AGOCG logo
Graphics Multimedia VR Visualisation Contents
Training Reports Workshops Briefings Index
Also available as an Acrobat File Back Next



Pratical Experience



VR in teaching

Multi-User VR

Multi-User VR in teaching


Observation Study


Data Capture

Ethical Issues




Case Studies Index

Multi-User Virtual Reality Technology as a Laboratory for Learning about Social Research: Issues and Prospects

11. Ethical Issues

Since the ethical issues involved in studying virtual worlds have been mentioned on several occasions, these issues deserve to be discussed at greater length: an important aspect of ethical conduct in research is the principle that research participants should be treated with respect and as autonomous agents (National Commission for the Protection of Human Subjects, 1978). From this flows the requirement for obtaining informed consent from research participants. Students who participated in the project did so on the basis of fully informed consent about the purposes and procedures of the study. It should be clear that, although most questioning was overt, in some cases those encountered by students in the virtual world were unaware of the purpose behind the questions being asked of them. We think, though, that as it was constituted the experiment did not raise major ethical difficulties. The questions asked were innocuous and of a kind frequently employed at the beginning of on-line conversations. Moreover, participants in a virtual world have a repertoire of actions for signalling when they do not want to maintain contact.

A second principle of ethical conduct in research is that no harm should come to research subjects as a consequence of their participation in a research project. Flowing from this principle is a concern for the privacy of research participants. Issues of privacy arise when personal information gathered for research purposes can be tied to a specific individual. (Aggregated information, e.g. about the number of users in a particular world at a particular time poses no threat to privacy.) In fact, little threat exists to the privacy of those individuals interviewed by students. Since users are represented in multi-user VR by avatars and refer to themselves by aliases, identifiable data are not available. In theory, some identifiable information is available to systems operators but interlocutors only have information that is volunteered to them.

The design of the experiment provided further protection to users: threats to privacy come into being through dissemination of research data. In this study, the primary interest was the behaviour of the students rather than other users. As a result, no use was made of the information they gathered from other participants in the world they visited. Nevertheless, it is generally considered undesirable to put students, especially undergraduates, into situations of potential ethical difficulty (Lee, 1987).

Potential ethical difficulties, together with the problem that some students had to wait to find someone with whom they could interact, suggest that it might be appropriate in the future to modify the sort of procedures used here. In particular, the teaching session could be held in an `empty world' and participation could then be restricted in the research exercise to class members, each of whom would be assigned an `identity' and a set of social attributes that had been generated factorially. The aim of the exercise would be for students to elicit through interviews as many attributes of the others in the world as they can. The added benefit here is that in debriefing, the data gathered could be measured against the characteristics previously assigned. Such an approach might also alleviate a further possible problem that has been found in MUDs or MOOs. Popular sites increasingly attract the attention of researchers in ways that alter the character of the site. In consequence, that which was initially of interest is destroyed. A further consequence is that participants in virtual worlds can increasingly become irritated by the presence of researchers and distrustful of them. We emphasise we have no evidence that anything like this was generated by our experiment. Indeed our student ethnographers were welcomed and helped as much as they could about the world they visited. Nevertheless, we suggest there is a need to move forward with caution.

Graphics     Multimedia      Virtual Environments      Visualisation      Contents