Lothar Mikos & Claudia Wegener (Eds.) Qualitative Medienforschung. Ein Handbuch [Qualitative Media Research. A Handbook]. Konstanz: UVK, 615 pages, hardcover, ISBN: 3-8252-8314-3, Euro 34,90

Abstract: This useful handbook introduces the present state of qualitative methodology in German media research. While its formal focus is on media, it covers many aspects which are of general interest. The chapters are short and provide useful examples of the application of different methods. The book proceeds from theory via data collection to analysis and meta-methodological reflections. In some chapters, it is not always clear whether data collection or analysis is the focus. The book also indicates three desiderata of qualitative methodology today: 1. the logics of sampling are often neglected; 2. the process of analysis beyond structuring the material is not sufficiently clear; and 3. comprehensive criteria for the validity of qualitative research are still lacking.

Key words: data analysis, data collection, epistemology, interpretation, qualitative media research, sampling, validity

This is a useful book with concise information on major approaches to qualitative media research. The handbook has an introductory character. The chapters are brief; they usually introduce a wider perspective and close with an example on how to apply a method. The range of approaches is wide. The authors are well known experts who have long standing experience in the field of media research and many of them have been influential in developing qualitative research in Germany, too. On this basis, the book provides useful and reliable information. It avoids jargon and is easily accessible. References at the end of each chapter, together with a list of literature and Internet links in the appendix, enable the reader to engage more profoundly with an approach, if required. Contents, price and hardcover quality make this book a useful purchase for any student of social sciences. [1]

The handbook documents a third phase of qualitative research in Germany. Early qualitative studies in psychoanalysis, the Marienthal or the Frankfurt school studies were driven by an interest to closely investigate cases with several methods in order to understand as much of an issue as possible. From the late 1960s onwards, intellectual effort was made to establish, justify and elaborate on a qualitative paradigm next to the dominant stream of research in the style of American behavioralism. Since the 1990s, qualitative research has established itself as a field in its own, not only through methodological reflection, but also through widespread application of its methods. The current book is an expression of this development. [2]

Its 55 articles (see table of contents in the Appendix) are mainly of high quality. The perspective is clearly on introducing approaches and demonstrating what they have to offer in methodological terms—this is a strength of the book. For a deeper understanding, one has to consult the literature that is provided in the list of references. Actually, a more appropriate subtitle would have referred to this book as an introductory handbook. [3]

Debates on qualitative versus quantitative research have reached their peak and today qualitative research is accepted as a normal set of approaches among others. Media research has long been dominated by quantitative methods and contributed much to their development in the whole field of social sciences. Today, sociology, education and also, in part, communication science (several authors of the book have been engaged in the introduction of cultural studies in Germany) make wide use of qualitative methods. In general, it is not always clear what is meant by qualitative research. Often it is associated with data gathering by ethnography and open interviewing as opposed to standardized surveys. It is important to note that qualitative methodology consists of data collection and data analysis. There has been much development in general methodology, strongly connected to epistemological considerations on how it is possible to perceive reality. Methods of data collection are also very developed; most difficulties are connected to rules for qualitative data analysis. [4]

The handbook goes beyond a distinction between theories, data collection and data analysis. It broadly introduces the theoretical background (Part 1) by presenting a number of approaches, some of them particularly connected to media research, others of a more general nature (theories of action, sociology of knowledge, case study method). It then turns to research design (Part 2) and presents twelve methods of data collection (Part 3). Before introducing methods of analysis (Part 5), it clarifies how data should be prepared for analysis (Part 4). A brief conclusion (Part 6) discusses the presentation of results and the important issue of validity. [5]

It is not always clear why some chapters are situated in certain parts of the book. One the one hand, content analysis is presented as more of an issue of research design than as a method for analyzing data, while role playing, on the other hand, is in fact a method of data collection, not a method of analysis as it appears in the volume. The part on data analysis mixes distinct approaches to analysis and chapters on particular media such as chats, films, television, photography, video games, film music, music videos, and cross media. These field specific chapters refer to several methods of analysis. For example the analysis of role play applies deep hermeneutics. Film, television and photography analysis uses iconography. Chat and forum analysis uses content analysis. [6]

Media research is not only a field by itself, but also functions as a central tool in disciplines like sociology, education and political science. Many aspects of political science have been analyzed by using media as data sources to get a handle on the world. We are, therefore, dealing with a key science that has much more to contribute than enlightenment on the state of media in society. [7]

