This study has methodological limitations that must be taken into consideration when interpreting the findings. One cannot over-emphasise the limitations of self-report as this may limit the validity of findings. Respondents for various reasons may under, or overestimate their practice. A methodological problem frequently associated with the use of self-reports in questionnaires, which may have been evident in the present study, is the inability to determine the extent to which responses accurately reflect the respondents' experiences and expectations of their PBL tutorial sessions. This warrants further research to examine the actual PBL process. It is also possible that medical educators in this study were not representative of PBL educators.
The response rate was low, despite our efforts to maximise it and this means that the findings should be interpreted with caution. Reasons for non-response are not known. Non-respondents to the survey may also be less interested or involved in PBL, and therefore the reported extent of the PBL approach in this study may be higher than in reality.
Regarding forum repliers, this was a convenience sample consisting of only 6 medical educators. The online forum discussions were convenient and provided a transcribed record. Drawbacks to participation in online discussions may be the same as for online education in general, that is, the inability to capture the richness and depth of meaning without visual and verbal clues.
To overcome these methodological limitations we suggest, therefore, randomised experiments which focus on the performance of PBL graduates and non-PBL graduates in the clinical workplace. This may optimise the accuracy of inferences about the PBL approach. Clearly, an important task facing researchers is the identification and control of those factors that may give rise to alternative explanations for the effects of PBL compared to non-PBL methods. Factors such as the educational background of the students, methods of student selection and the learning culture of the institution are all potentially important. In addition perhaps more emphasis should be placed on researching the comparative learning processes that PBL and non-PBL students engage in. For example PBL students engage in considerably more verbal discourse, questioning and reasoning episodes than traditional students. Perhaps this develops additional cognitive and interpersonal skills not necessarily acquired to the same extent by more didactic and teacher-centred learning methods.
The descriptive analysis of this study showed that many participants valued the PBL approach in the practice and training of doctors. However, some medical education scholars held contrasting views upon on the importance of the PBL approach in undergraduate medical education. Among the medical educators surveyed, 38.5% had a neutral experience of PBL as a student-oriented educational approach. This finding is not consistent with the common characteristic of the PBL approach, indicating its student-centred nature [19
]. Although 46.2% of participants valued PBL as a student-oriented approach, the question that comes to mind is why do a group of medical educators feel so uncertain about it? Further research should examine this. What is surprising is that more than 61% of medical educators disagreed that the facilitator needs to be an expert in the subject matter of the case despite the fact that the majority of participants had a medical health professional qualification. The issue of content knowledge compared to process expertise is still challenging. Some evidence shows differences in favour of content experts when compared with process expertise [20
]. For example, Eagle et al
. concluded that twice as many learning issues were identified by groups led by content experts [21
]. Consistent with the results of these studies, Schmidt et al
concluded that students guided by subject experts spent more time on self-directed learning and achieved somewhat better scores on high stakes tests than students guided by non-expert facilitators [22
]. However, a study by Silver and Wilkerson indicated that content expertise resulted in more tutor-directed discussion in a PBL course [23
]. Taken together, these studies may suggest that both subject and process expertise are required by facilitators.
The results of this study indicate that the participants had a neutral view of the efficiency of traditional learning compared to a PBL tutorial. As such, participants had a neutral view of the claim that knowledge is better acquired in PBL-based course rather than a lecture-based one. These findings add to most previous research studies by demonstrating that there is no difference between the knowledge that PBL students and non-PBL students acquire about medical sciences [24
]. Although studies show that group learning in PBL may have positive effects, much more empirical evidence is needed to obtain deeper insight into the productive group learning of a PBL tutorial [25
]. One may argue that the process of PBL needs to be rigorously investigated in order to offer reasons for believing that it is designed to help student construct an extensive knowledge base and to become doctors dedicated to lifelong-learning. It is therefore important to further explore the nature of the learning acquired from PBL courses compared to traditional instruction courses.
With respect to graduate entry PBL, this study did not show that the policy of admitting graduates versus school-leavers to medical programmes was perceived as effective in creating better doctors. Interestingly, no previous PBL studies have explored differences between graduate entry PBL and school leaver programmes, although this study revealed that graduate entry PBL is not perceived as a more effective way of increasing the number of doctors in the UK by the majority of responders. In addition, this study revealed that there was a majority perception that graduate entry PBL will produce doctors who have come from a greater variety of educational backgrounds. However, will graduate entry PBL create better doctors compared to school leaver programmes? Sophisticated methodological approaches are required to answer this question.
The descriptions of medical educators about the PBL approach focused on the process of PBL, the characteristics of a good PBL facilitator and the advantages and disadvantages of PBL. It has been well documented that the facilitator role is central to PBL. The adoption of the role requires an understanding of epistemological and ontological issues about teaching and learning in medicine. In the epistemological sense PBL students are novices and the knowledge facilitator should assist them in restructuring new knowledge based on their prior declarative and procedural knowledge. In the ontological sense perceiving a new reality by students is important and the role of the facilitator is to assist students to explore reality in different ways. As the importance of faculty development in PBL was valued by participants in the forum discussion this may suggest more facilitator development workshops to help achieve competence as skilled facilitators of the PBL process. Such workshops may uncover conflicting roles of tutors in the steps of the PBL process. As Irby indicated, identifying and practicing these roles (mediator, challenger, negotiator, director, evaluator and listener) is a key skill of effective facilitation [26
In addition to this, one medical educator had a negative approach about PBL, and reflected: "PBL is still unclear in GEM
". It seems that some medical educators have negative perceptions of the ontological assumptions of PBL. For instance, a qualitative study was conducted to explore how a cohort of tutors made sense of PBL. In this study, one participant stated: "absolutely not, no views not really changed at all. I'm still not convinced that PBL, despite the fact that [I will tutor again] is the proper way of teaching"
]. Altogether these findings concerning academic achievement are slightly in favour of non-PBL programmes.
When asked about their experience in a PBL tutorial course, medical educators indicated they had few negative feelings with respect to facilitating self-directed learning and student learning. There are several possible reasons for this. Firstly, in the beginning of the course, it seems that the students find adopting a self-directed problem-based approach to learning difficult as they "do not know what they do not know". This may be attributed to the fact that students may have a restricted personal knowledge of the complexity of the "case". Secondly, students may not have clear objectives for the behaviour that they have to achieve, particularly in clinical settings, as mentioned by one participant. Thirdly, learning styles, both deep, surface and 'strategic', are determined at secondary school, and it is also difficult to influence learning styles even with a PBL curriculum [28
In this study, a few participants suggested combinations of pedagogical strategies, where several PBL courses are offered along with courses presented in a more traditional way. There is no evidence that indicates how a hybrid curriculum can make students better doctors compared to other approaches. However, a recent study concluded that changing curricula in medical education reform is not likely to have an impact in improvement in student achievement [17
]. The authors suggested that further work ought to focus on student characteristics and teacher characteristics such as teaching competency.