Peer feedback needs practice and faculty support to be effective in simulation based education.

Background Peer evaluation drives effective self-assessment, encourages autonomous learning and enhances both metacognitive skills and critical reasoning skills. 1 Students, by commenting on the work of others, gain a better understanding of the criteria required for successful performance and develop skills of objective judgment which can be transferred to the assessment of their own work. 2 Methods Peer feedback was introduced to simulated patient history taking consultations. Students were required to evaluate the observed performance and provide verbal feedback to their peers. This was compared to facility review and end of term examination results. In interview students indicated that they benefited from the opportunity to engage in peer feedback. Students reported that they felt more comfortable receiving feedback from peers than providing feedback to peers. Results 65% of students rated peer performance as excellent which did not correlate with summative OSCE results. When students did mark the borderline grade, a significant difference was found for one individual element of the feedback score. Students ticked "lacked confidence/fluidity" significantly more frequently than any other of the 7 elements suggesting they felt more comfortable relaying this element of performance feedback as it did not address content understanding or knowledge. Conclusion Ability to give constructive feedback should be viewed as an essential skill but for success students need to be taught how to give effective peer feedback. Emotions and loyalties affect student’s unwillingness to find fault with a fellow student’s work. Students may need more opportunities to conduct peer assessment to become familiar and comfortable with this process. Co-operative learning in simulation can combine with peer feedback to produce effective social constructivist approaches; however faculty input is required to monitor and validate the feedback.


Background
Medical education is rapidly evolving and the last decade has seen major changes in terms of how medical doctors are educated with undergraduate medical curricula evolving from predominantly didactic, discipline-based formats to case based, patient centered, integrated and centrally administered programs. New teaching methods adopted in response to information overload on traditional medical courses employ small group experiential, and facilitated learning instruction methods which incorporate an increased emphasis on self-directed learning.
Simulation based education (SBE) is now a central thread in the fabric of health professional education 1 having gained tremendous momentum over the last decade as a powerful method of reinforcing clinical knowledge, improving communication, and teaching decision-making skills.
Learners reach competence through deliberate and repeated practice, breaking down skills into parts and practicing these parts repeatedly whilst eliciting feedback and reflecting on performance to guide the focus of subsequent practice. Success requires students to acquire reflective learning skills, to assess accurately their own strengths and weaknesses and to take responsibility for their own professional development. Educators are increasingly encouraging their students to take responsibility for not only their own learning but also for that of their peers. The peer evaluation process is suggested to drive effective self-assessment and encourage autonomous learning. Students, by commenting on the work of others, gain understanding of the quality criteria required for successful performance and develop skills of objective judgment which can be transferred to the assessment of their own work 2, 3 . By participating in peer assessment students gain enhanced metacognitive skills and critical reasoning skills 4 .

4
This paper addresses collaborative learning in small groups with simulated patients (SPs) and examines the effect of peer feedback in creating collaborative learning groups for SBE.
The key objectives of this work are: To understand what opportunities students require to become familiar, comfortable and effective with the peer evaluation and feedback process and to discuss issues that restrict full participation.
To examine how simulation learning might be enhanced to make the experience more purposeful and valuable for observers, and help them contribute more confidently and specifically to feedback discussions.
To understand what tools are required to help students observing their peers to become better attuned to what is going on and at noticing what is important.

Study design
This study was designed as a sequential transformative mixed-methods approach employing constructivist grounded theory 5 . A realist evaluation approach was taken focusing on causal mechanisms and exploring the necessary conditions for success.

Setting
The study was carried out with third year medical students at the Royal College of Approximately seven weeks later upon completed of their clinical attachment each student attended a record of in-year performance (RiYP) meetings where they were asked to reflect on their experiences of the peer evaluation and their perceived usefulness of these sessions.

