Qualitative methodology includes a variety and diversity of methods, procedures, and research designs. All kinds of qualitative methods have in common that their main research aim is a deeper understanding of the research object. Therefore, they are nonstandardized tools that can be adapted flexibly to every kind of research object, which can better be called research subjects because qualitative methods do not measure them objectively but interact with them, insofar as method is not a neutral tool in order to gain knowledge about researched subjects but is part of the social reality investigated.
Aims Of Qualitative Methodology
Qualitative methods try to discover new hypotheses rather than testing hypotheses deductively derived from known theories; they explore new phenomena and describe them intensively (“thick description”) and from different perspectives. This is essential for the most prominent approaches in qualitative methodology, such as grounded theory, ethnography, case studies, social hermeneutics or phenomenology, feminist methodology, or action research. They all have several research aims in common.
First and basically, qualitative methodology is directed at the understanding of the social world that qualitative researchers explore and investigate. Social research, therefore, tries to reconstruct the social constructions of people’s cognitions, emotions, communications, or actions. Understanding and meaning cannot be taken for granted but need a great effort to recover within the research process by way of intensive interaction between researcher and the people researched (or the texts researched). Epistemologically, meaning is itself not a given object but must be “negotiated” in social interaction. Understanding can be characterized as the process of exploring and (re)constructing meaning.
Second, every piece of scientific knowledge and the process of gaining it are necessarily context-bound and in consequence strictly relative to this context. Context originates both from within the social reality and social interactions investigated and from the interaction between the researcher and the researched subject (people, actions, etc.) or object (texts, visual material, etc.). The researcher’s methods and tools for gaining knowledge are not considered neutral or technical instruments but establish, develop, and affect the research context. Even basic key terms and concepts used in research projects are not only defined by the researcher but elaborated within the data collection process through communication and “negotiation” with the informant or research subject being investigated.
Third, research outcomes are not only achieved results within a research process, but can be characterized as being dynamic knowledge in an epistemological sense. This is the reason why research instruments have to be open, flexible, and adaptive to the research subject or object. Research instruments are not the fixed tools of operationalized theories that have to be completely and finally developed before fieldwork begins. Instead, they have to be changed (and changeable) within research interaction according to the requirements of the individual and case-specific research situation.
Fourth, the procedure of qualitative research is primarily inductive insofar as it starts with research questions and temporarily “ends” with hypotheses. The research process is spiral instead of linear because there is a permanent dynamic inherent, which includes developing, testing, and changing research questions, premises, hypotheses, instruments, and tools. At the end of the study there should be a better and deeper understanding of the research subject investigated. Consequently, research methods need to be reflected at each stage of the research process with respect to their contribution to the research results.
Different Qualitative Approaches
The different approaches mentioned above emphasize different aspects of qualitative research, although there are many commonalities in fundamental research aims and in epistemology.
Phenomenology is the study of human phenomena and experiences in everyday life. As social phenomena are not entirely conscious for the people experiencing them, the researcher has to explore different perspectives to gain a full picture of the phenomena being investigated. The foreground of the participants’ life-world can be researched by asking the participants about their subjective perceptions. The background of the participants’ experiences can be explored by asking other participants about their understanding of the researched person’s actions or by observing the participants’ actions within their social context. The main aim of phenomenological research strategy is to gain a holistic impression and a deeper understanding of the phenomenon researched and its nature by changing the perspectives of research.
A very similar approach, although less philosophical and more practical, is ethnographical research, which emphasizes field research in natural settings and the investigation of people’s everyday lives. The researcher adopts an insider’s perspective, trying to understand how people understand and construct the world they are involved in. Thus, ethnographical investigators primarily rely on participant observation, in-depth interviews. and analysis of (private) documents; data collection is eclectic. Ethnographic researchers do not confine themselves to a single and short-term research contact but establish a longer contact and intensive relationship with the field under study (community, group, organization, etc.). Another characteristic of ethnographic research is reflexivity, which is particularly relevant for the researcher’s ambivalent role and status within the field as both (distant) observer and (close) participant.