The media, usually understood as mass media, is a multifaceted notion. VOLLBRECHT introduces this field in the only general article. He distinguishes material (air, paper), communicative (speech, pictures), technical (microphone, camera) media and media as institutions (a particular newspaper or the broadcasting system), and then presents a number of media concepts. These refer to elements such as sender, receiver, channel and code which together constitute the broadcasting system. Another perspective looks at media as organizations, institutions and social systems. Mass media is characterized by a dispersed audience which can, in a critical perspective, be distinguished from the sphere of producers and the products that are transmitted. Finally, media didactics and media pedagogies are mentioned. Unfortunately, this chapter does not deal with the relationship between media and politics or society; media effects or new media such as the Internet or cell phone-based media are not touched upon either. Thus, the only specific chapter on media remains largely general and does not tell much about critical questions that arise in the context of mass media. [8]

Since qualitative media research is closely connected to theories of action, these are covered in the following chapter. In an informative article, KROTZ limits the perspective to those action theories that refer to meaning as it is produced in action, experience, thinking and communication. The general aim of qualitative research is to understand the constitution of meaning by the subjects under investigation. For example, symbolic interactionism asks how meaning is constructed in action; cultural studies analyze how readers produce texts in the act of reception; phenomenology looks for objective meaning in subjective acts; and structural anthropology tries to understand action through close observation and description. These approaches have been influential in German media studies by introducing the issues of perspective, situation and identity into the field. KROTZ notes that the main difficulty with these theories is to draw conclusions from action to underlying structures. Other articles that deal with specific media—like films, video games or music to mention only a few—are located in Part 5, dealing with analysis. [9]

Part 1 further presents a number of theories that deal with media from different perspectives. In the chapter on structure-analytical reception research, NEUMANN-BRAUN argues that usage of media usually starts with an active decision and includes a relationship to others that does not stop when the act of reception is finished. It continues through a phase of appropriation (Aneignung) that connects the message to the daily life world of the individual. In a similar vein, BACHMAIR notes that media develop into a frame of reference for communication and understanding among individuals and, as such, are constitutive for everyday life. He refers to social situations that are based on media, like media events or collecting Pokemon cards, that provide symbolic material for exchange. Qualitative research is appropriate for these type of investigations because they try to reconstruct the meaning that interacting with media provides for individuals. [10]

This is also obvious in media biographic research. Remembering the reception of media products not only supports the construction of identities, but also enables individuals to express what cannot be said verbally. Examples are referencing favorite hits of teenagers and other phases in the life cycle (SANDER & LANGE). The strong connection between media and daily life can also be found in media ecology (GANGUIN & SANDER) and in the chapter by MIKOS that explicitly deals with the role of media in daily life. [11]

Theory establishes the connection between the observed and general conclusions derived from it. The research process also rests on practical issues. WEGENER and MIKOS give advice on how to design a study while the rest of Part 2 introduces media production (KÜBLER), reception (PROMMER & MIKOS), content analysis (WEGENER), triangulation (TREUMANN), comparative cultural analysis (VOLKMER) and case studies (BAUR & LAMNEK) as options from which to choose. Part 4 covers other practical issues such as writing protocols, transcriptions of interviews, describing the nature of the data and coding the material. The most central one is the chapter on coding. Coding aims at identifying larger units of meaning. Often, (although not necessarily), it reduces the material to a manageable amount. The chapter provides for a good example of how a coding process can look like in practice, but does not consider more intricate issues in coding. [12]

Some advice can be obtained by consulting the chapter on grounded theory which is among the best of the book. In ten pages, LAMPERT manages to present the history and basic idea of grounded theory, introduces its basic concepts and procedures, without omitting (in contrast to other chapters) critical remarks on its limitations. Central aspects are simultaneous data collection, coding and analysis by ongoing comparison. The criterion of theoretical saturation is helpful for finishing the process of data collection and analyses; however, in practice it is hard to reach this point, in particular in funded research with a fixed time frame and budget. Open coding, axial coding and selective coding are introduced as ways to proceed from the utterances of objects of research to a central category that allows the researcher to produce a hypothesis on the topic under investigation. Many elements of grounded theory have been integrated into other qualitative approaches. Beyond describing the central role of grounded theory for the development of qualitative procedures, the chapter falls short of anything specific about its application in connection to media. This is also the case with other chapters in the book. [13]

