Peer Feedback and Reflective Practice in Public Service Interpreter Training

The paper discusses the importance of student-generated feedback, that is, peer feedback and self-assessment in public service interpreter training. The importance of peer feedback and self-assessment is widely recognised in teaching and learning and benefits include: pro moting analytical and critical thinking skills, students’ active participation in the learning process, promoting a collaborative model of teaching and learning, students’ responsibility and autonomy, to name but a few. However, their beneficial character can also be observed in public service interpreter training . The aim of the pilot study conducted among trainee inter preters (MA students) of public service interpreting course was to examine interpreting qual ity and compare positive (strengths) and negative aspects (weaknesses) of trainee interpreters’ performance identified by them by means of peer feedback and reflection (self-assessment). The trainees participated in simulated public service interpreting sessions and later were asked to reflect on their own as well as their peers’ performance. As seen from data analysis, there are discrepancies between peer feedback and reflection in the perception of students’ strengths and weaknesses and a negative trend can be observed in the case of reflection.


Introduction
Assessment is considered an essential aspect of the process of teaching and learning. The aim of assessment, when used effectively, is to assist students to learn, enable teachers to monitor students' progress, identify their strengths and areas for improvement as well as inform teachers whether students achieved their learning objectives, to name but a few. An aspect which is inextricably https://orcid.org/0000-0003-0874-1770 134 linked to assessment is feedback, which can be provided by a teacher, peers or students themselves. It is vital to note that "the responsibility for learning is in the hands of the teacher and the learner, and therefore they both must act to have the best outcomes for learning" (McFadzien, 2015, p. 17, emphasis added K. H.). Therefore, it is essential to allow learners to take a certain amount of control of their learning. This can be achieved by peer feedback and students' self-assessment. Learner independence, which is one of the goals of student-conducted feedback, appears to be particularly crucial at the tertiary level, where students are required to take the initiative and responsibility for their own learning.
There are mutual benefits of student-conducted assessment both for students and teachers, as rightly stated by Brew (1999): Assessment and learning must increasingly be viewed as one and the same activity; assessment must become an integral part of the learning process.
[…] When teachers share with their students the process of assessmentgiving up control, sharing power and leading students to take on the authority to assess themselves-the professional judgment of both is enhanced. Assessment becomes not something done to students. It becomes an activity done with students. (Brew, 1999, p. 169) The fact that assessment should be "an activity done with students" appears to have applicability particularly in the context of public service interpreting classes, which are practical in nature and where trainee interpreters need to obtain hands-on experience useful in their future practice. Not only do they need to possess knowledge and skills in interpreting, but also ability to reflect critically on their own performance in order to identify and evaluate the areas for improvement and devise their own action plan, as well as listen to, understand and respond to the remarks of their future clients about their performance. It is argued in the paper that those skills can be learnt and achieved by means of reflection and peer feedback, which are the scope of the paper.
Considering the importance of the topic in the interpreter training, there appear to be limited comparative studies on peer feedback and self-assessment (Bartłomiejczyk, 2009;Fowler, 2007;Hartley et al., 2003), particularly in the field of public service interpreting training, which is the scope of the paper. Previous studies, mostly related to conference interpreting, generally tend to concentrate on one method, that is, either self-evaluation (Russo, 1995;Bartłomiejczyk, 2007;Postigo Pinazo, 2008;Z. Lee, 2015;Y. Lee, 2005) or peer feedback (Wang & Han, 2013;Pallero Singleton, 2015;Su, 2019) or compare self-assessment and teacher's assessment (Y. Lee, 2016;J. Lee, 2018).
The aim of the pilot study conducted among trainee interpreters was to examine interpreting quality and compare positive (strengths) and negative (weaknesses) aspects of trainee interpreters' performance by means of reflection (self-assessment) and peer feedback. The study aimed to address the following research questions: -What aspects of their own and their peers' performance do trainee interpreters perceive as strengths and weaknesses? -Are there any discrepancies between peer feedback and reflection? Which components of trainee interpreters' performance tend to be prioritized over others? -Do trainee interpreters tend to be critical and report more negative than positive aspects of their own and their colleagues' performance?
