In total, 16 GTAs were interviewed for approximately 20 minutes each (postdoctoral tutors were not interviewed). The interviews were recorded, but all data remain anonymous. The participants all volunteered their time in response to an e-mail request, and the interviews took place in November 2006. No incentive was offered for participation, but the number of participants (16) represented a positive response rate of 75 per cent. The teaching experience of participants varied from several years to new GTAs in their first semester of teaching. Five of the interviewees were in their first year of teaching, six were in their second year and the remaining five had already had at least two full years in the role.
The structure of the interviews led to a broad range of ideas and issues being discussed. However, the interview data coalesced around two main areas: the role of a seminar leader, how this is understood and what it involves; and training and support issues related to this role. The following two sections report the results of the interviews.
Being a seminar leader
There was general agreement among the GTAs that the role of a seminar leader should be as a facilitator rather than as a teacher, but there were widely differing interpretations of what this meant. ‘I'm not there to teach but to help students learn’, was a stock phrase, variations of which were used by virtually all participants. This perhaps reflects the fact that it is a common piece of advice from module leaders to GTAs. The relatively small amount of contact time seminar leaders have with their students (typically one hour per week) was also regularly cited to support the contention that seminars are not a time for teaching students ‘everything they need to know to pass the exam’ but for encouraging them to develop their ideas and critical thinking skills. However, understanding of ‘facilitation’ varied widely. One GTA described facilitating learning as providing ‘a very clear structure – a framework within which students can learn’. This was based on the belief that students require a clearly structured environment in order to learn, and that it was therefore the duty of the seminar leader to provide one. This involves making what is required from students explicitly clear, and appealing to their self-interest by making their priorities your own:
‘I see my role as a facilitator, rather than someone who provides the information, I see myself as someone who enables them to access the information, and form opinions and be able to analyse the information ... At the end of the day they are not assessed on their seminar contributions, they are assessed on their essays and book reviews, so that's really where their efforts are going to be targeted – as much as they should be doing everything – that is where their efforts are targeted, so you might as well accept that and help them get the best marks they can’.
In this model, the focus is almost entirely on the course content. The assessment for the course is regularly referred to in seminars, and the way in which preparing for seminars helps students succeed in the assessed work is emphasised. Useful sources are highlighted, and handouts detailing what is expected and required in order to do well in the assessed work are provided. This ‘carrot and stick’ approach, this GTA believed, leads to students being well prepared for seminars, and therefore a better quality of seminar. The duty of the seminar leader is also to be well prepared, so he or she can actively lead the seminar and thereby provide a structure that covers all the necessary material.
At the other end of the spectrum, facilitation was regarded in a much more ‘hands-off’ way. One interviewee imagined a ‘perfect’ seminar to be one ‘in which I do as little as possible ... which means that everyone would have to feel confident enough to participate’. They argued that in first-year seminars in particular, the development of useful academic skills is more important than ensuring that every aspect of the course content is covered. By socialising new students into the academic world, you are providing them with a valuable service that will set them in good stead for the remainder of their academic career.
‘When I first started teaching I think I approached it very much from the perspective of getting the knowledge across – that is, the content of the course. Over time I have changed it so that now I see it much more about skills– so I get them to work in groups, take control of the seminar, lead debates, and so on. The knowledge comes out through the seminars, but it is very much skills-focused rather than knowledge-focused ... If they get the skills the knowledge will follow I think ... It is also the skills which are more important: employers want skilled graduates’.
An ideal seminar would therefore be one in which students would lead the seminar themselves and learn through having to organise independently their own learning experience. While the discussion generated may not reach such depth as in a more teacher-led environment, it was felt that the benefits of getting students to learn independently would outweigh this.