In Part 3 the handbook presents a number of methods for data collection. Among the most important are different forms of interviewing (qualitative, narrative, expert, group, online). The initial chapter offers a limited perspective on qualitative interviewing. It introduces openness, flexibility, communication, explication, process orientation and reflexivity as basic characteristics (KEUNECKE) and then presents the narrative, problem-centered, focused and expert interview. Unfortunately, KEUNECKE only uses secondary sources and does not present the problem-centered interview according to WITZEL (2000) who developed this method. This results in a description that contains elements that are more characteristic of the narrative interview (constraints of condensing and detailing). The article offers useful advice for conducting interviews such as how to deal with recording, how to stimulate narration and which style of language to choose. It would have been useful to present further ideas such as asking for details instead of accepting seemingly clear answers, avoiding interview guide bureaucracy (HOPF 1978), how to take notes while keeping up communication and, probably most important, that there is no such thing as a perfect interview. It is not by chance that DEXTER (1970, p. xiii), author of one of the most influential interview guides, states that the experience of numerous interviews has taught him most of what he knows about the issue. In a similar vein, HERMANOWICZ' strategy number 25 reads: "Practice, practice, practice" (2002, p.497). [14]

Three chapters on content analysis cover its procedure (MAYRING & HURST), the application of computers (KUCKARTZ) and its general position in media studies (WEGENER). The latter distinguishes primary content analysis that is applied to media products like newspapers or television programs from a secondary use as a method to analyze transcriptions of interviews. This distinction, however, has no further consequences for the application of the method. Content analysis can be used not only to describe the structure of a text, but also to make inferences about the author (communicator) or the recipient (effects). The latter is common but risky and requires support through comparison with other studies. Finally, content analysis is used to make inferences about the state of society at large. [15]

There are two chapters on the case study method. The chapter by NEUSS largely draws on basics of qualitative methodology and justifies the case study as a method to detect experiences and results close to reality. BAUR and LAMNEK discuss different understandings of what possibly constitutes a case. Independent from selecting a person, a group, an institution or a process as a case, detailed documentation of interaction and the detection of a plausible plot are fundamental for the analysis. This proceeds by comparison, either between cases, between case and theory or by varying variables within the case. [16]

Many approaches seem to be compatible with one another and are frequently combined. Methods for visual data often also use content analysis. Discourse analysis explicitly puts itself in juxtaposition to content analysis, arguing that the latter can describe the contents of texts but fails to explain the generation of meaning of the very content. Discourse analysis, itself a universe of approaches, deals with "naturally occurring talk," i.e., practices of speaking with regard to implicit rules and principles of classification that limit the range of what we can know and communicate in society. DIAZ-BONE presents a Foucauldian version of discourse analysis. After having selected relevant material, it tries to identify repeated topics, analyzes their occurrence and finally identifies underlying schemata. The crucial question is whether the identified order of knowledge enables or limits other non-discursive practices. The example of two music genres demonstrates how different meanings are ascribed to similar practices through discourses (see also DIAZ-BONE, 2005). [17]

One of the few methods that provided very early rules is objective hermeneutics. The central idea is that agency reflects latent structures which can be reconstructed through successive narrowing of possible meanings embodied in action. It remains contested whether it is possible to seriously reconstruct objective structures from a case structure. HAGEDORN demonstrates how ways of reading a text are increasingly narrowed through sequential analysis. WINTER presents interpretative ethnography, an approach that tries to establish a dialogical relationship with the subjects of research and enhances the understanding of the other by intensive reflection on the researchers own experiences during field work. The researcher is a supporting participant. S/he tries to develop sensitivity for the strange world of the other that s/he can only understand in contrast and comparison to her/his own life world. Auto-ethnography enables her/him to grasp differences between her/his own and the subject's perceptions. [18]

An innovative approach that surely is going to gain significance in the future is qualitative online interviewing. EHLERS presents a taxonomy of methods. Synchronicity is characteristic for chat interviews, audio and video conferences and collaborative platforms such as the use of a whiteboard where participants' writings are simultaneously visible for others. Other methods are interviewing by e-mail, the use of discussion forums and so called Wiki platforms for unstructured discussion via the Internet. These enable researchers to reach subjects without traveling and they allow researchers to save utterances for documentation. The main difficulties are related to the setting. On the assumption that subjective meaning is connected to the situation and context of an utterance, online research has to explicate the context by using intensive verbalization. Participants also have to be familiar with the technologies used, and the methods require an intensive introductory phase to assure technical functioning and create an atmosphere for discussion. [19]