The paper is organized in the following way. The paper begins by describing the role of assessment and feedback in the process of teaching and learning. Then, it discusses the importance of peer feedback, self-assessment and reflective practice in public service interpreting together with their limitations. Next, the paper describes the pilot study used, its analysis and results. Concluding remarks and suggestions for further research are discussed in the final part. It should be noted that it is beyond the scope of this paper to examine in detail the reflective practice applied during classes, which is a part of a larger study. Only the evaluation stage of Gibbs' (1988) reflective cycle, which focuses on identifying strengths and weaknesses, will be commented upon.

The Role of the Assessment and Feedback
Assessment is a crucial element of education and an essential part of the teaching and learning process. Summative assessment, which tends to be most frequently applied in the classroom, allows teachers to establish students' knowledge and skills (Taras, 2005), achievement and progress (Anderson, 1989(Anderson, , 1990 as well as identify their strengths and weaknesses. Formative assessment, on the other hand, given to students throughout the course, enables teachers to observe how and if they are progressing and assists in improving their performance. It also allows teachers to reflect on their teaching, adjust the methods used or modify the content of the course (Harmer, 2007). Both types of assessment play a pivotal role in students' motivation, their achievement of goals and can contribute to their level of satisfaction, or dissatisfaction, with the course. The paper focuses on formative assessment, which in the context of interpreter training, is used to provide continuous feedback on trainee interpreters' progress (Z. Lee, 2015).
However, when it comes to motivating students and contributing to their progress, it seems it is not merely assessment itself that counts, but feedback they receive (McFadzien, 2015;Shin et al., 2016). Feedback, which is considered "the most powerful single moderator that enhances achievement" (Hattie, 1999), can be teacher-as well as student-conducted. While the importance of teacher feedback is unquestionable, it is often suggested to complement it with student-conducted feedback, that is, peer feedback and self-assessment (Black & Wiliam, 1998). Not only does it actively involve students in the process of learning and fosters "a relationship of engagement between lecturer and students" (D'Hayer 2013, p. 328), but also contributes to the overall amount of feedback students are exposed to (Black & Wiliam, 1998). There is ample evidence of the advantages of student-conducted feedback such as, student autonomy and responsibility, "a reflective approach to learning" (Cao, 2017), problem detection and solving or mutual learning both for those receiving and providing feedback, to name but a few. A more detailed discussion is provided in the next section with reference to Interpreting Studies.

Importance and Benefits of Peer Feedback, Self-assessment and Reflective Practice for Public Service Interpreters
The paper argues that the ability to reflect on one's own performance and evaluate others should become a part of formative assessment and interpreter training. They are as crucial as systematic maintenance and improvement of interpreter's knowledge and skills-due to the fact that interpreting is a practice profession (Dean & Pollard, 2013), where reflective practice is considered a pivotal element of everyday practice. Highlighting the unquestionable value of training and formative assessment, Niska rightly states: "I don't think any test can be a substitute for proper training, nor is testing per se a remedy for a lack of interpreters. Tests don't produce interpreters; proper education does" (1998, p. 275, emphasis added K. H.). There are numerous benefits of studentconducted feedback which can be observed in the case of interpreter training and these are detailed below. First, attention is drawn to peer feedback and next reflective practice is commented upon.
Peer feedback creates a learning environment which is learner-centered and collaborative, and knowledge becomes constructed through social sharing and interaction (Liu et al., 2001). The need for a constructivist approach to translator training has been observed by Kiraly (2000, p. 194) who maintains that it aims at "emancipating learners and to making them able to think for themselves and to depend on each other, on their individual capabilities for independent learning." In such an environment, trainee interpreters become actively involved with their peers and learning becomes "an interactive constructive process" (Kiraly, 2000, p. 39) and "takes place as the outcome of active mental processing and when learners perceive meaningful connections between new and acquired information" (Moser-Mercer, 2008, p. 10). As a result, trainees identify problems and attempt to solve them and therefore feedback becomes beneficial both for its recipients and providers. This in turn, leads to increasing trainees' responsibility and autonomy in the learning process and also enables them to concentrate on their own learning.
