clarify how much of the information obtained from the interviewee can be published in a non-anonymous form (e.g. as quotations). In some cases, it may be necessary to have the interviewee expressly authorise the use of the minutes or interview transcript for analysis or publication purposes. The expert should, in all cases, be given the assurance that all data will be [Page 664]treated in confidence. Issues of anonymity have become even more severe through the general use and availability of the Internet. This makes experts even more easily identifiable, ultimately worldwide. Consequently, expert interviews should not be used, or if so, very carefully, as demonstration material in courses or lectures. Quotes might unintentionally spread far and wide via the Internet or social media and could endanger the integrity of the interviewee.
It has been pointed out that expert and elite interviews have become diverse in the last few years. There is not just one way of doing expert or elite interviews. This holds also true for the analysis, be it with or without computer assistance. There is no standard procedure for analysing expert interviews. In principle, all qualitative social research analysis methods can be used, for example, the code-based procedures common in grounded theory or qualitative content analysis, or the sequential analyses applied in the hermeneutic sociology of knowledge or objective hermeneutics. A combination of different methods is also admissible.
The diversity of expert and elite interviewing provokes the question whether a methodological canonisation of these forms of interviewing is in sight. We would cautiously answer this question with no. The unfolding of the methodological debate has led to the refinement of the methodology, and thus to a greater variety of the overall aims and applications of the expert interview, the interview strategies and the consideration of general issues. This can be interpreted as beneficial for the plurality of methods and methodologies. The most suitable form of expert or elite interviews ultimately depends on the actual research project, its goals and the particular research questions at stake.
1. On the distinction in research logic between ‘contextual knowledge’ and ‘operational knowledge', see Meuser and Nagel (2009).
2. Gender relations can play a twofold role in the interview. First, most of the experts are men, as there are relatively few women in management positions. Second, the probability of the participants ‘doing gender’ (i.e. assuming gender-specific roles, particularly in a mixed-gender setting) becomes highly likely. Doing gender can also become manifest in the content of the interview, for example, by using particular gendered metaphors (‘the firm as the mother of its employees’ etc.).
3. For ordinary in-depth interviews one basic rule is to avoid the ostentatious presentation of knowledge and expertise: ‘Researchers need a degree of humility, the ability to be recipients of the participant's wisdom without needing to compete by demonstration of their own’ (Legard et al., 2003, p. 143).
4. See for the design of a topic guide Arthur and Nazroo (2003); for topic guides in expert interviews see Gläser and Laudel (2004, p. 59ff).