Digital Ethnography, Practice and Ethics

Date Posted
March 27, 2008

The Internet has had many effects on research methods and design. The onset of the Internet has brought on new data to analyse, research methods, analysis and tools for research, as well as challenging some of the traditional research methods and ethical issues associated with it.

The introduction, expansion and increasing penetration of the Internet has created a larger volume of diverse, easily accessible data to research and analyse. Researchers now have access to additional research subjects in new ways, to broader samples with greater cultural diversity (Clark 2002). The Internet also has it’s own unique communication tools which allows researchers to have immediate access to transcribed conversations, archives, action recording and observation (Negretti 1999). Along with this it raises new issues for researchers relating to anonymity, different biases and behavioural changes in the subjects.

The Internet breaks down geographical boundaries due to its networked nature (Johnson 1998). For example it can cut out the expense of postage and the poor response rate for mailed surveys, allowing surveys to be easily completed and returned immediately on-line. It also therefore makes it easier for respondents to be geographically diverse which significantly broadens the cultural base from which a sample can be drawn (Sheehan & Hoy 1999). However this also raises its own challenges for researchers as each culture interacts on the Internet differently, leading to culturally different levels of “honesty”. Subjects from autocratic cultures would be wary of being traced via the Internet and they would censor their responses accordingly (Frankel 1999).

Researchers using the Internet also find they have access to new tools, unavailable in traditional research. One of these is the “native textual medium” which allows the synchronous transcribing of conversations. Not only is this instantaneous, it is archived, concrete and objective, removing the human error factor. Therefore material can be responded to asynchronously, allowing in depth replies and more detailed decision making. In addition, the ease of reproduction allows independent researchers to validate qualitative research (Papacharissi & Rubin 2000).

Another tool is web server data logs (traffic/linking) which automatically track and analyse users actions and behaviours on the Internet (Srivastava et al 2000). Such data is difficult and expensive to replicate in the physical world. Furthermore, automatic analysis of user behaviour is possible, and this can quickly aid in evidence-based decisions.

The ease of anonymity has affected research design and methods in many ways. Subjects often provide more information when posted anonymously, which can be helpful or hindering. Anonymous subjects make it hard to gain informed consent for research. Although many aspects of the Internet are anonymous by default and others are made anonymous intentionally by the user or venue. However, often anonymity is more perceived than a reality since information about the user can be sourced with relatively little effort by the sophisticated user (Freeman & Bamford 2004). This has caused many new ethical issues for researchers to consider in their methods and design or research .

Unfortunately the Internet also creates research biases to such as those who have access to the Internet and those likely to interact with the research and/or give permission for themselves to be researched (Solomon 2001). However, often the geographical and cultural diversity of the Internet seems to compensate for and overshadow this built in bias (Yang 2000).

The virtual medium of the Internet can affect research designs – removing the physical self can cause some people to interact differently. Furthermore because visual cues (body language) are absent, actions and communications may be interpreted differently (Freeman & Bamford 2004). Users honesty may vary as well as their personal investment and involvement in the medium. As visual and audio technologies on the Internet expand, this limitation may reduce to a degree.

Although traditional methods of research can be transferred onto the Internet, changes in their execution, design, ethical frameworks and analytical properties are needed (Nosek et al 2002). The data collection methods used by researchers such as surveys, interviews, focus groups, ethnography, action and observational research need to be analysed in the light of this.

On-line surveys are very attractive to researchers as they provide “convenient, verifiable, low-cost delivery and return systems as well as easy access and feedback mechanisms” (Kaye & Johnson 1999). Surveys can be conducted more efficiently and effectively than ever before, but this comes at a cost due to the bias of Internet responders (Zang 1999).

Interviews are affected greatly by the change of medium to the Internet. Traditionally face-to-face interviews can often be a poor representation of an individual or perspective, but on-line an interviewee has the tools to represent themselves differently. Subjects can author their words, re-read, iterate, revise, access resources. This reduces the likelihood of the subject being distracted or intimidated by the physical presence of the interviewer or second-guessing and misreading non-verbal cues (Crichton & Kinash 2003).

Traditionally focus groups have been seen as one of the more applicable, beneficial and expandable forms of research and market research (Schneider et al 2002). On-line focus groups are one of the newest, most widely used innovations in qualitative research (Morgan 1997). On-line focus groups can be both synchronous and asynchronous through bulletin board systems, discussion boards, email/usenet groups, instant messaging, chat rooms and video/audio conferencing. The asynchronous nature allows respondents more time to put together in-depth arguments (Kozinets 2002).

In Ethnographic research the researcher is immersed fully in the research environment, analysing it as it progresses, yet with having minimal effect on the research outcomes. This is made easily possible due to the non-physical nature of the Internet, where a researcher can lurk and not be seen if they choose. However, the research design and methods must account for the ethical issues this raises (Beaulieu 2004).