In her study, Pallero Singleton (2015) aimed to obtain opinions of conference interpreting students on peer feedback and its usefulness by means of an online survey and an email interview. When it comes to interpreting skills, students in Pallero Singleton's study noticed the benefits of peer feedback in terms of language skills, analytical skills, interpersonal skills, and speaking skills (i.e., presentation and delivery). She has also found out that students hold positive opinions and recognise the value of peer feedback since it contributes to learner autonomy and self-reflection and that it is "mutually beneficial." Some of the comments given by the participants were as follows: "Giving feedback can help you to reflect on your own performance or on techniques." "Feedback is something positive-regardless its content-and it is interesting even for the person giving it." The educational value of peer feedback for both parties is also recognised by Fowler (2007, p. 261), who claims that "[a]ssessing one's peers also makes one more aware of the shortcomings in one's own work, and so this task acts as a learning instrument for the assessor. It is just as challenging a task as doing the translation […]." Due to the fact that peer feedback tends to engage a group of students, which was the case in the current study, comments which interpreting students receive are likely to be more individual (Bijami et al., 2013) and personalized. Students may express their opinions in a distinctive way and focus on different aspects of the same skill or behaviour. Not only may peer feedback be considered more individual, but also more comprehensible, since it is conducted by peers, not the teacher. Students are likely to use less metalanguage, and thus their comments may be better understood, especially by weaker students. Furthermore, as Clarke (2008, as cited in Cao, 2017 notes, "looking at the work of others can help pupils to understand the different approaches they could have taken [...] that there are different ways of achieving success." In other words, by analyzing strengths and weaknesses of their peers, students are likely to either aim at avoiding their mistakes or incorporating their peers' strengths into their own practice, which leads to active learning. Finally, receiving peer feedback (and also teacher feedback) can develop in trainee interpreters the ability to respond to constructive feedback, a skill which is highly beneficial for future interpreters. In their future career, the trainees will need to possess the ability to accept criticisms and dissatisfaction from their clients, be open-minded as well as learn how to respond to them. It is also crucial to develop their awareness of low quality feedback that may be merely negative or "in the form of complaints" (Lee, 2005), thus hardly constructive. Trainee interpreters should learn how to respond to such feedback and not get discouraged by it.
As regards self-assessment, many of the benefits are akin to peer feedback, such as ensuring student autonomy (Hartley et al., 2003), greater understanding of learning objectives, standards and goals they are aiming for (Clarke, 2008;Y. Lee, 2005), contributing to students' progress due to their active involvement (Cao, 2017) or awareness of both strong and weak points of their own performance (Russo, 1995).
Given that self-assessment applied in the study focuses on reflective practice, benefits of reflection in the context of interpreter training will be commented upon here. Reflection, as noted by Boud, Keogh, and Walker (1985, p. 19), "[…] is an important human activity in which people recapture their experience, think about it, mull it over and evaluate it. It is this working with experience that is important in learning." However, since the experience alone does not automatically need to lead to learning (Gibbs, 1988), the aim of reflective practice is to have a careful look at one's own experiences and analyse them to improve one's own performance as well as avoid the mistakes and decisions that have been made. There are numerous tools for reflection such as portfolios, logs, logbooks, peer discussion, group discussion, reflective journals, and reflective essays (Z. Lee, 2015), guided commentaries (Norberg, 2014), self-assessment reports (Y. Lee, 2005), blogs or diaries.
The reflective model suggested by Gibbs (1988), applied in the study, can be considered useful for trainee interpreters. The model (Figure 1) promotes experiential learning and by means of reflection "learners can gain new insights into the experience" (Z. Lee, 2015, p. 38). Due to its cyclical nature, students are able to focus on the experience step by step and in detail. It is argued that it is the stages of evaluation, analysis, conclusion and action plan which are particularly important for students to actively explore the experience (Gibbs, 1998). Learners consider strong and weak points of their performance (evaluation), draw conclusions from the experience (analysis), consider alternative steps which they could have taken (conclusion) and think how they can act in the future should a similar situation happen (action plan). The current paper does not aim to investigate all of the stages of the reflective cycle, but draws attention only to the evaluation stage, that is, describing positive (strengths) and negative (weaknesses) aspects of trainee interpreter's performance.  (Gibbs, 1988) In addition to autonomy and active involvement, reflective practice promotes trainee interpreters' responsibility for their own learning, since it is only for them to decide how much they will learn from the experience. Their success depends on how specific and clear they are when describing their experience to start with, how well they identify their strengths and weaknesses and alternative steps to be taken, and finally how clear and specific their action plan will be. Consequently, reflective practice focuses on problem detection, problems solving and critical thinking skills, and can also assist interpreters making informed and balanced decisions. Such skills are fundamental for interpreters, particularly in public service settings, where interpreters are often faced with dilemmas, not necessarily of a linguistic nature.
