Introducing peer observation of teaching to GP teachers: a questionnaire study

Abstract

In medical education programmes which rely on clinical teachers spread across diverse sites, the application of peer observation of teaching offers the potential of both supporting teachers and maintaining quality. This paper reports on a questionnaire survey carried out with general practitioner (GP) teachers of medical undergraduate students from King's College London School of Medicine at Guy's, King's College and St Thomas’ Hospitals. The aim of the study was to determine GP teachers’ views on a proposed programme of peer observation of their teaching. The majority of GP teachers identified benefits of the proposed scheme with 69% saying it would help improve the education of future doctors. However, despite seeing the benefits, less than half wished to take part in the programme. Two thirds cited time and paperwork as major disincentives to taking part and 62% said that they felt it would make them feel under scrutiny. No associations were found between measures of workload and willingness to take part. This suggests that a fundamental fear of scrutiny and criticism may be the main hurdle to be overcome in implementing the scheme. Imposing peer observation on GP teachers in the form proposed could create suspicion and distance between the university department and practice-based GP teachers and may even result in a loss of teachers. The introduction of peer observation is more likely to be successful if GPs’ apprehensions are addressed. Using peer observation to strengthen the process of quality assurance may undermine its role in the support and development of clinical teachers.

Introduction

In 2001, United Kingdom (UK) Departments of General Practice and Primary Care and 3900 associated general practices contributed an average of 9% of all teaching in medical school curricula (SAPC, 2002). At King's College London School of Medicine at Guy's, King's College and St Thomas’ Hospitals (KCL), we have a network of some 600 primary care teachers who teach 14% of our medical school curriculum to our 2000 students. Eighty percent of these are general practitioners (GPs) from 250 teaching practices. Sixty five percent of the practices are in southeast London; the remaining 35% are scattered throughout southeast England and other parts of the UK. The numbers involved and the geographical dispersion of these teachers present a complex challenge when it comes to supporting them, delivering a consistent curriculum and ensuring that teaching is of a high quality. To help meet this challenge, we plan to introduce a new system of peer observation of teaching (White & Stephenson, 2000).

Table

Practice points

• The majority of GP teachers believed peer observation would help them to reflect more on their teaching, to set clearer learning goals with students, and to try out new teaching methods.

• Only half of teachers however would wish to take part in peer observation. Main obstacles were given as time and workload but widespread anxiety about scrutiny was apparent. Teachers were questioning of the motives of the university department.

• Peer observation was designed to meet the twin aims of teacher development and quality assurance. Teachers’ views suggest these two aims may conflict.

• By imposing peer observation, we could create distance and suspicion between the university department and practice-based teachers.

• Teachers’ fears about scrutiny must be addressed and we need to either redefine the role of the observation in relation to the quality assurance process or use it for teacher development only.

The proposed system is consistent with, and complementary to, other significant developments in this field. The UK Government Department for Education and Skills (DFES) white paper of 2003 (DFES, 2003) and the Higher Education Academy (HEA) consultation paper of 2004 (Universities UK, 2004) recommend that, from 2006, all new teachers in higher education will be required to attend accredited teaching programmes. Peer observation of teaching and training is a key element of these programmes and is already widely used in UK and international institutes of higher education (http://www.heacademy.ac.uk/988.htm). In a further development that will encompass these plans, the Strategic Learning and Research (StLaR) Human Resources Plan Project, commissioned by the StLaR Advisory Group, is developing a framework for the guidance, maintenance and development of the training and educational arms of health, social care and education organisations (StLaR, 2004).

The system of peer observation we are introducing, and which is the subject of this paper, was originally called Supervised Teaching Practice (STP) (White & Stephenson, 2000). We have now adopted the term ‘peer observation’ as this is in line with current educational terminology. The implications of the previous use of the term STP are addressed further in the discussion. The goal of peer observation is the development of the teachers’ capacity to reflect on their teaching. This is achieved by sharing the task of reflecting on process, by identifying principles underlying the teaching process and by identifying examples of good practice and areas for development. The key components of the system are described in Figure 1. The system would be centrally directed and monitored but would be delivered primarily by community teachers, trained and supported in their peer observer role.