Although recording is simpler on-line, the physical world has many valuable interactive, inter-personal features that can be studied within focus groups, interviews and ethnography. The physical interaction may be lost when studied on-line. Research design and methodology should take this in to consideration before basing too much on this kind of on-line research (Kozinets 1998).

Action research is made easier on-line as researchers can implement change and easily watch and effortlessly record the changes. “Participation is one of the most important characteristics in action research. The action research process is not static. It is more cyclic, alternating between action and critical reflection” (Auer & Follack 2002). On the Internet a researcher can analyse user behaviour, website traffic, users comments as well as being involved in discussion with the users and having it automatically transcribed for further analysis.

Traditionally it is almost impossible to have easily accessible observation methods, especially without high costs and substantial invasions of privacy. The Internet changes this for a researcher as all user actions are automatically recorded – every time a file or piece of information is requested from or sent to a server an IP address is recorded. IP addresses can, at best, be traced to an individual machine, but for many on-line services that require a log in, the specific users actions can be recorded and traced to that user. This is unique, useful and yet dangerous for the privacy of an Internet user and again research design and methods must take into account the ethical issues involved (Srivastava et al 2000).

Web server data logs provide an enormous vault of objective data. This ranges from entry/exit pages, number of visits to length of visits and much more. This kind of information can be useful for a researcher to determine what information is popular, needs more/less detail, should be removed/modified, is not receiving enough visits, and where their strengths/weaknesses are (Srivastava et al 2000). Researchers should take note that although there is a comprehensive database of objective information, much of it may be inaccurate and somewhat irrelevant if not placed against other more subjective data (opinions and perspectives).

The technologies and research methods of the Internet when used comprehensively can improve analysis methods. The range and depth of information now available through the Internet can help make more informed decisions with more data. Different types of data (surveys, focus groups, web server data logs etc…), juxtaposed together can provide different angles to draw from when analysing research results and making decisions (Jadad et al 2000).

Many of the Internet communications are textual and these can be analysed in various ways. Research methods should be designed to also take into account the different perceptions of the same textual expressions. Many Internet users are not authoring material with the idea in mind that it is a publication (Harter & Park 2000). Therefore there is often information left out and assumed knowledge which can lead to other respondents or researchers getting the wrong impression of a user’s textual expression. A lot of emphasis is put on the methods of data collection and often the concepts and interpretation are left behind.

The Internet has automatic computations using many tools to capture an analyse information. Search engines employ computer programs (robots) to read web-pages and follow links, in the process information is archived and indexed. This process allows people to easily search information. This is a great tool for research as specific, desired information can be found with ease (Gordon & Pathak 1999). Boolean searching assists in narrowing down information with specific parameters and goals in mind.

University research requires the following four critical aspects to be ethical “[1] subjects must know the purpose of the study, [2] possible benefits/harm; [3] participation is voluntary and [4] they can withdraw without penalty any time” (Fraenkel & Wallen 1993). Many ethical issues arise with the technical capabilities and failings of the Internet coupled with users lack of understanding of these technologies. Some technical capabilities raising ethical issues include permanence, search capabilities, ease of reproduction and lurking. Furthermore, the possible technical failure and limitations must be considered in putting users first (Nosek 2002).

On the Internet data is permanent unless deleted by an administrator, or by a user if there is a mechanism in place. However, even if data is eventually deleted there is a good chance that there is an archived record of it. This permanence of data can be dangerous for a user’s reputation. For example if a user makes a sexist comment on-line, even unintentionally, it will stay there as a record and can be found years later and used against the person (Mann 2002).

Data permanence is magnified with the improving web searching capabilities. For example a researcher may publish a comment authored on the Internet, a user’s name or pseudonym, date and/or place where the comment was made. The search capabilities of the Internet allow the source to be easily found (Yang 2005, Kozinets 2002).

However if the original source of on-line content is erased and not automatically archived, its non-existence still cannot be certified. Inherent in the Internet is the ease of reproduction – forwarding emails, copy and paste, download and upload. This becomes even more dangerous to the author if it is reproduced without appropriate context.

As noted earlier, innate in the Internet is the ability for a researcher to be present, “lurking” and yet unnoticed (Beaulieu 2004). Although often seen as a great advantage, as not to disrupt the research subjects (Tolich & Fitzgerald 2006), yet has the potential to breach the University research code of ethical practice as subjects do not know that they are being studied and therefore there is no knowledge of the purpose of the study, benefits/harm, they do not voluntarily participate and therefore cannot withdraw (Fraenkel & Wallen 1993).

In fact there are many on-line factors which make it difficult to abide by traditional ethics principles. Researchers face a hard task in obtaining and validating consent from Internet users to be studied (Ess 2002). They must also consider the factors of timing, the medium used, the nature of the on-line venue, authority of consenters, validation and perceptions of authors in their research design and methods.