Data Variables
The completed peer evaluation forms were compared with faculty evaluation of the same student SP encounters using the same assessment forms. The summative marks for history taking at the end of term exam are also presented. In addition, results from the course evaluation questionnaire are presented.
Two questions were asked in semi-structured face-to-face interviews conducted by CC with all 254 students conducted as part of the student's record of in-year performance (RiYP) meetings. This was a five minute commentary at the end of an in formal feedback 6 meeting.
Three follow on focus groups with a total of 14 students were carried out (CC, AA) to further explore key concepts using questions constructed around the themes identified in the interviews. Students were invited by email to volunteer to attend these focus groups.
Students were provided a participant information leaflet that explained the reasons why the study was being carried out, by whom and how. Participants signed a consent form and attended one focus groups which took on average forty five minutes to complete

Bias
Selection bias maybe considered a limitation of this study in that students volunteered for the focus groups. Voluntary participation was a condition of our study ethics permission.
However, the full cohort of students did participate in individual interviews as part of their course. The principle reason for this work is to understand how learners cooperate in simulation to drive learning and as a simulation specialist and principle investigator in this work CC may unconsciously add an element outcome-reporting bias. Participants had two scheduled classes with CC prior to the interviews and focus groups and knew her as their teacher. AA was a class mate of one cohort of students.

Analysis
The interview answers were captured by note taking and focus groups were fully recorded by voice recorder. Transcripts were not returned to the participants and participants did not provide feedback on the findings. Data was anonymized and although quotes are

Results
From the course evaluations the majority of students agreed that simulation practice enhanced their consultation skills and clinical application of their medical knowledge.
Students agreed that the simulated sessions provided a realistic and safe environment to practice communication and history taking skills ( Figure 1).
From interviews and focus group discussions, key descriptive themes were deduced several of which have been previously been identified 10 ( Table 2)

Peer feedback is valued
Students recognized value in peer feedback and indicated that they benefited from receiving peer feedback.
"Yes it is really important because there is something that you do not notice that you do, that other people can see, then you get an idea about the bigger picture rather than the thing that you are focusing on" Students appreciated the opinions of their peers in the main and reported that having completed the exercise, they are more likely to practice history taking presentations at the hospital site with a colleague and to request peer feedback.

Nervous and embarrassed
8 Students generally value anonymity and some students reported that they were nervous and embarrassed having to perform in front of their colleagues.
"There is the nerves when you walk into a room and you are like ummmmmmm?" Students reported that they are more comfortable receiving feedback from peers than providing feedback to peers. Giving peer feedback was an unfamiliar and stressful activity for the students and the video recording of the simulated sessions added further to their stress.
"got stressed -like you wouldn't imagine, honestly"

Emotions and loyalties
Emotions and loyalties affect student's unwillingness to find fault with a fellow student's work. Students are reluctant to damage personal relationships, and they struggle with the prospect of embarrassing or insulting their peers whom they have to continue to study with for the rest of their program.
"It's not worth it".
"I do not want to offend them and then be stuck with them for whatever amount of time and then they're gonna hate me" "Also when we first met everyone in the peer group no one knew each other. We did not know each other so everyone was afraid to say what they want to say as advice"

Preparation and skills required to give feedback
The overarching theme that emerged was that students felt unprepared to provide good quality feedback. There was significant discussion about the quantity, standardization and emphasis of the feedback. Students also demonstrated significant concern for the wellbeing of other students as a result of poorly constructed feedback.
"You do not want to discourage them at all. Maybe they are better in other situation"

Faculty involvement
Attitudes towards peer feedback varied depending on a student's previous educational experiences and culture. The belief that formal education is about learning from experts is deep-rooted in the student psyche. The theme of faculty oversight or requiring more faculty input was recurring.
"We went and observed students taking a history that was the extent of it --so there was no actual teaching" "I need to hear ...more focused and reliable advice and this will be from the tutor" "Having the tutors to give us feedback is more reliable because they have more experience" "But we need like-improved and higher knowledge to correct us"

Forming and Norming
An insight into the importance of group formation emerging from the discussions.
Students want faculty to form their simulation teams.
"No no no. it is too much drama in doing your own group.

Forced into it is better"
"If we made our own groups we will not take it seriously" "Gonna be like-oh yeah that was great and we move on with our lives"

Co-operative learning teams
A safe learning environment and thus effective learning through peer feedback in SBE is dependent on the formation of cooperative learning teams. Students reported that this was happening gradually over repeated experiences. The key to forming these groups effectively and accelerating the breaking of barriers is pre-training on the format of tasks. The value placed on peer feedback also grows as relationships grow within the group.
"Whereas now I am with these people that I have spent so much time with, they tell me stuff that I need to improve, -I take it as a thank you, you guys are trying to help me.

