Ethnographic Methods for the Study of Electronic Communities

By Sirkku Männikö and Rob Procter

The origins of ethnography lie in social anthropology. In recent times, however, ethnographic methods have found an increasing role within IT design and development (Anderson, 1994). In such cases, an ethnographic researcher is, in an integrated way, engaged in describing, planning and participating in a process of change, rather than merely observing and describing an existing setting.

The basic conditions for an ethnographic study can be summarised as follows:
  • The study is based on the researcher's comparatively intense and prolonged involvement in the studied environment.
  • The study demands a holistic, contextual attitude towards the problem area, rather than analytically picking up and studying a few aspects of the environment of the problem, the researcher should consider these aspects in relation to the socio-cultural context.
  • The study describes and analyses the setting from the participants' point of view rather than from the observer's. This demands an exploring and open manner from the researcher. The ethnography always must start at the participants' point of view and be based on this point, but can leave it behind at a later stage.
  • The ethnographic part of the evaluation may consist of a number of different methods of inquiry and data collection, including: participant observations, informal interviews, behavioural observations, structured interviews, questionnaires and analysis of documents. Data analysis follows a systematic and structured approach that is usually grounded through the use of various types of triangulation. In this way, comparisons may be made between the activities of an individual over time and in different settings, and between different individuals in similar settings, both to challenge the researcher's own assumptions, and to guide further data gathering.

Generalisability
The strength of qualitative methods like ethnography lies in the fact that by studying one case in detail the greater phenomenon can also be described. The results of ethnographic inquiry are often assumed, however, to be resistant to generalisation and therefore of questionable use for IT design. This is a mistake [Button and Dourish 1996]. First, ethnography expressly recognises the notion of "cohort independence", that is phenomena that are not exclusively determined by the specific circumstances in which they are observed [Garfinkel and Sacks 1970]. Second, and related, in ethnography, a specific case is understood to be a part of a greater phenomenon or a context . As with qualitative methods generally, by careful choice of case studies it is possible to produce findings that are transferable to other settings.

Time perspective of findings

Qualitative methods like ethnography are used to be able to understand the (social) conditions and affordances that make the achievement of a specific behaviour, interaction or process possible.. Based upon the thick descriptions  generated by ethnographic field work, systematic and detailed analysis can bring forth useful discoveries and explanations of these processes. In conventional ethnographic approaches, a prolonged period of time is usually needed for discovering and describing these processes in full and the same applies to finding changes in processes. However, experiences with applying ethnographic methods to IT design point to the possibilities of adapting these methods to suit the time scales of IT projects [COMIC 1994] without prejudicing their strengths. In this project we will similarly adapt our methods to match our time scales and needs.

Reproducibility
In the Natural Sciences reproducibility is control: the findings should be reproducible, thereby increasing the their credibility.
This attitude belongs to the positivistic paradigm that sees the world as unchangeable and constant. Qualitative research holds the opposite assumption. The world is constantly moving and changing and besides the social world is always being constructed and reconstructed. This makes the whole idea of reproducibility very difficult. If you in spite of this still want to control that reproduction is possible you can use the same theoretic of perspective as the natural scientist, if you are to follow the same general rules for collecting data and for analysis and you should get the same theoretical explanations to the phenomenon under the same circumstances as in the original one.

Participation
Participant observation is an important part of all qualitative research. Immersion in the setting allows the researcher to hear, see and begin to experience reality as the participants do. Ideally, the researcher spends a considerable amount of time in the setting studying the daily life and behaviour of the participants.

Observation
Observation is a fundamental and crucial method in all qualitative research. It is used to discover complex interactions in natural social settings.
Observation entails the systematic noting and recording of events, behaviours and objects in the social setting chosen for study.
Through observation, the researcher learns about behaviours and the meanings attached to those behaviours. This method assumes that the behaviour is purposive and expressive of deeper values and beliefs. Observation can vary from extremely detailed notation of behaviour, guided by checklists to more holistic descriptions of events and behaviours. In the early stages of ethnographic inquiry, the researcher enters the setting with broad areas of interest and without predetermined categories or observations checklists. Thus the researcher is able to discover the recurring patterns of behaviour and relationships. After such patterns are identified and described through early analysis of field notes, checklists become more appropriate and context-sensitive. Focused observation is used later in the study, to check analytic themes to see if they explain behaviour and relationships over a long time or in a variety of settings.
Also during an in-depth interview it is important to observe the participant's body language and feelings.

In-depth interviewing

The in-depth interview is described as "a conversation with a meaning".  The researcher can start with a few general topics to uncover the participant's meaning, perspective. It is after all the participant's point of view you want to find in this kind of study. It is most important that the researcher shows respect towards the participant and shows that the information given by the participants is important and valuable. Through the interview you get a lot of information in a short time. Immediate follow-up and clarification are possible. Combined with observation, interviews allows the researcher to understand the meaning people hold for their everyday activities.
Interviewing has its limitations and its weaknesses. Interviews involve personal interaction and, they are demanding on the scientist and the participant. The analysis of the data is time-consuming and one might even question the quality of data found this way. Because of this it is important to combine different methods to be able to guarantee the validity of the study.

Document Analysis

Since the computer conferencing tools used in this project are text based, they naturally lend themselves to a systematic examination of the forms of communication , including structure and content. . This more objective approach allows the researcher to obtain an objective and quantitative description of the organisation and content in the various forms of communication. One of the strengths of content analysis is probably the fact that it is so discreet, it can be conducted without disturbing the setting in any way. Privacy is not an issue in this case as the documents are, by the nature of the setting, implicitly public in character.
After gathering the material the researcher determines where the greatest emphasis lies after the data has been gathered. Besides the method of the procedure is explicit to the researcher.

References
Button, G. and Dourish, P. (1996). Technomethodology: Problems and Paradoxes. In Proceedings of the ACM Conference on Human Factors in Computing System Design, (Vancouver, April), ACM Press.
COMIC (1994) Field Studies and CSCW. COMIC Project (Esprit BRA 6225) Deliverable D2.2.
Garfinkel, H. and Sacks, H. (1970) On formal structures of practical actions. In J. McKinney and E. Tiryakrian (Eds.) Theoretical Sociology. New York: Appleton Century Crofts.