Initiating undergraduate medical students into communities of research practise: what do supervisors recommend?

Using the questionnaire design tool SurveyMonkey and comprehensive lists of contact details of staff who had supervised research projects at the University of Edinburgh during 1995 - 2008, current and previous supervisors were invited to recommend procedures which they had found successful in initiating students into the research culture of a team. Text responses were then coded in the form of derivative recommendations and categorized under general themes and sub-themes.Using the chi-square tests of linear trend and association, evidence was found for a positive trend towards more experienced supervisors offering responses (χ2 = 16.833, p < 0.0005, n = 215) while there was a lack of evidence of bias in the gender distribution of respondents (χ2 = 0.482, p = 0.487, n = 203), respectively. A total of 126 codes were extracted from the text responses of 65 respondents. These codes were simplified to form a complete list of 52 recommendations, which were in turn categorized under seven derivative overarching themes, the most highly represented themes being Connecting the student with others and Cultivating self-efficacy in research competence.Through the design of a coding frame for supervisor responses, a wealth of ideas has been captured to make communities of research practise effective mediums for undergraduate student learning. The majority of these recommendations are underpinned by educational theory and have the potential to take the learner beyond the stage of initiation to that of integration within their community of research practise.It has been recognized that the pursuit of new ventures is an integral part of human existence [1]. As Wenger explains, however, it is when these ventures are pursued effectively with others that participants may gradually assume particular types of practises and types of social interaction which ensure survival in their individual roles and progress towards their respective endpoints. In the current study, we understand this pr