Because they are my friends and you are not as nervous around them any more either"
From the video recordings, it is evident that students are more inclined to provide constructive feedback verbally after a session than complete the paper evaluation form.
The quality and quantity of interactivity and feedback vary dramatically but as there was an expectation of input from the student who had completed the task each observing student had to engage with the process and produce comments. The construction of peer feedback is a high-level activity that is cognitively demanding and engagement and collaborative learning were evident.
In written feedback 65% of students rated peer performance as excellent-"everything done correctly", which did not correlate with the tutor assessment of the same event or the summative end of term exam results for history taking. When students did mark the borderline grade, a significant difference was found for only one individual element of the feedback score. Students marked "does not show the appropriate level of confidence/fluidity" significantly more frequently than any other of the seven elements suggesting they felt more comfortable relaying this element of feedback as it does not point to understanding or knowledge (Figure 2).

Discussion
The literature on peer review has focused on students' ability to grade the work of others.
From our results this approach may be flawed as our student awarded grades did not correspond to faculty grading. However, opportunities for students to engage in interactive and collaborative activities with their peers have been shown to contribute to better learning outcomes, including the development of higher order thinking 11. SBE is ideally placed to provide these opportunities and we have previously demonstrated that students like discussing learning points with peers in SBE 12 In constructivism, faculty and peer support contribute to learning through the concepts of scaffolding, cognitive apprenticeship, tutoring, cooperative learning and learning communities 13,14 . SBE in small groups has the potential to change both the students' and faculty's role through shared inquiry thus "dissolving the Atlas Complex" where faculty is normally required to shoulder the entire responsibility for the learning process 15 . However our work and others have found that learners viewed faculty members as the ultimate knowledge validators 16. Critical thinking, reasoned evaluation and leadership skills are core competencies which may be developed through the collaborative learning experience of peer observation and feedback in the context of clinical simulation. The skills gained from the experience of collaborative learning are highly transferable to team-based work environments 17 .
An ability to construct feedback is a fundamental requirement in professional settings, and learning how to give constructive feedback should be viewed as an essential skill, and for success, students need to be taught how to give effective peer feedback. Peer observation and feedback moves away from learning as an individual activity and students need many opportunities to conduct peer assessment to become familiar and comfortable with this process.
Reflection is an essential activity in developing the cognitive schemata for problem solving and decision making in medical communication, fostering the application of different behaviors, actions and outcomes if the situation was to be encountered again 18,19 . However, critical self-evaluation is vulnerable to cognitive bias and often avoided 20,21 . For accuracy, meta-cognitive judgements of one's own performance should be accompanied with systematic and intentional elicitations of the views of others 22 .
Formative assessment is an invaluable part of the learning process if the learner is able to understand the feedback and act upon it. Constructivists regard learning as an active process where learners discover principles, concepts and facts for themselves and constructivist models stress the need for collaboration among learners 18  Students need to be taught the necessary skills for effective collaboration, particularly those skills that will help them succeed in a group environment 24 25 . The experience of observation can equip students with the meta-skills to learn by watching 21,26 . Being involved in peer feedback, can reframe student's views of feedback as a dialogic, participative process, and help them begin to recognize the importance of taking deep approaches to learning and viewing the subject matter through a different lens 27 . To aid translation of the simulation experience to the clinical environment, students need to be supported with practical strategies to direct their attention and develop a nuanced professional gaze that renders their work environments more pedagogically rich 28 . Our students report altered attitudes to working with fellow students on the wards indicating that they would be more likely to request a peer to observe them and provide feedback having completed the simulation exercises.
Tools are required to help students who are observing their peers during simulation become better at noticing what is important and more attuned to what is going on.
Tailored resources will help to make the simulation experience more purposeful and valuable for student observers, and help them to contribute more confidently and specifically to feedback discussions.

Conclusions
In conclusion, co-operative learning in SBE can be augmented by peer feedback in small groups to produce effective social constructivist approaches. Engaging students in reviewing and giving feedback to each other in a safe simulated environment may help develop social cohesion, foster learning communities and drive meta-learning. There is significant variation among students with respect to skill decay and this relates directly to   Peer assessment marks comparted to formative and sumative faculty assessment.
Peer assessment marks were compared to the marks of the faculty facilitators who observed the session and to the end of term summative OSCE scores for the history taking station. No correlation was observed between the peer assessment score and the faculty facilitator's score. No correlation was observed between the peer assessment score and the end of term OSCE *: z score = -9.7 vs summative OSCE. The faculty facilitator's mark did correlate with the OSCE score. ^: z score = 1.0 vs summative OSCE. N=254

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download. Tables.pdf