DICHANZ briefly introduces the Delphi method as a means for forecasting. Repeated waves of expert surveying are used to identify challenges and likely developments of an issue. He gives the example of the future role of school books in connection to Internet media. Video production by teenagers (WITZKE) and children's drawings (NEUSS) can be used in youth research. These methods are prone to articulate feelings and views beyond the use of verbal language. Both methods are used in combination with verbalization. Role playing (STAHLKE) and scenic play (TILEMANN) are other methods to capture unconscious and emotional states better than interviews, since the latter only allow for a retrospective view. These methods of data collection produce rich material and have to be complemented by techniques of interpretation. Documented by video, they have to be used in connection with methods of film/video analysis. STAHLKE proposes to apply deep hermeneutics (Tiefenhermeneutik). Unfortunately, her article does not really demonstrate how this should yield the results she claims. The conclusions are fairly general and, as such, can be derived through any act of interpretation that compares sequences of and assumptions made in the material with life world knowledge. This problem is true for other methods as well. Partly this may be explained by lack of space, as these articles are rather short, but it is also related to the problem of interpretation as presented below. [20]

Another short article impressingly demonstrates how music in films and videos can be analyzed. BULLERJAHN presents a number of ways in which music appears in the media. The analysis she proposes is partly a technical one which requires profound musical competencies, thus goes beyond social sciences. She turns to musical styles, sequences, sound, instruments used, melody and interpretation of the song, technical effects, and visual presentation. Functions range from economic support for a film to persuasive stimulation of emotions and identification. [21]

Thinking aloud (also called protocol analysis) can be used to document mental processes that accompany an activity. BILANDZIC distinguishes between thinking aloud while an activity is performed, e.g., in a factory, from verbalizing afterwards what was on one's mind. The latter has been used to investigate zapping in front of the TV screen. A video sequence has been recorded while watching TV and afterwards this video is commented by the recipient. The central difficulty is letting the stream of thoughts flow without controlling it. The method seems to be a contradiction in itself since thinking is fundamentally different from speaking. But the protocol documented in the article is promising. Although many utterances are very general ("this was interesting") it might be useful to identify less conscious aspects of behavior that are hidden and cannot be identified only by interviewing. [22]

Altogether there are a number of approaches which are less common that deserve attention. Often they should be used in combination with other methods. Mixed methodology, therefore, does not only concern qualitative and quantitative designs, but also different methods within each paradigm. One has to remember though, that it is easy to argue for a combination of methods. In research practice the application of unconventional methods poses some difficulties. Combining them with conventional methods requires additional resources, using them as a single method is often risky because of insufficient experience with the potential of such methods. [23]

One of the central problems of qualitative methodology has been its difficulty to provide for rules of procedure that not only allow for intersubjectivity, but also make it possible to learn a method in the first place. The problem is not central in data collection; the crucial issue is data analysis. How do we proceed from the description of the observed to a judgment about the meaning of the data? A "constructivism" that refuses to speak of ultimate meanings only relocates the problem; the researcher has to make a judgment. S/he has to offer interpretations of reality in a plausible way. Talking of preliminary conclusions is basically a rhetorical turn, rather than a way out of the difficulty to justify how these relate to the outside world. Of course, our conclusions are the best we can offer at a present stage—but research projects are usually completed with the presentation of a final report. The rest is up to history. [24]

A common claim advanced by qualitative methodology refers to its ability of capturing reality of social processes better than quantitative approaches, because they are viewed from the subject’s side. This claim somewhat contradicts the epistemological basis of qualitative research which, according to FLICK, does not aim at producing a realistic picture of society (Abbild) but aims at reconstructing meaning as subjects produce it in interaction. [25]

Despite differences in epistemological positions between different approaches, tangible analytical operations applied by them are often fairly similar. The analysis of video games (EICHNER) distinguishes between the game world, the player figure, modality of action, design of the game and the real-life context in which a game is embedded (online, LAN-party). Self-produced videos are analyzed according to length, pictures, camera activity, effects and tone in each take (WITZKE). The analysis of chats and Internet-forums proposes to categorize the material according to topics (SCHWEGRAF & MEIER). The central technique is usually structuring the material after specific aspects. [26]

Structuring the material is also the core operation of qualitative content analysis (MAYRING & HURST). The results of a descriptive content analysis are categories that inform about the structure. A further step would be counting of frequencies, i.e., the transformation of qualitative into quantitative analysis. Also the interpretation of pictures in the frame of the sociology of knowledge proceeds by a detailed description of elements of the picture (REICHERTZ). In the process of description, extra-empirical theorems, e.g., from iconology, are applied in order to transfer ideas rooted in cultural history to concrete empirical material. All these approaches present only slightly different ways of structuring the material. [27]

FLICK argues for triangulation as a way to improve the quality of research. Different methods can be applied to support conclusions when findings are similar. They can also be used complementarily to shed light on different aspects of an issue or their results can contradict each other, thus challenging the theoretical understanding of an issue. Often different approaches are combined in a hierarchical fashion, meaning that a central method is accompanied by a method of lesser importance. Whether quantitative approaches, which see data as a pure representation of reality, or qualitative approaches, which understand reality as constructed, can be combined is an open question of epistemology. [28]

