based on Kozinets 2002

Netnography is a research method derived from the technique of ethnography developed in anthropological research, filtered through the increasingly complex opportunities for information about communities and on the world wide web. Ethnography, the anthropological basis for this method, has gained popularity in sociology, cultural studies, consumer research, and various other social scientific fields (Kozinets 2002). It is comprised of both fieldwork, the study of the distinctive meanings, practices, and artifacts of particular social groups, and the representations based on the fieldwork. Rather than commence from a positivistic or scientific realism point of view, where a researcher would begin with theory, develop hypotheses, design and execute experiments to test hypotheses, and draw conclusions, ethnography approaches construction of meaning in social groups in an open-ended way. The development of information, field notes, documentation, and data is predicated on developing relationships of trust with the subjects of the study, in order to develop streams of information that are not constructed or manufactured, however innocently, especially for publication. It relies heavily on the"acuity of the researcher-as-instrument" (Sherry 1991, p572) and is framed in unavoidable ways by the interests and skills of the researcher. Metaphor, hermeneutics, and analytical interpretation of data are employed. This takes a lot of time and casts the researcher necessarily in the role of participant-obeserver as well as interpreter. As communications begin to establish a picture of the social construct and dynamics, the researcher must acknowledge and own his or her own point of view, as the development of representations may only be built through their own constructions of reality, their own perspectives and experiences. Of course, the same caveat may be applied to any kind of research, positivist or otherwise, because people are human beings, even when they wear white lab coats and aim to be objective and detached. That said, ethnographic research is nevertheless grounded in knowledge of the local, the particular, and the specific (Kozinets 2002). The grounding in knowledge of ethnographic research attempts to follow the path of particularized understanding that has been codified as "grounded theory" (Glaser & Strauss 1967).

These methods yield rich qualitative findings that follow rather than lead the subject group, and embrace varieties of approaches that facilitate adaptability to a wide variety of circumstances, leading to its popularity and applicability in understanding the behaviors and construction of meanings for social groups all over the world since the 19th century.

Field procedures in ethnography include developing cultural entee, gathering and analyzing data, ensuring trustworthy interpretation, conducting ethical research, and providing opportunities for culture member feedback (Kozinets 2002). These procedures have been developed in the course of time and face-to-face research, but there is increasing impetus to cast the web communities on the Internet as appropriate subjects for study (Granitz & Ward 1996; DeCindio et al 2003; M. Venkatesh 2003; Fischer & Bristor 1996). Social exchange on the Internet is a dynamic phenomenon, but prior research shows that, following the 'Pareto' rule of 80-20 which is operative in almost all consumer marketing, the primary and weighted component of information intereresting and significant to marketing professionals is likely to be there (Kozinets 1999).

Prior to embarking on this adventure, the researcher must have 1) specific questions; and 2) identified particular online forums appropriate to to answering the specific questions. In order to determine appropriateness, the researcher must learn everything possible about the forum and the individual participants comprising the major component of the discourse. These tasks are facilitated by online search engines.

TYPES OF ONLINE COMMUNITIES

For the purposes of marketing and advertising research, at least five types of web communities may be identified. All of these forums are populated by people who have self-segmented by whatever parameters form the subject of the group. These comprise 1) bulletin boards/newsgroups/usergroups organized around a particular product, interest, or service (e.g. Yahoo has 2228 registered participants for the Collectible Card Games Group alone, and over 46000 registered for rock and pop music artists); 2) independent web pages/web rings composed of thematically linked web pages such as epinions.com and www.dir.webring.yahoo.com; 3) lists/listservs which are thematically linked e-mail lists (such as collectible card games and rock stars), and may be found through search engines such as egroups and liszt.com; 4) multiuser dungeons; and 5) chat rooms which (to date) tend to be oriented toward specialized topics (sex, fantasy, online dating), but may yield interesting advertising and marketing data nevertheless.

CRITERIA FOR SELECTION AND CULTURAL ENTREE

Posed with choosing between several online community options for research, the investigator may parse the choices using the folowing preferential heuristics: 1) relevance to research topic; 2) relatively higher traffic; 3) relatively higher number of posting contributors; 4) relatively richer or more detailed description or elaboration of discourse; and 5) relatively higher number of between-member interactions relevant to the research question--in short, the richest and most dynamic field for addressing the research question. The researcher should be familiar with the characteristics of the group (membership, market-oriented behaviors, interests, and language) before initiating contact.