It is important to stress that not only does reflective practice contribute to trainee interpreters' learning, but also plays a crucial role in professional development since it can promote good professional behaviour and routines.
Self-evaluation enables trainees as future interpreters, to become reflective practitioners and raises their awareness of how vital it is for practice professions (such as teachers, doctors, and public service interpreters) to reflect on their action for the purposes of "professional growth and improved work outcomes" and effective decision-making in their everyday practice (Dean & Pollard, 2013, p. 140).
As described above, peer feedback and self-assessment can be considered highly beneficial for interpreters. However, what might bring even more benefits would be combining them together with teacher feedback, because then "[…] the trainee has the opportunity to gather feedback about her own performance from a range of different sources, all of whom may have a different perspective" (Fowler, 2007, p. 257, emphasis added K. H.). Incorporating all types of feedback is also likely to contribute to increasing students' confidence in their own skills and to a belief that "they have the potential to perform better tomorrow than today" (Choi, 2004, as cited in Lee, 2016, p. 93). Three-dimensional feedback, however, which is a part of a larger project of the author, is beyond the scope of the paper.
Student-conducted feedback can have numerous benefits for interpreting trainees and teachers alike, as was demonstrated in this section, provided that certain conditions are fulfilled. These include knowledge of "the theoretical aspects of interpreting" (Fowler, 2007, p. 261), students' knowledge and understanding of the assessment criteria (Fowler, 2007;Y. Lee, 2005;Pallero Singleton, 2015), clear and transparent assessment criteria (Bartłomiejczyk, 2007;Fowler, 2007;Su, 2019;Y. Lee, 2005; Z. Lee, 2015), prior instruction by the teacher and student training in how to provide peer and self-assessment (Fowler, 2007;Y. Lee, 2005, p. 4;Black & Wiliam, 1998), and "careful coaching and supervision" (Y. Lee, 2005, p. 3). These conditions are vital since learner autonomy is not an automatic skill for every student (Y. Lee, 2005). Moreover, "the fear of the unknown" and lack of experience and/or knowledge may result in students' lack of interest or discouragement in such types of feedback.
Having discussed the importance and benefits of student-conducted feedback in the context of public service interpreting training, let us now turn to the next section, which attempts to outline the design of the study.

The Study
The intent of the study was to examine interpreting quality and compare positive (strengths) and negative (weaknesses) aspects of trainee interpret-Peer Feedback and Reflective Practice… 141 ers' performance by means of reflection (self-assessment) and peer feedback. Qualitative data collection methods applied in the study were retrospection and observation. The study, as stated in the introduction, aimed to address the following research questions: -What aspects of their own and their peers' performance do trainee interpreters perceive as strengths and weaknesses? -Are there any discrepancies between peer feedback and reflection? Which components of trainee interpreter's performance tend to be prioritized over others? -Do trainee interpreters tend to be critical and report more negative than positive aspects of their own and their colleagues' performance?
Purposive sampling was employed to select the participants. Participants in the study were final year postgraduate students (n = 40) attending the Translation and Interpreting program with Chinese, German, and Arabic at the Institute of English (University of Silesia in Katowice). Trainee interpreters were attending a public service interpreting course, which runs for one semester during their final year. Due to a large group size, during public service interpreting classes trainees were divided into three groups. The trainees were familiar with the teacher, who delivered translation and interpreting classes for them in the previous years. They had prior interpreting experience and had attended simultaneous and consecutive interpreting classes for six semesters (at the undergraduate and postgraduate level). They also took translation classes (general and specialized translation) and subjects such as theory of translation and interpreting and methodology of translation and interpreting research. The students' working languages were Polish and English.
For the purpose of the study, trainee interpreters participated in simulated scripted role plays (healthcare setting) performed in a triad. The interpreters provided bi-directional translation (English-Polish, Polish-English) and the mode of interpreting was consecutive with or without notes. The average length of a simulated role play was seven minutes. For the purpose of the study, due to a large group size, trainees were divided into three groups of ten students.