REICHERTZ criticizes three common strategies to legitimize findings. The first rests on the charisma of the researcher who, in the fashion of a genius, has developed a "great idea." Quite provocatively, REICHERTZ claims that this strategy is usually chosen by those who claim objectivity in their findings, but who fail to demonstrate how they arrived at their conclusions through the research process. A second strategy turns to procedures. Phenomenologists claim knowledge of the things "as such," multi-perpectivists try to understand things better by looking at them from different angles while deconstructivists raise awareness for perspectivity in every claim for reality. A third strategy rests on the ability of scientific discourse to identify valid results. This approach is also problematic because discourse is a social process that does not necessarily function according to scientific criteria alone. Peer review as the best of all bad methods would be a case in point. According to REICHERTZ, better research can be identified when it uses naturally occurring data, continues sampling until saturation is achieved, validates findings by applying them to texts in order to test the fit of interpretations and involves colleagues in interpretation in order to expose findings to further scientific discourse. [29]

FLICK closes the book by going beyond procedural advice. He claims that qualitative research should look for coherence and validity as its main criteria of goodness. He regrets the lack of criteria for the preference of one method over another. Unfortunately, the present book does not supply criteria either. The criteria that he does propose are fairly conventional (appropriateness in terms of issue, population, question and level of established knowledge). If ever this handbook sees a second edition, the editors should ask their contributors to reflect on these criteria and include their answers into every chapter! [30]

FLICK further calls for quality management of the research process which covers steps such as reflecting on the goals of research and application of methods, responsibilities and standards. This implies that goodness is an issue where agreement has to be reached among scientists in each single case. Standardization as a way to control for context variation is usually not applicable to qualitative methods because they explicitly aim at capturing context and subjective relations. Valid interview data can be achieved through an interview that allows the interviewee to express himself without feeling coerced to communicate strategically. This includes a trusting relationship between researcher and subject. This is convincing with regard to data collection, but what about analysis? The central idea is to provide for a transparent research process at all stages. Such a general proposal is surely justified and protects the researcher from drawing conclusions he might personally like when they are not supported by his material. It does not, however, enable him to judge beforehand whether his work can satisfy an external scientific community. [31]

Given that the handbook is not a contribution to academic debate, but aims at familiarizing researchers with the potential of qualitative media research in the first place, this reader would have liked a more consistent conclusion. Pointing at the openness of debate on validity and generalization does not carry far. It would have been more useful if the distinguished author of a number of textbooks and edited volumes on qualitative methodology had presented his personal view of useful criteria for intersubjective research. If senior scientists are not able or do not dare to choose among the variety of proposals for useful criteria, how could students be expected to make their way through the fog? [32]

From a broader point of view, a handbook devoted to qualitative media studies raises questions about in which ways the qualitative is integrated into the field of media research. We are dealing with an interdisciplinary field that spreads across disciplines such as sociology, political science, education, psychology, literary studies and even economics. Thus, it is not surprising that media studies are often not specific about media. Although there are qualitative "milestone" studies (JENSEN, 2002, pp.157-160), media often serve as an indicators for larger societal issues in other disciplines. It is common that, for example, discourse analysis uses newspaper articles, television dialogues or even pictures to analyze general issues such as racism or innovation ideologies. In the same vein, analysis of media content is used to make inferences about general societal problems. The use of language in newspapers has been an issue for gender studies, not with the aim to analyze the media system, but rather gender relations. This is possible because media has become an integral part of our lives, as the catch phrases media or information society suggest. [33]

Today the qualitative media perspective has a place in mass communication theory. McQUAIL (2000, pp.12-13) describes three essential approaches to media studies. These are the structural approach dealing with the societal system of the mass media, the behavioral approach focusing on individual behavior in connection with media reception, and the cultural perspective centered around language and meaning. At least the last is closely connected to qualitative methods. A challenge to the traditional transmitter model also came from constructivism which called the existence of a unified reality into question and, consequently, called for attention to the production and interpretation of reality by, and in interaction with, mass media. Thus, the media is seen not only as a transmitter of messages but also as a field for producing and exchanging meaning. Connected to the cultural view is an interest in hegemonic structures. This goes along with increased attention for marginalized groups in the public sphere such as women, the poor, non-whites or sexual minorities. [34]