DATA COLLECTION AND ANALYSIS

Netnography generates two kinds of data: 1) transcripts copied from the forum itself; and 2) researcher's written notes on observations of the community and its members, interactions, and meanings. Capturing the text of the discourse is infinitely easier for the researcher in this method because it is the nature of the beast to create a standing written record of all communications between members, as contrasted with traditional ethnographic techniques which usually involve a person with a notepad writing down everything he/she can hear or remember. Clearly, the acquisition of primary data is more accurate and less time consuming, and the researcher may be faced with the problem of data overload. Culling data should be guided by the research question and other relevant parameters such as subject willingness to be interviewed, subject articulation, and time.

A taxonomy of postings may be developed to classify entries, beginning with determination of social or informational in content, and whether the posting is on- or off- the topic of research interest. All of the data taken together may provide a context for a "grand pass," but the researcher might choose to focus on specific types of information to answer the research question. Also of interest would be information/categorization on the posters of information--possibly on the basis of participation levels/roles/consumption activity. Categories of posters identified (Kozinets 1999) include 1) tourists--lack deep interest and strong social ties to the consumption activity; 2) minglers--have strong social ties but minimal interest in the consumption activity; 3) devotees--strong consumption interests but few attachments to the online group; and 4) insiders--strong social ties to the online group and to the consumption activity. For marketing research, insiders and devotees represent the most important data sources, and these groups have been shown to exhibit devotion, enthusiasm, active involvement, and sophistication in their user segments.

Data collection should continue as long as new insights on important topical areas are still generated, following grounded theory (Glaser & Strauss 1967). It may be useful to quantify numbers of: items read, messages, web pages, participants by type, etc. Netnography uses particular ties to specific online consumer groups and the revelations found in the depths and insights therein to form interesting and useful conclusions about how the group constructs itself, how it develops social meaning, and how it derives personal gratifications.

ENSURING TRUSTWORTHY INTERPRETATION

Data must be contextualized, often concurrently with data collection, and the challenge of netnography is to do so without the attendant social and physical cues of face to face research. Instead, netnography relies on the context, metaphor, symbols used to interpret meaning, and various social science research software designed to expedite coding, content analysis, data linking, display, and theory-building functions (Paccagnella 1997).

"Trustworthiness" is the goal of the quality of the data and of the interpretations offered--moreso than the concept of "validity," which can be elusive in a naturalistic context. To ensure trustworthiness, researchers should follow conventional procedures of research protocols (Lincoln & Guba 1985; Wallendorf & Belk 1989), of which there are a variety.

RESEARCH ETHICS

As in any emerging field, there are questions that attend to the issues of ethics. Two major issues are emerging: the first raises the question of what is public and what is private online. Some researchers may feel a moral obligation to obtain explicit permission from the authors for publishing logs in academic papers (Marvin 1995) while others collect logs without asking permission but then have computers, not people, read them (Danowski & Edison-Swift). The second issue revolves around the privacy of the users, take precautions such as changing names, pseudonyms, or 'handles' and removing addresses from the logs. At the heart of this concern is the issue of 'informed consent' in Cyberspace. To preserve the integrity of the source, it is important for the researcher not to disrespect the subjects and thus poison the research well. The Forum on the Ethics of Fair Practices for Collecting Social Science Data in Cyberspace (Thomas 1996) reviews the variety of positions about ethical guidelines in online social research, and identifies three basic approaches: deontological, teleological, and postmodern.

Suggested guidelines for ethical research procedures for marketing researchers using netnography include: 1) the researcher should fully disclose his or her presence, affiliations, and intentions to online community members during any research; 2) the researchers should ensure confidentiality and anonymity to informants; 3) the researcher should seek and icorporate feedback from members of the online community being researched; and 4) the researcher should take a cautious position on the private vs public medium issue, including the securing of permission (informed consent) to use any specific postings to be directly quoted in the research, personal stories, or other online artifacts covered under copyright law.

MEMBER FEEDBACK

Allowing informants to check reports and interpretations of the research is considered integral to the integrity and trustworthiness of the work (Arnould and Wallendorf 1994; Hirschman 1986; Lincoln & Guba 1985). These reviews provide another opportunity to authenticate events and meanings with the subjects, and may be relatively easy and timely to task (subjects will all have email and access to a computer, for instance).