GTAs' understanding of the purpose of seminars and their role as a seminar leader greatly influenced how they organised and structured their groups. Those who emphasised content and knowledge tended to have a clearer plan or idea about how the seminar should proceed, to ensure the necessary material was covered. Some suggested that this may necessitate curtailing discussions or particular activities after fixed periods of time, because of the need to ‘move on to the next thing’ and get through the full plan of the seminar. The GTAs who put a greater emphasis on skills also used seminar plans, but these were based more around different activities, related to, but not entirely driven by, the course content. For example, small-group work was cited as a means not just to cover a particular topic, but to ‘get all students participating – it's much harder not to talk when you are in a group of three’– and to develop academic skills: ‘students are more willing to challenge each other's ideas and debate in smaller groups’. These GTAs were also more willing to let a particular activity overrun if they thought it was going well, and for the content of the seminar to be determined more by the students. One noted how they try to manage discussions:
‘Module leaders appreciate that there is no way that you can cover all angles of a seminar topic in the space of 50 minutes. If the discussion was straying too far away from the topic (even if it was interesting) I'd probably try and pull it back or raise one of the seminar questions that we hadn't covered – it's important not to let people go off at too much of a tangent – but equally you don't want to squash novel ideas or different lines of thought’.
This quote demonstrates how in reality most GTAs sought to strike a balance between seminar activities and skills and breadth of content coverage. Those who professed to believe in an active learning process driven by student involvement were also anxious to ensure that course content was covered, and the debates and discussions reached a sufficiently sophisticated level. Similarly, GTAs who claimed to be largely concerned with ‘covering the course’ also highlighted the importance of student involvement. As such, sacrificing discussion of some suggested seminar questions was seen by one as an ‘acceptable compromise’ if others were dealt with in greater depth or provoked particularly interesting discussion. Preparation and participation on the part of students were cited by all as the key to successful seminars. Knowing when to intervene, when to join a discussion and how to manage a group (particularly students who are either dominating a group or unwilling to participate) were seen as vital skills that a seminar leader needs, and which are largely derived from experience.
Experience (and with it, confidence) appears to be a significant influence on how tutors approach seminars. As a general rule (although there were a couple of exceptions to this) less experienced tutors were more concerned with covering as much of the course as possible in their seminars, and were more anxious about being ‘caught out’ by a student, and not being sufficiently knowledgeable themselves. More experienced tutors were less concerned about this, as one commented:
‘They [the students] do sometimes seem to think that you do have all the answers, which is not necessarily the case! I think it's a good skill that you learn that if you haven't got the answer, don't try and waffle your way through it, just say “no, I don't know the answer but I'll come back with it next week”, or say “what does everyone else think?” ’
Some GTAs also explained their tendency to over-prepare for seminars and focus heavily on course content in terms of a ‘fear’ that students would not be prepared, and would therefore be unwilling to participate. Again this was more prevalent among less experienced tutors. One described how it was difficult to apply their ideas about the teaching and learning experience in practice:
‘I'm trying to excite students, to create an environment where there's a meeting of minds. The seminar idea to me seems to be about teamwork and sharing and benefiting from each other's knowledge, creating a dialogue and debate, and getting different points of view to come out ... and that I think is very important, as so much of the learning process is very individualised and competitive. So I'm interested in encouraging co-operation between learners’.
In practice, however, this interviewee acknowledged that an ‘individualised and competitive’ form of learning was heavily ingrained in many students, and they were therefore not initially very responsive to this more inclusive and co-operative approach. Consequently, content and assessment tended to dominate seminars, as these were the primary concern of the students. In part this is a reflection of the learning experience most students have had at school or college, and part of the challenge of teaching new undergraduates in particular is socialising them into the university learning environment (Leamnson, 1999).
Small-group teaching has been described as the ‘engine room of instruction’ in politics (Wood and Moran, 1994), and is widely recognised as a critical part of the university learning experience (Newble and Cannon, 1995; Ramsden, 2003). As GTAs are at the forefront of this delivery, how they choose to organise their seminars is of great significance. The variety of techniques employed by the interviewees again reflected their pedagogical understanding and level of experience as tutors. Group work was mentioned by most of the interviewees as a method they had used in their seminars, varying from short ‘buzz groups’ to formalised group activities and presentations which may take place over a number of seminars. Typically, those GTAs primarily concerned with covering the course content seemed to have the most rigid seminar plans, while more experienced tutors discussed their willingness to adapt seminar structure according to how they ‘felt it was going at the time’. The majority of interviewees noted that they were relatively free to structure seminars how they wished, and welcomed this flexibility as an expression of confidence by module leaders in the ability of tutors to deliver seminar programmes in their own way.