Grounded theory emphasizes the spiral and strictly inductive research process: Instead of establishing hypotheses at the beginning of a study, researchers start their investigations only with open research questions. The research process is considered a permanent alternation of data collection and data analysis. Thus, sampling procedures are not planned systematically in advance, but the researcher enters the field with a convenient sample of first interviews, observations, or gathered document materials. Provisional theoretical concepts are identified from the first distinct events or incidents in the data. Then further data are gathered by either the same method or different methods and are analyzed in detail. New questions arise from the data, which need to be answered with the help of further data, and so on. The process of data collection and data analysis comes to an end when no further concepts or dimensions emerge from the data and theoretical saturation is reached. Of course the researcher still may discover a “negative case” that does not fit the concepts elaborated. Then the theory must be modified or extended.
Case studies privilege in-depth inquiry over coverage when trying to understand the case and its complexity rather than to generalize from the case to a certain population of individuals or organizations. As is true for the approaches previously described, the participants’ interpretation of their actions or of their worldview is at the center of the investigation; case studies are rather long-term investigations than one-shot studies; and they follow an inductive logic. The case study approach slightly differs from the other approaches, insofar as its explorative character is less demanding. The results of case studies are considered descriptive (rather than leading to hypotheses or theories), heuristic and provisional (rather than saturated), and particular (rather than generalized).
Another claim for qualitative methodology is made by feminist methodologies and by action research. Although feminist methodologies are not methodologies in a strict sense but can be understood as a (meta-)theoretical approach or perspective, they suggest a preference for qualitative methodology and its practical or political consequences. They sensitize the researcher to the politics and power influences underlying each methodology and each method. As research and methodological rules are mostly masculine – most researchers have been male – the dominating role of objectivity in research and methodology is itself masculinist. Instead, feminist methodology insists on the positionality (“standpoint epistemology”) and reflexivity (“logic of self-reference”) of social research and the researcher. It is a critical methodology and has normative implications for social practice, claiming justice and equality between sexes. Implications for research designs include the empowerment of the people researched and the cooperative production of knowledge.
This is also and particularly true for action research, which aims to change practice with the help of and for the participants researched. Social scientific research causes political consequences in two different ways: with respect to the relationship between researcher and researched and with regard to the change or improvement of social practice outside the research process. Action research helps the participants researched to reflect their own practice and to change a criticized situation. In this perspective, all steps in research, from the early stage of defining the research problem to the final data analysis, result from the processes of negotiation and cooperation between researcher and participant.
Methods In Qualitative Research Practice
Prominent methods in qualitative methodology are participant observation, in-depth interviews, narrative interviews, qualitative content analysis, discourse analysis, conversation analysis, social scientific hermeneutics, etc. They will be described shortly.
With the help of observation, events and actions can be investigated. Observation can either be direct (nonparticipant) or participant. There is a preference for participant observation in qualitative research, particularly when an ethnographical approach is chosen, because in direct observation the observer is cut off from the people he or she observes and therefore cannot ask the meanings of or reasons for their actions and behaviors. Practically, researchers or observers first have to gain access to the field, which implies defining their role as more or less active participants in the field or situation observed. Then observers make a record of their impressions of the situation. As human behavior is complex, observers permanently select particular elements of the situation to be observed. Furthermore, observing includes interpreting the meaning of the impressions made, insofar as observation is an observer’s subjective, selective, and sense-making activity. In qualitative observation there is no standardized schedule with predetermined categories; instead observers record relevant impressions with detailed but unstructured field notes and sometimes with the help of video equipment. Although qualitative observation is unstructured in terms of measurement, it must be well prepared according to the research aim, to the role the observers take within the field they observe, and to the technology the observers use to record their observation.
In in-depth-interviews the researcher or interviewer uses an interview guide consisting of open-ended key questions and of probing follow-ups. The guide has a flexible order and structure so that the interview can be conducted as a dialogue with the respondent. It enables the respondents to express their point of view, to emphasize the aspects relevant for them, to clarify and illustrate the meaning of their responses, etc.