Qualitative media studies have come from outside of the discipline. Parallel to surveys that accompanied the introduction of radio and television, thinkers of humanistic origin like BENJAMIN or HORKHEIMER and ADORNO, later BARTHES and ENZENSBERGER, have influenced media research. While their contributions were mainly theoretical, behavioralist research dominated the empirical search for media effects. The limitations of the simplistic stimulus-response model eventually urged media research to look for micro practices in the uses of media as well. Reception analysis that concentrated on daily life arrangements and the decoding of messages is where qualitative aspects most obviously come into play. The humanistic tradition pays attention to language and other symbols. This, together with discursive studies, is another domain of qualitative approaches. [35]

A distinct qualitative methodology was only applied from the 1980s onwards, strongly influenced by cultural studies. By the early 1990s a critical mass of journal articles and textbooks indicated a "qualitative turn" in mass communication studies (JANKOWSKI & WESTER, 1991, pp.71-72). This notion, however, may seem slightly exaggerated because the quantitative paradigm never stopped growing. It was paralleled by an increase in studies based on interpretive designs. Consequently, McQUAIL still today speaks of a dominant and an alternative paradigm (2000, pp.51-52). [36]

More than half of the authors in "Qualitative Media Research" are located in media and communication departments. Another indicator of an institutionalization of the qualitative perspective is provided by the member survey of the German Communication Association (DGPuK). About half of those who participated in the survey consider the humanistic and cultural studies perspective to be of importance (PEISER, HASTALL & DONSBACH, 2003, pp.320-328). There is even a significant number of researchers who actually use qualitative methods; their number is only slightly smaller than those who apply quantitative methods, and many, in fact, do both. In contrast to the large amount of researchers with a humanistic-qualitative profile, however, publications of this sort are marginal in the association's journal, "Publizistik." The same is true for the second important journal (Medien und Kommunikation, published by the Hans-Bredow-Institut). In both journals, the vast majority of the more than 100 articles that have appeared since 2003 are based on a quantitative design. Qualitative methods are used in less than ten percent of all articles. In fact this number is even less than during the period 1989-1991. It seems that qualitative media research is practiced widely and has developed its own community. Although it overlaps with the mass communication research community, qualitative media research remains marginal within the institutional core of media and communication research. [37]

The same trajectory applies to qualitative media research generally as to other research processes. Research question, literature review and the choice of theory are the first steps taken by a researcher sketching the design of a study and a sampling strategy. A third step is data gathering and the preparation of the material for analysis. Phase four engages the researcher in data analysis followed by drawing conclusions and then reporting of findings. This plan may be modified by loops creating shortcuts between data collection and analysis, re-formulation of the research problem and engagement with theory at different stages. Because of their general significance, three neglected issues in MIKOS and WEGENER's handbook, namely sampling, the explication of strategies for analysis and criteria of goodness should be mentioned here. [38]

Formulating a research design involves questions, concepts and a sampling strategy. This is the stage where a researcher decides whether to conduct a case study, a survey, an experiment and whether to use triangulation, a comparative design or any combination of these. Since qualitative procedures usually do not rely on random sampling, the selection of cases can be done in different ways. Criteria can be theoretical considerations, extreme, typical, similar or critical cases as well as maximum variation or even convenience sampling. In most approaches the sampling strategy is considered to have an impact on the generalizability of the findings, but too often this does not receive the attention that it deserves. Unfortunately, MIKOS and WEGENER's handbook is no exception. Some argue that qualitative analysis only ought to be interested in understanding one particular case in depth. This is the model of psychoanalysis or cultural anthropology. But the popularity of qualitative methods cannot be explained by a view that declines the issue of generalization. The common strategy is to generalize findings to settings of similar context conditions (SCHOFIELD, 2002, p.178). [39]

Selecting cases and making contacts is not only important in terms of theory, but also time consuming. It is so central to the research process that it deserves explicit treatment. A common solution is multi-step sampling. The selection of cases is often connected to contacts, particularly in media studies. Although media is a broad phenomenon, in many instances sampling faces the difficulty of finding individuals who have been exposed to a particular media product. This differs from educational research, for example, where the target population is usually connected to an institution; it is often challenging to find interviewees suitable for research on particular radio programs or video games. Snowball sampling is the most feasible option but it typically remains locked in into networks of similar cultures or life styles, since patterns of media use are strongly connected to social structure and, as such, one factor that constitutes social milieus (SCHULZE, 1993). [40]