Figure 1. The stages of peer observation.

There is a gap in our knowledge about the transferability of peer observation of teaching to settings outside the university and particularly its acceptability to those professionals for whom teaching is only a small part of their workload. In the case of our GP teachers, the competing demands of the 2003 NHS contract for GPs, together with their other professional responsibilities, present potential obstacles to the introduction of this innovation. We report here on the results of a survey of GP teachers’ views of teaching observation and feedback, of their willingness to take part in such a programme and of the opportunities and challenges it would present. By identifying these, we aim to design a peer-observation programme that can best meet the needs of this particular group of teachers.

Method

We conducted a mailed questionnaire survey of the 481 active undergraduate GP teachers contracted to the Department of General Practice and Primary Care at KCL to provide undergraduate teaching in any of the five years of the MBBS curriculum. Our aim was to assess the acceptability of peer observation (based on a written description) to teachers outside the university department and to identify factors that would influence its implementation.

We drafted a questionnaire on the willingness of GPs to take part in peer observation of teaching, their views and experience of peer observation, their personal and workplace characteristics and their learning styles. To test the relevance of the questionnaire, and to identify the issues that were most important to the GP teachers, we ran a focus group.

Twenty GP teachers were invited to the focus group. They were chosen to represent the diversity of teachers within the department in terms of gender, range of experience, length of time teaching for the department, year groups taught and location and size of practice. Seven teachers agreed to take part and were asked to complete the draft questionnaire before the focus group. The discussion was facilitated by an experienced group-leader, audiorecorded and transcribed. A member of the research team (LA) made notes during the meeting. Discussion addressed the concept of peer observation (called supervised teaching practice or STP in the focus group and questionnaire), its benefits, the obstacles to it and the implications for GP teachers. Each section of the questionnaire was then discussed in turn considering content, clarity and ease of completion. The questionnaire was amended in response to the focus group comments and piloted with ten GP teachers in another London medical school.

The final questionnaire assessed the personal, workplace and workload characteristics of the GP teachers, their previous experience of teaching and mentoring and their teaching styles using Prosser & Trigwell's (1999) Approaches to Teaching Inventory (to be discussed in a future paper). The acceptability of peer observation was assessed through the GP teachers’ responses to a list of component tasks (e.g., ‘observed teaching live’) recorded on a five point Likert scale (totally ill at ease to totally at ease). A Likert scale (totally disagree to totally agree) was also used to record the teachers’ responses to a series of statements on the feasibility and usefulness of peer observation (e.g., ‘peer observation would encourage me to try out new teaching methods’). Respondents were asked in an open question to cite the main reasons why they would or would not wish to take part in peer observation. The questionnaire was mailed to GP teachers with follow up reminders to non-responders at three and six weeks.

Data analysis

Qualitative data generated by the focus group were transcribed and read closely by two members of the team (LA, PW) who, first separately and then in discussion, agreed a list of key themes. The qualitative data from the open questions in the questionnaire survey were then added to this framework and the data from the two sources were considered together.

Quantitative data from the questionnaire survey were processed and analysed using version 11.5 of SPSS (SPSS Inc., Chicago). Associations (univariate and multivariate logistic regression) were sought between the reported willingness of GPs to undertake peer observation and their personal, workplace and workload characteristics, their previous experience of teaching, mentoring and supervision, their attitudes to the processes of peer observation and their perceptions of the opportunities and challenges of peer review.

Results

Three hundred and sixteen GP teachers responded (65.7%). Fify-nine percent were male, the mean age of respondents was 46 years (range 28–65), 62% worked eight or more clinical sessions per week, 53% came from postgraduate GP training practices, 46% had three years’ or less experience of teaching, 37% came from practices of six or more partners, 20% were trainers, 19% had qualified overseas and 8% were single-handed practitioners.