Narrative interviews do not even use interview guides as employed in in-depth interviews, because narrative interviews aim at gaining complete stories told by the respondents. If the respondents talk about their lives, narrative interviews are used for biographical research, life history, or oral history. Storytelling is an open-structured but rule-based dialogue or conversation. It is the aim of biographical research not only to gain true data about the informants’ life but also to get an impression of how the story of their life is told, how narratives are constructed.
Unlike individual interviews, focus groups consist of moderated discussions among a group of six to ten respondents or discussants. Instead of an interviewer, a moderator manages the conversation by asking questions and developing the discussion. The main aim of a focus group session is to find out how strong arguments are, to test new ideas or products, etc., and this is achieved by discussants’ mutual stimulation and controversial debate. As group dynamics emerge spontaneously, the results of the discussion may differ from group to group. Therefore, several groups should be selected to compare the outcomes. Depending on the research question, the samples of focus groups can either be homogeneous or heterogeneous with regard to demographic or other characteristics, real-world groups (families, neighbors, cliques, etc.), or artificial groups (consisting of people who do not know each other). All kinds of qualitative interviews are recorded with the help of an audio tape (sometimes even a video tape), then transcribed and finally analyzed.
Qualitative content analysis is a method to develop categories from texts or visual material including newspaper articles, radio or television programs, and web pages (primary material), but also transcribed interview texts, documents, or video-taped observations (secondary material). Categorization means a step-by-step reduction, abstraction, and generalization of the textual or visual material. This coding procedure is an inductive and spiral process, because it starts from the material and repeatedly goes back to it in order to check that the categories evolved from the material are correct and do not bias the original, context-bound meaning. Categories summarize, explicate, and structure the original material. They can be sorted on different levels of abstraction. Often categories may be combined in order to develop patterns or typologies of texts or other manifest material (photographs, films, documents, etc.) in further steps of analysis.
A more open and less rule-based kind of interpretation or data analysis is at the center of (social scientific) hermeneutics, which is particularly used in analyzing narratives or stories from interviews or texts. Instead of reducing propositional units to more abstract categories, hermeneutic interpretation emphasizes the functional relationship between units and the whole text, narrative, or story. Both formal and content-related characteristics of the language used in interviews, texts, or visual material are brought into play for the interpretation and understanding of meaning and social reality represented in and by the material analyzed.
Conversation analysis can be regarded as a particular kind of hermeneutic analysis that is related to dialogical structure, such as talk, conversations, debates, etc. Language not only represents social reality but is also an expression of human (communicative) action. Participants’ utterances and interactional forms are embedded in social rules, structure, and order. With the help of linguistic or semiotic tools these latent structures and patterns can be uncovered. Conversation analysis is not only based on communication represented in texts but also takes into consideration nonverbal and extra-verbal overt behaviors.
Discourse analysis combines a text-immanent hermeneutical approach with a contextrelated interpretation. Similar to conversation analysis, it characterizes discourse not only as a text but as a social practice in everyday life, involving ideology, authority, social rules, power, and even audience. Unlike conversation, discourse represents the struggle for arguments, status, and hierarchy and emphasizes sociological questions rather than linguistic ones. For discourse analysis both naturalistic material (documents, media contents) and interviews may be used. In addition to coding processes employed in qualitative content analysis, discourse analysts combine manifest characteristics of the text with latent power analysis (e.g., in the tradition of Michel Foucault). This is the reason why discourse analysis often follows a critical approach, uncovering the ideology and power relations behind the text.
As texts or other documentary material cannot be separated from their social production, documentary analysis also explores the social context of texts for a better understanding of the meaning expressed within them. Therefore, it is necessary to investigate the characteristics of the producers of texts and documents. Documentary analysis is particularly relevant for historical research and for media research in order to evaluate the status, truth, and relevance of documents, media contents, and their production.