There are basically three choices in data collection. Direct collection of (verbal) data occurs in different forms of interviews such as the semi-structured, expert, online, group and narrative interview. Also the Delphi method and thinking aloud fall into this category. A second option is to carry out an observation. This will typically result in a protocol that consists of textual data. A third choice is to collect artifacts. This is important in media research where films, photos, videos and also drawings, video games and even role playing can constitute data. Again, these usually have to be transformed into textual data by means of sequence protocols, transcripts or detailed descriptions of the interaction. Thus, according to ALTHEIDE (1996, p.2), qualitative media analysis is always document analysis. It deals with "symbolic representation[s] that can be recorded or retrieved for analysis." Although this view enlarges the notion of the document beyond recognition, it rightly points to the fact that even in qualitative approaches, it is usually not sufficient to provide "only" for detailed descriptions. At some point one has to establish the meaning embedded in the material. [41]

For the crucial stage of data analysis, qualitative research does not rely on correlation coefficients that indicate significance within data. It aims at meaningful interpretations. MIKOS and WEGENER introduce options like conversation analysis, qualitative or computer aided content analysis, grounded theory, objective hermeneutics, discourse analysis and interpretative ethnology. One might add rhetorical and semiotic analysis. They also include a number of media specific approaches (chat, film, games, music, cross media) although these are not original methods of analysis. Ultimately, data is transformed into a textual form to make it available for analysis. Qualitative media research, however, sometimes allows analysis to begin directly, without, for example, transforming photos into a text. The analysis then treats photos as documents. [42]

The examples given above demonstrate that the crucial operation of ascribing meaning (what means what) has not been sufficiently explicated as of yet. Even after decades of reflection, the act of interpretation (which is at the heart of qualitative methodology) remains somewhat opaque. When it comes to describing how to proceed while conducting an interpretation, other influential authors of the qualitative tradition also resort to very basic techniques. The core of interpretation, according to DENZIN (2002, pp.349-350), is "bracketing the phenomenon" –by which he refers to "reducing it to its essential elements and cutting it loose from the natural world so that its essential structures and features may be uncovered." This is followed by constructing the phenomenon in the sense of "putting the phenomenon back together in terms of its essential parts, pieces and structures." HAMMERSLEY and ATKINSON (1995, pp.209-218) see generating concepts and developing typologies as central operations of data analysis. KVALE (1996, p.190) suggests that "The analysis proper involves developing the meanings of the interviews, bringing the subjects' own understanding into the light as well as providing new perspectives from the researcher on the phenomena." KVALE proposes five approaches to analysis: condensation, categorization, narrative structuring, interpretation and an ad-hoc mixture of all of them. MILES and HUBERMAN (1994, p.245) provide perhaps the most extensive list of tactics for generating meaning out of the material, among which are noting patterns, clustering, making contrasts and comparisons as well as subsuming particulars under the general. [43]

In other words, the central operations in analysis and interpretation are connected to structuring the material in order to reduce it to its essential elements. Mostly this refers to different ways of structuring and compressing the material into larger units of meaning. In many cases this resembles the operation of summarizing the manifest contents. The "analysis" of meaning is often left to the individual creativity of the researcher. It rests heavily on her/his theoretical assumptions and context information connected to the situation from which the material is extracted. [44]

Another reoccurring issue in qualitative research is the debate about criteria. In a field that has been long dominated by the positivist paradigm this is not easy to solve. One reason for the difficulty to reach professional standards of quality is that most of qualitative research is conducted outside institutional funding. In his contribution, REICHERTZ argues that today there is not too little, but too much qualitative research. Data is not collected systematically, discussion of the nature of the material is omitted, methods of analysis are chosen without clear criteria and single cases are presented without sufficient justification as ideal types. A similar criticism was raised more than a decade ago (HOPF & MÜLLER, 1994). This ongoing failure is not due to lacking information. Since then, a number of good textbooks have been published and, not least, FQS has contributed to an intensive coverage of qualitative issues. The problem reflects the lack of agreed criteria and terminology amongst the qualitative community. Similar procedures appear under different names, some applying quantitative social research criteria, while others rejecting their suitability. The challenge remains. It seems that the important epistemological debate, originally introduced in order to justify the position of qualitative research amidst the dominant quantitative paradigm, has turned against the interpretive scientific community itself. It has led to a multiplicity of theoretical justifications, concepts and applications that are sometimes difficult to distinguish from "empty theoretical phraseology" (HOPF & MÜLLER, 1994, p.71), limiting understanding between different schools of thought. [45]

Today, methodology is highly sophisticated in terms of theoretical reflection. Much research practice, however, is guided by ad-hoc procedures that combine elements of different approaches, not necessarily in a reflective way. Sampling procedures, data collection and analysis are often not justified by the research question or the nature of the material, but by the theoretical tradition on which a researcher embarks. It would be beneficial to point out more explicitly what the potentials and limitations of chosen procedures are and how they are compatible with the nature of any given research question. Qualitative media research uses methods like conversation analysis, objective hermeneutics, discourse analysis, ethnomethodology, and rhetorical analysis, all of which have been developed by other disciplines. Nonetheless, it also has something to offer to methodology more generally. In particular, qualitative media research informs the wider research community about forthcoming approaches to social reality that will influence research practice. One example is data collection through online interviewing. [46]