The benefits of peer observation of teaching

There was a broad consensus amongst GP teachers about the potential benefits of peer observation both for the students and for the teachers (Table 1). These benefits included the identification of clearer learning goals with the students, more reflection on their teaching and encouragement to try out new teaching methods. The majority of teachers (72%) agreed that STP would provide a way of addressing problems in their teaching. Forty-eight percent of all respondents (69% of GPs who were in single-handed practice) agreed that peer observation would decrease their isolation as a teacher.

Table 1.  GP teachers’ views of the potential benefits and costs of supervised teaching practice

Statements about supervised teaching practice	Agreed somewhat or totally %	Neither agree nor disagree %	Disagreed somewhat or totally %
STP would increase my confidence as a teacher (n = 303)	56	31	13
STP would not add to what I do now (n = 302)	14	24	62
STP would encourage clearer learning goals with the students (n = 302)	73	19	8
STP would take up too much time (n = 304)	66	24	10
STP would help me reflect more on teaching (n = 303)	75	18	5
STP would help improve the education of future doctors (n = 304)	69	24	8
STP would encourage me to try new teaching methods (n = 304)	71	25	8
STP would make me feel under scrutiny (n = 305)	62	22	16
STP would provide a way of addressing problems in my teaching (n = 303)	72	21	7
STP would decrease my isolation as a teacher (n = 302)	48	32	19
STP would involve me in too much paperwork (n = 303)	66	28	7
STP would provide me with a good written record of my teaching (n = 302)	59	35	7
STP would require skills I don't yet have (n = 301)	32	45	23

The perceived benefits of peer observation were reflected in individual statements made in the questionnaire open questions and the focus group.

• A desire to know more about their teaching skills:

  ‘I think that some form of assessment [of teaching] is good from my own point of view. Because I don’t know if I am doing it right … I don’t know if I do it better than anybody else.’ (GP 1, focus group.)

• A desire to improve the quality of their teaching:

  ‘Always useful to have someone else's observations; would probably make me more learner centred and would probably make me reflect more on my teaching work.’ (GP No. 63, questionnaire.)

  ‘Make me more critical of my teaching.’ (GP No. 231, questionnaire.)

• A desire for interaction with other GP teachers:

  ‘STP would improve the relationship amongst various teachers.’ (GP No. 254, questionnaire.)

  ‘I don’t feel isolated but STP would make me feel part of the local teaching community.’ (GP No. 348, questionnaire.)

  ‘Chance of peer review of teaching broadens horizons—part of a team.’ (GP No. 266, questionnaire.)

Willingness to take part in peer observation

Whilst there was a very high level of agreement about the need for, and the benefits of, peer observation, more than half of the GP teachers were not yet ready to commit to the programme. One hundred and forty six (49%) of the teachers wished to take part in peer observation, 136 (45%) did not wish to take part and 18 (6%) were undecided. The decision to take part in peer observation was independently associated with working in a single handed or two partner practice (OR 2.61, 95%CI 1.07–6.43), working in a postgraduate GP training practice (OR 2.24, 95%CI 1.15–4.35), having no shortage of space in the surgery (OR 2.73, 95%CI 1.43–5.22) and with having received feedback on a personal development plan that was perceived as helpful (OR 4.8, 95%CI 1.24–18.55). Table 2 shows GP teachers’ views about how at ease they felt with the main activities of peer observation.

Table 2.  GP teachers’ views about how at ease they felt with the main activities of STP

	Degree of ease with activities of STP
Activity of STP	Fairly or totally ill at ease %	Neither at ease or ill at ease %	Fairly or totally at ease %
Being observed teaching live (n = 305)	25	23	52
Observing another tutor teaching live (n = 305)	7	20	73
Videotaping own teaching session (n = 305)	19	20	61
Observing another tutor's videotaped teaching session (n = 305)	3	18	79
Audiotaping own teaching session (n = 305)	19	25	57
Reviewing another tutor's audiotaped teaching session (n = 305)	9	23	68
Recording own teaching session in writing (n = 304)	12	25	62
Reviewing another tutor's written record of a teaching session (n = 302)	8	25	67
Receiving feedback on own teaching (n = 303)	5	19	75
Giving feedback on another tutor's teaching (n = 304)	11	27	61
Receiving feedback on own teaching from an educationalist (n = 303)	12	15	73
Having a written assessment of teaching provided to the department (n = 301)	14	28	58
Providing a written assessment of another tutor's teaching to the department (n = 303)	24	35	40

