In this paper the authors report on a piece of action research (using the critical incident technique) in the area of PhD supervision. Twenty eight supervisors, most of whom are co-supervisors, were instructed in critical incident technique (Flannagan, 1954) and then asked to write incidents from their supervision experience. The incidents could be about mundane matters but they became critical, in the meaning of this technique, when they were analysed or ‘critiqued’. The incidents we collected were critiqued in four stages. First, by the individual authors who wrote down their own analytical reflection of the incident. Second by focus groups consisting of five authors per group. Third, by a plenary session that discussed the reports from each group; and finally, by the researchers, who accessed the written incidents and individual reflections as well as the plenary discussion and summaries from the focus groups. The incidents were classified according to type and their actual and potential consequences examined. Possible assumptions underpinning the incidents were also analysed individually, in focus groups and by the researchers. Our aim was to heighten the supervisors' awareness of issues in supervision and to give them a tool for future reflection and analysis. Many of the incidents were problematic even though the informants had been asked to simply write down an event they remembered as a supervisor or as a former PhD student. Issues and conflicts relating to co-supervision were a dominant theme. Supervisors were helped via this action research to find ways of preventing or overcoming this and other types of conflict. One recommendation from our study was that a transparent structure and set of rules governing co-supervision be defined and implemented.

Reference: Flannagan, J (1954). ‘The critical incident technique’ in the Psychological Bulletin, volume 51.

The aim of this action research project was to heighten the awareness of PhD supervisors concerning issues in supervision and help them to improve their practice. Action research, which has its theoretical and philosophical roots in critical theory (Lewin, 1946 and Habermas, 1972) is collaborative, cyclical and intended to make a change for the better in professional or other human endeavors. In our case the twenty eight supervisors cooperated in this research on the understanding that by undertaking workshops connected with it and being able to discuss progress and final results from it they would develop critical, analytical and reflective tools that they could continue to use in their future supervision.

Critical incident technique is attributed to the American psychologist, John Flannagan (Flannagan, 1954). In the words of a later researcher it is ‘an exploratory, qualitative method of research that has been shown to be both reliable and valid in generating a comprehensive and detailed description of content domain’ (Woolsey, 1986, p 242). The analysis of data has to be governed by the overall aim of the research and tied closely to the content domain, which, in our case, was supervisory practice in a Swedish engineering university. In our study we involved twenty eight supervisors, most of whom were acting or co-supervisors. The situation in Sweden is that an academic can not act as the main supervisor until he or she has docent competence or has gained that title. In reality many young researchers co-supervise PhD students who are formally enrolled as the students of the Professor in charge of the research group.

Our informants undertook workshops in critical incident technique and were then asked to write incidents from their experience as supervisors or students. The incidents we collected were critiqued in four stages. First, by the individual authors who wrote down their own analytical reflection of the incident. Second by focus groups consisting of five authors per group who discussed the individual analyses. Third, by a plenary session where reports from each group were presented; and finally, by the researchers themselves who accessed the written incidents and individual reflections as well as the plenary discussion and summaries from the focus groups. The incidents were classified according to type and their actual and potential consequences examined. Possible assumptions underpinning the incidents were also analysed individually, in focus groups and by the researchers.

Nineteen incidents were described from the perspective of the supervisor, eight were recollections from a student’s perspectives and in one incident the narrator’s perspective was impossible to determine. Even though a limited set of critical incidents has been analysed, some trends were clear: multiple supervisors are often the basis for conflict; task related problems are common; multi-disciplinary research may increase task related problems; the supervisors lack of scientific experience, especially in the student’s chosen area, can lead to complications; and supervisors often see student behaviour as a source of conflict. Of the twenty eight incidents that were analysed six were written by female narrators (21%), nineteen incidents were written by male participants (68%) and the gender of three narrators could not be inferred from the incidents (11%). The majority of the incidents relate to interaction between male-male protagonists, which probably reflect the distribution of supervisors and students at this university. Only one of the incidents describes a gender-related issue. A female supervisor feels uncomfortable as a male student indicates his liking for her and gives her presents; eventually she feels bribed by the male student. Given a recent survey on sexual harassment at the university this type of incident was under represented in our study. The reasons for this were a source of discussion in the focus groups.

We concluded from this study that writing about and analysing incidents in supervision increases one’s awareness of the process and problems associated with educating researchers. Participants in this project have also had the opportunity to recognize and reflect on conflict situations in each other’s critical incidents. This provides a broader perspective of putative problems in research supervision. With increased awareness and a larger portfolio of putative conflict situations, supervisors (and students) may earlier recognize problematic situations and issues. An early diagnosis (Tripp, 1996) can help prevent the escalation of small matters into severe conflicts. Thus the generation and analysis of critical incidents, even of such a limited set as represented in this research, is a powerful tool for gaining insight into factors that lead to successful supervision. Participation in such action research (judged on the basis of an evaluation of the workshops that was a part of this project) can also facilitate growth as a supervisor in postgraduate education. For students on the other hand, the availability of a collection of such narratives could increase their understanding into the complex task of a supervisor. Moreover, recognition of student-supervisor problems and the realization that one’s problems as a supervisor are not unique, can contribute to the well being of young supervisors. In the next cycle of this action research PhD students will also be included as informants. A third group that was not included in this study are senior researchers who formally act as main supervisors while doing very little practical supervision. Gaining their perspective via incidents or interviews would be of particular interest for further investigating one of the main categories that emerged in this study, namely, triangular conflict between student, co-supervisor and the main or nominal supervisor. This will be the focus of future research.

References

Flannagan, J (1954). The critical incident technique. Psychological Bulletin, vol 51.

Habermas, J (1972) Knowledge and Human Interest; Theory and Practice. tr. J. J. Shapiro, Heinemann, London.

Lewin, K (1946). Action research and minority problems. Journal of Social Issues, vol 2.

Tripp, D. (1993) Critical Incidents in Teaching, Developing Professional Judgment, Routledge, London.

Woolsey, L. (1986) The critical incident technique: an innovative qualitative method of research. Canadian Journal of Counseling, vol 20, no 2.