"Qualitative Medienforschung" enables students to quickly grasp the nature of different approaches and can be used as a handbook throughout their studies. It is even useful for researchers who specialize in quantitative methods, because it informs the reader about the potential of complementary qualitative approaches in an accessible way. This review has mentioned a number of shortcomings, but part of the criticism is for the qualitative research community in general. In particular, the act of interpretation in data analysis and the criteria for validity and generalization are not transparent and clear enough, yet. Although a handbook promises to present the state of the art in its field, it is unrealistic to really expect the full range of aspects to be covered. The editors of the present handbook have managed to compile very accessible articles that usually avoid overweight theoretical outlines. Brief systematic descriptions are complemented by examples that enhance the understanding of the methods presented. Despite the criticism of certain details, the book has accomplished its aim of presenting a well grounded introduction to qualitative media research that presents theoretical foundations, central fields of application and detailed advice for practical research. [47]

Dexter, Lewis Anthony (1970). Elite and specialized interviewing. Evanston, IL: Northwestern University Press.

Diaz-Bone, Rainer (2005, October). Zur Methodologisierung der Foucaultschen Diskursanalyse [48 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research [On-line Journal], 7(1), Art. 6. Available at: http://www.qualitative-research.net/fqs-texte/1-06/06-1-6-d.htm [Date of access: November 03.06].

Hammersley, Martyn & Atkinson, Paul (1995). Ethnography. Principles in practice. London: Routledge.

Hermanowicz, Joseph C. (2002). The great interview: 25 strategies for studying people in bed. Qualitative Sociology, 25(4), 479-499.

Hopf, Christel (1978). Die Pseudo-Exploration – Überlegungen zur Technik qualitativer Interviews in der Sozialforschung. Zeitschrift für Soziologie, 7(2), 97-115.

Hopf, Christel & Müller, Walter (1994). On the development of empirical social research in the Federal Republic of Germany. Soziologie, Special Edition 3, 52-80.

Jankowski, Nicolas W. & Wester, Fred (1991). The qualitative tradition in social science inquiry: Contributions to mass communication research. In Klaus Bruhn Jensen & Nicolas W. Jankowski (Eds.), A handbook of qualitative methodologies for mass communication research (pp.44-74). London: Routledge.

Jensen, Klaus Bruhn (2002). Media reception. Qualitative traditions. In Klaus Bruhn Jensen (Ed.), A handbook of media and communication research. Qualitative and quantitative methodologies (pp.156-170). London: Routledge.

Kvale, Steinar (1996). InterViews. An introduction to qualitative research interviewing. London: Sage.

McQuail, Denis (2000). McQuail's mass communication theory (4th ed.). London: Sage.

Miles, Matthew B. & Huberman, Michael A. (1994). Qualitative data analysis. An expanded sourcebook. (2nd ed.). London: Sage.

Peiser, Wolfram; Hastall, Matthias & Donsbach, Wolfgang (2003). Zur Lage der Kommunikationswissenschaft und ihrer Fachgesellschaft. Ergebnisse der DGPuK-Mitgliederbefragung 2003. Publizistik, 48(3), 310-339.

Schofield, Janet Ward (2002). Increasing the generalizability of qualitative research. In A. Micheal Huberman & Matthiew B. Miles (Eds.), The qualitative researcher's companion (pp.171-203). London: Sage.

Schulze, Gerhard (1993). Die Erlebnisgesellschaft. Kultursoziologie der Gegenwart (4th ed.) Frankfurt/M.: Campus.

Peter HILGER, Diplom-Sozialwissenschaftler, PhD Student at the University of Helsinki, Department of Political Science. Research on citizen media, research cooperation and the politics on civic engagement. Course lecturer at the Department of Political Science, University of Helsinki.

Department of Political Science
University of Helsinki
Itämerenkatu 8 C 59, FI-00180 Helsinki
Finland

Hilger, Peter (2007). In Search for Criteria: The State of Qualitative Media Research. Review Essay: Lothar Mikos & Claudia Wegener (Hrsg.) Qualitative Medienforschung. Ein Handbuch [Qualitative Media Research. A Handbook] [47 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 8(3), Art. 7, http://nbn-resolving.de/urn:nbn:de:0114-fqs070373.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License