Qualitative experiments use a particular stimulus, as do standardized experiments, but they explore the range of subjects’ reactions rather than test hypotheses. Within ethnomethodology Harold Garfinkel has developed so-called breaching experiments to reconstruct everyday rules: By breaching commonsense expectancies, questioning “normal” behaviors, or disrupting ordinary rules, it is possible to make self-evident practices conscious for the practitioner and to demonstrate the effectiveness of invisible rules, structures, and orders. Thus, experimenters use incomplete or inconsistent instructions. Another technique is for researchers to engage in an everyday conversation insisting on the clarification of the other person’s position, which obviously disrupts the conversational rule not to question clearly comprehensible statements.
As research instruments are open and flexible, so too are sampling procedures. In qualitative research they are mostly deliberative or purposive. The sample cannot be representative in a quantitative sense but should either be homogeneous, in order to find the small distinction among the sample units, or heterogeneous, in order to cover a maximum of variation and find the commonalities (and differences) among the sample units. A special case of heterogeneous sampling is extreme-case sampling, which is appropriate when the full range of possible answers, positions, and actions needs to be explored. Very often sampling procedures are driven by convenience because or if access to the field or population investigated is hard to gain. However, sampling procedure is far from being arbitrary; instead it is coherent and consistent with the research questions and the theoretical aim of the study.
In sum, qualitative research requires the researcher’s fundamental openness within a research process. In a cognitive sense this means that the researcher is open to unexpected aspects and events. In a social sense this means that there are no strict rules either for the researcher (interviewer, observer, coder or interpreter) or for the person interviewed or observed. Critics of qualitative methodology object that the lack of rules may cause arbitrary research results. Furthermore, if these results are strictly context-bound and can only be interpreted within their context, the idea of deciding between true and false results must be abandoned. Finally, the results of qualitative research cannot be generalized, as they are based on only a small number of cases.
Evaluative Criteria For Qualitative Research
However, there are evaluative criteria for qualitative research dealing with these problems: First, the most general criterion of qualitative methodology is clarity of presentation, which seems trivial as it is an overall criterion for every scientific effort. Process-related reporting of the research process is typical of qualitative methodology, in contrast to a more outcome-related reporting within quantitative methodology. Other researchers can comprehend not only the operationalization of theoretical constructs but each step of the process and can criticize its plausibility.
A second criterion may be called communicative validity or validation, which means that the researcher’s understanding of an answer, an observation, or a text interpretation will be checked with the research subjects’ own interpretation or with other researchers’ inferences. On the whole, the qualitative research process includes open and authentic data collection, which inherently promotes validity because it relies not merely on a sketchy research contact but on intensive and extensive interaction.
Third, generalization and generalizability can be achieved if the results and conclusions are transferable to other settings and contexts than those under study. Therefore, the sample should be heterogeneous enough with respect to the research question, the context should be elaborated intensively, and the concepts should be abstracted at a high level. Even in case studies results are generalizable, if the case is sufficiently complex to discover general structures rather than single social practices.
Fourth, reliability of data implies the stability of the results within the research process. Also, the systematic use of different methods and research instruments (“triangulation”) can provide reliability if the outcomes of different methods are consistent or lead to a coherent holistic impression of cases investigated.
References:
- Creswell, J. W. (1994). Research design: Qualitative and quantitative approaches. Thousand Oaks, CA, London, and New Delhi: Sage.
- Denzin, N. K., & Lincoln, Y. S. (eds.) (2000). Handbook of qualitative research, 2nd edn. Thousand Oaks, CA: Sage.
- Punch, K. F. (2005). Introduction to social research: Quantitative and qualitative approaches, 2nd edn. London, Thousand Oaks, CA, and New Delhi: Sage.
- Ragin, C. C. (1994). Constructing social research: The unity and diversity of method. Thousand Oaks, CA: Pine Forge Press.
- Seale, C., Gobo, G., Gubrium, J. F., & Silverman, D. (eds.) (2004). Qualitative research practice. London, Thousand Oaks, CA, and New Delhi: Sage.
- Somekh, B., & Lewin, C. (eds.) (2005). Research methods in the social sciences. London, Thousand Oaks, CA, and New Delhi: Sage.