Obstacles to taking part in peer observation

Sixty-two percent of respondents agreed somewhat or totally with the statement that peer observation (STP) would make them feel under scrutiny. The teachers’ decision not to undertake peer observation was independently associated (multi-variate logistic regression) with unease with the processes of peer observation in terms of ‘being ill at ease video recording a teaching session’ (OR 3.73, 95% CI 1.79–7.77) and ‘being ill at ease receiving feedback on teaching from an educationalist’ (OR 3.21, 95% CI 1. 20–8.60), but not from a peer.

Two thirds of GP teachers (66%) agreed that peer observation would take up too much time and involve too much paperwork. However, we found no significant associations (univariate and multivariate logistic regression) between GP teachers’ reports of high workload (number of clinical sessions per week, number of patients seen in morning and afternoon/evening surgeries, number of patients per whole time equivalent partner, length of appointments, type of on-call system, number of out of hours sessions, number of students taught) and their decision whether or not to take part in peer observation.

Disquiet about the impact of peer observation was reflected in individual statements in the open questions of the questionnaire and in the focus group:

• Time pressures:

  ‘In a job where I feel I barely keep my head above water, I feel (at the present time) supervised teaching practice would feel like an extra burden rather than a useful tool.’ (GP No. 65, questionnaire.)

  ‘I am so overworked and stressed anything else will tip me over the edge.’ (GP No. 279, questionnaire.)

• Being observed in their role of teacher:

  ‘Big Brother is watching you scenario.’ (GP No. 352, questionnaire.)

  ‘Find being observed threatening, always have and find that I invariably do things differently.’ (GP No. 60, questionnaire.)

• The motives of the medical school department:

  ‘What has it come to replace anyway? It might be that you are doing quality control. But have there been complaints about teachers.’ (GP 2, focus group.)

  ‘… you feel there is a different agenda … .’ (GP 4, focus group.)

• Risk of losing teachers with this initiative:

  ‘Would feel ill at ease, would probably put me off teaching’ (GP No. 243, questionnaire.)

  ‘… my impression is that you would lose a lot of tutors through this … I think the medical school has to remember that we don’t have to teach’ (GP 5, focus group.)

Discussion

Peer observation of teaching was seen by most teachers as likely to encourage clearer learning goals with students, to help them reflect more on their teaching and to help improve the education of future doctors. However, these benefits were insufficient to persuade more than half of them to commit to peer observation. Although a majority of GPs identified time constraints as the main obstacle to participation, unease about being observed or video recorded in a teaching session was also very evident.

The issue of time appears to be complex. There was no association between measures of workload and a decision not to take part—the busiest GPs were no more or less reluctant to take on peer observation than GPs with lesser workloads. Whilst it is undeniable that the perception of workload and the reality of it are often not related, it is possible that in citing time and workload these GPs may have been drawing attention to what was already a major issue—the stress they experience from being GPs and the need to manage competing demands on them. Sandars & Boreham (2002) have drawn the analogy of the community tutor as juggler. This seems to match the experience of some of our teachers. Stress, anxiety and depression have been significant problems amongst GPs (Caplan, 1994; Rout, 1999; Calnan et al., 2001). High workload, inadequate remuneration, threats to clinical autonomy, fear of complaints and the unrealistically high expectations of others, have variously emerged as job related stressors within general practice (Appleton et al., 1998; Sibbald et al., 2000; 2003). It was clear that some GP teachers saw peer observation as mirroring and adding to these general stressors.