Training and support for GTAs
Previous research has considered the issue of training and development of GTAs in higher education. For example Peg Boyle and Bob Boice (1998) have suggested the development of systematic mentoring for new teachers, and Kim Korinek, Judith Howard and George Bridges (1999) have proposed a ‘developmentally based’ programme for GTA training in sociology. On the question of whether or not GTAs should face a more comprehensive compulsory training programme (either before beginning to teach or during their time as a GTA in the department), opinion was distinctly mixed. Approximately half the interviewees were against such a programme (mainly citing time pressures) although there was a greater willingness to accept the idea if the training was paid. There was also a widespread feeling that although the department's training requirements were relatively light, the system generally worked quite well and the quality of GTAs and the teaching they delivered was quite high.
In this respect the department is lucky in that it attracts a sizeable pool of talented research students from which to draw tutors. Perhaps surprisingly, two interviewees expressed concerns about the quality of some of their fellow tutors, and suggested that the department's quality control mechanisms are insufficient. They suggested that the interests of GTAs were elevated above those of undergraduate students, in that Ph.D. students were generally given the opportunity to teach without first being subject to a formalised assessment of their ability to do so effectively. The result could be some students getting a ‘poor deal’, particularly in the context of higher tuition fees and the changing student demands that have resulted from reduced financial support (Rolfe, 2002). However, it should be noted that this was very much a minority viewpoint, being raised by just two of the sixteen interviewees.
The main quality control mechanism in the department is the teaching observation. All GTAs are observed annually (as are all teaching staff), usually by the module leader of one of the courses they teach. One commented that ‘It's a useful process, particularly quite early on. Things were pointed out to me that I hadn't really thought of, but now seem quite obvious! So I think my teaching has improved as a result of it’. This position was supported by the majority of interviewees. None objected to the idea of this form of peer observation in principle, and most were strongly in favour of it as a way of finding out if their teaching was up to a sufficient standard, and as their only real feedback mechanism. However, specific problems with the process were raised. Several GTAs felt that the process was one with inbuilt diminishing returns: it became less useful each time it occurred. This is probably partly a reflection of their own greater experience as tutors – as they became more experienced, they would have less need for basic ‘pointers’ or seminar room ‘tips’, as they should already be armed with a greater repertoire. Three interviewees also pointed out that if they taught the same module for two or three years running, the result could be that they were observed by the same person (the module leader) each year. This was generally felt to be less useful than the opportunity to be observed by someone else and gain a different perspective. The timing of the observation was also a potential issue. Too early, and a new tutor might be greatly unnerved by the process, too late and input that might have benefited the tutor earlier was unnecessarily delayed.
Related to the issue of peer observation is that of the teaching mentor. Several GTAs suggested that a more formal system of mentoring may be a useful innovation:
‘I think it would be good if we all had a mentor who wasn't our Ph.D. supervisor, but was someone who you could talk about your teaching with, and if there was a proper system of feedback, assessment and review of teaching progress each semester ... this could help people improve their performance’.
At present, the first point of contact for a GTA in the department for teaching matters is the module leader of whichever module(s) they are currently teaching on. The advantage of this is that the module leader has a close and direct interest in the teaching of his or her module; the potential disadvantage from the perspective of the GTA is that the relationship is inevitably focused on module delivery. As such, the focus on the personal development of the GTA as a teacher may be limited. A mentoring system – where a GTA retained the same mentor throughout his or her employment in the department, might lead to a more structured system of personal development more akin to that available to full-time staff.