In conclusion, one of the most significant considerations for further research on the Internet will be the ethical issues involved. While the introduction of the Internet has significantly assisted researchers in more easily collecting and analysing a broader range of data, the challenge for researchers will be to integrate this opportunity into their current ethical and research frameworks.

Date Posted
March 27, 2008

References

• Beaulieu, A. 2004. Mediating ethnography: objectivity and the making of ethnographies of the Internet. Social Epistemology. Communications Magazine, IEEE. Volume: 35, Issue: 5. pp52-57
• C Auer, M Follack. 2002. Using Action Research for Gaining Competitive Advantage out of the Internet’s Impact on Existing Business Models. The Proceedings of the 15h Bled Electronic Commerce Conference. eReality: Constructing the eEconomy. Bled, Slovenia, June 17 – 19, 2002. bledconference.org
• Clarke, P. 2000. The Internet as a medium for qualitative research. South African Journal of Information Management. generalupdate.rau.ac.za
• Crichton, S. Kinash, S. 2003. Virtual Ethnography: Interactive Interviewing On-line as Method. Canadian Journal of Learning and Technology. collection.nlc-bnc.c
• Ess, C. and the Association of Internet Researchers (AoIR) 2002. Ethical decision-making and Internet research: Recommendations from the aoir ethics working committee. Approved by AoIR, November 27, 2002. Available online: http://www.aoir.org/reports/ethics.pdf
• Fraenkel, JR. Wallen, NE. 1993. How to Design and Evaluate Research in Education.
• Frankel, M. & Sang, S. 1999. Ethical And Legal Aspects Of Human Subjects Research On The Internet, AAAS Workshop Report
• Freeman M & Bamford A. 2004. Student choice of anonymity for learner identity in online learning discussion forums, International Journal on E-Learning, pp. 45-53
• Gordon, M. Pathak, P. 1999. Finding Information on the World Wide Web: The Retrieval Effectiveness of Search Engines. Information Processing and Management.
• Harter, SP. Park, TK. 2000. Impact of prior electronic publication on manuscript consideration policies of scholarly journals. Journal of the American Society for Information Science.
• Jadad, AR. Haynes, RB. Hunt, D. Browman, GP. 2000. The Internet and evidence-based decision-making: a needed synergy for efficient knowledge management. Canadian Medical Association Journal, 2000.
• Johnson, DR. Post, DG. 1998. The New ‘Civic Virtue’of the Internet. The Emerging Internet.
• Kaye, B. K. Johnson, T. J. 1999. “Research Methodology: Taming the Cyber Frontier-Techniques for Improving Online Surveys.” Social Science Computer Review 17: 323-337. Southern Illinois University
• Kozinets, RV. 2002. The Field Behind the Screen: Using Netnography for Marketing Research in Online Communities. Journal of Marketing Research, 2002. economia.unipv.it
• Kozinets, RV. 1998. On Netnography: Initial Reflections on Consumer Research Investigations of Cyberculture. Advances in Consumer Research.
• Mann, C. 2002. Generating data online: ethical concerns and challenges for the C21 researcher. Keynote lecture, Nordic Conference on Internet Research Ethics, Trondheim, Norway, June1-2, 2002.
• Morgan, DL. 1997. Focus Groups As Qualitative Research. Sage Publications
• Negretti, R. 1999. Web-based activities and SLA: A conversation analysis research approach. Language Learning & Technology, llt.msu.edu
• Nosek, BA. Banaji, MR, Greenwald, AG. 2002. E-Research: Ethics, Security, Design, and Control in Psychological Research on the Internet. Journal of Social Issues. ingentaconnect.com
• Papacharissi, Z. Rubin, AM. 2000. Predictors of Internet use. Journal of Broadcasting & Electronic Media.
• Schneider, SJ. Kerwin, J. Frechtling, J. Vivari, BA. 2002 Characteristics of the Discussion in On-line and Face-to-Face Focus Groups. 2002. Social Science Computer Review, Vol. 20, No. 1, pp31-42. Sage Publications
• Sheehan, KB. Hoy, MG. 1999. Using e-mail to survey Internet users in the United States: Methodology and assessment. Journal of Computer Mediated Communication.
• Solomon, DJ. 2001. Conducting web-based surveys. Practical Assessment, Research & Evaluation.
• Srivastava, J. Cooley, R. Deshpande, M. Tan, PN. 2000. Web usage mining: discovery and applications of usage patterns from Web data. ACM SIGKDD Explorations Newsletter.
• Tolich, M. Fitzgerald, MH. 2006. If ethics committees were designed for ethnography. Journal of Empirical Research on Human Research Ethics. Univ California Press
• Yang, Z. Zhong, S. Wright, RN. 2005. Anonymity-preserving data collection. Conference on Knowledge Discovery in Data. portal.acm.org
• Zhang, Y. 2000. Using the Internet for survey research: A case study. Journal of the American Society for Information Science. Volume 51, 1. pp 57-68. John Wiley & Sons, Inc. School of Library and Information Science, Kent State University. Kent