The lack of association between time (and workload) and the reported willingness (or unwillingness) of the GP teachers to take part in the initiative may also be explained by the GPs’ usage of time and workload as shorthand for different concerns that are more difficult to acknowledge. Peer observation would make many feel under scrutiny. Unease with being videoed and receiving feedback from an educationalist were associated with unwillingness to take part in peer observation. The quotes illustrate the depth of anxiety about outside scrutiny for some of the teachers. This unease strikes at the heart of peer observation whose primary aim is to support and develop teachers.

These observations emphasise the ‘internal’ factors that impact on stress, including low levels of confidence, and fear of criticism, especially in a work environment perceived to have considerable external job-related stressors. Firth-Cozens (1997) in a ten year follow up survey of stress in general practitioners found that a high level of self-criticism as a student was highly correlated with current stress levels. Lack of confidence has been identified as a significant issue in relation to GPs’ perception of their teaching skills (Gray & Fine, 1997; Hartley et al., 1999; Sandars & Boreham, 2002). When confidence is low, it may be that expectations of criticism increase. There is a danger that processes designed for support and development will be seen negatively in terms of being for quality assurance and then primarily as instruments to identify failing tutors. The focus group's questioning of departmental motives was reflected subsequently in the concerns of questionnaire respondents.

This form of peer review was originally designed to meet the twin aims of teacher development and quality assurance. Teachers’ reactions suggest a conflict between the two aims. This scheme is unlikely to succeed if it is seen to be conveying quality assurance in the guise of tutor support. Such a scheme will need to be presented and organised in a way that builds confidence and erodes fears. The motives of the university will need to be transparent in terms of using peer observation for teacher assessment and accreditation. The role of observation will be put at risk if it is seen to be allied to a form of quality policing. It may be preferable not to use peer observation in tandem with the quality assurance process.

Limitations of the research

We did not use a previously validated questionnaire though the questionnaire appears to have high face validity. The response rate of 67% was high for this type of mailed survey to GPs, and the question completion rate was also high at 94%. We suspect that the name ‘Supervised Teaching Practice’ itself had a negative impact on the responses of GP teachers’ in the questionnaire; it may have emphasized the supervisory aspect of the initiative at the expense of its supportive elements and led in turn to an anxiety response in some participants. Replacing the word ‘supervised’ with ‘peer observation’ may elicit a different response from GPs. The views of the GP teachers were obtained in response to a written description of the process of peer observation and may be different from the views of GP teachers obtained after participation in an actual model of peer observation.

Conclusions

Imposing peer observation on GP teachers in the form proposed could create suspicion and distance between the university department and practice-based GP teachers and may even result in a loss of teachers.

Although time and workload are the apparent obstacles to the implementation, a fundamental fear of scrutiny and criticism may be the main hurdle to overcome in getting teachers to take on peer observation. The introduction of peer observation is more likely to be successful if GPs’ apprehensions are addressed.

Acknowledgments

This study was funded by the Guy's and St Thomas’ Charitable Foundation. We are grateful to Virginia Morley who ran the focus group, to Anne and Matthew Bracewell who helped with the data processing and analysis, and to the general practitioners who attended the focus group and completed the questionnaires.

Additional information

Notes on contributors

Lesley Adshead

LESLEY ADSHEAD, BA, MSc, CQSW, was a Research Associate in the Department of General Practice and Primary Care, King's College London School of Medicine at Guy's, King's College and St Thomas’ Hospitals, UK at the time of this research. She works as a part time lecturer in social work, teaching specialist palliative care and is doing doctoral research at King's College London.

Patrick T. White

PATRICK WHITE, MBChB, BAO, MRCP, FRCGP, is a Clinical Senior Lecturer in the Department of General Practice and Primary Care, King's College London School of Medicine at Guy's, King's College and St Thomas’ Hospitals, UK, and a general practitioner in south London. His research and teaching interests are in chronic disease management.

Anne Stephenson

ANNE STEPHENSON, MBChB, PhD, ILTM, is Senior Lecturer in General Practice and Director of Community Education at the Department of General Practice and Primary Care, King's College London School of Medicine at Guy's, King's College and St Thomas’ Hospitals, UK. She also works as a general practitioner in south London. Her research and teaching interests are in professional development.