The scripts for the role plays belonged to similar genre (medical interpreting) and had similar degree of difficulty and length for all students. Trainees had freedom in the selection of primary participants for their role plays. Since the aim of the study was to investigate positive (strengths) and negative (weaknesses) aspects of trainee interpreter's performance, they were asked to undertake self-assessment in the form of written reflection as well as provide written peer feedback to their colleagues.
In order to provide peer feedback, participant observation was employed. While watching the performance of their colleagues, trainees were asked to note down comments focusing on strengths and weaknesses (as many as, in their opinion, would be observable). Rules of peer feedback were made explicit to students (Cao 2017). They were asked to provide constructive, specific, and honest feedback for every trainee who acted as a public service interpreter and advised to avoid vagueness. It is worth noting, however, that trainees had prior opportunities for peer and self-assessment, observing and assessing their peers during simulated role plays practiced throughout the semester, since, as Fowler (2007, p. 257) rightly states, "in general, the more opportunities for observation of interpreted role-plays they have, the more they will learn." They also had prior experience in providing feedback during simultaneous interpreting classes in the past. What is more, trainees were familiar with the assessment criteria for the course (Fowler, 2007;Su, 2019;Y. Lee, 2005;Z. Lee, 2015) and aspects of the public service interpreter's role and behaviour, and were asked to apply them while providing peer feedback. Such knowledge is essential so that students are aware of what to comment upon (Norberg, 2014). Introducing peer feedback ahead of the introduction of reflective practice is recommended by the author of the paper on the grounds that providing peer feedback is likely to be easier for students (Cao, 2017). To ensure the spontaneity of expression, it was decided to allow students freedom with regards to the choice of language, or mix of languages to provide their peer feedback, the practice also suggested by Bartłomiejczyk (2007).
To undertake self-assessment after the completion of the interpreting task, trainee interpreters were asked to produce a piece of reflective writing. Reflective writing is relatively common to many courses, for instance at British universities, where it often becomes a core feature of the assignments, yet is not so prevalent at the universities in Poland. Therefore, it was decided to attempt to incorporate it into interpreter training practice. The primary aim of the task was to enable the trainee interpreters to think critically about the experience, that is, the interpreting task, and learn from it. Reflective writing was adopted for the purpose of the public service interpreting course to allow trainee interpreters-future interpreters, to become reflective practitioners and raise their awareness that it is vital for practice professions (such as teachers, doctors, and public service interpreters) to reflect on their action for the purposes of "professional growth and improved work outcomes" and effective decision-making (Dean & Pollard, 2013, p. 140). Nevertheless, it is argued that the ability to conduct such practices effectively in the future depends on trainee interpreters' exposure to it during training.
After completing their interpretation in the role plays, trainee interpreters were asked to write a reflection. It was decided that written format will be beneficial for the trainees since previous studies show that reflective practice in interpreter training takes the format of diaries, blogs, logbooks, reflective journals, and reflective essays (Z. Lee, 2015), self-assessment reports (Y. Lee, 2005) or guided commentaries with respect to translator training (Norberg, 2014). Due to the fact that trainees have not previously been exposed to reflective writing and due to time constraint, a more structured approach was adopted. Trainees were presented with a handout with Gibbs' reflective cycle, divided into six stages (see Figure 1), and questions corresponding to each stage, which they were asked to answer. As stated previously, the paper does not aim to focus on all of the stages of reflective writing, but only draws on the data from the evaluation stage, which aimed at stating positive (strengths) and negative aspects (weaknesses) of trainee interpreters' performance. 1 In the evaluation stage, trainees were asked to enumerate at least two or three items in each category. At the beginning of the class, they were briefed how to attempt the task and had an opportunity to familiarise themselves with the materials as well as to ask questions. While completing their reflective writing, trainees were asked to conduct self-evaluation "with the target audience in mind" (Su, 2019, p. 180). Such a practice was employed to draw trainees' attention to a degree of a public service interpreter's responsibility in the interaction and consequences of their actions. To ensure the spontaneity of trainee interpreters' input, it was decided not to impose any time limit. In other words, students were able to take as much time as needed to provide a written reflection of their performance. The majority of trainees seemed to spend an average of twenty minutes on the task. Similarly to peer feedback, they were allowed freedom with regards to the choice of language used (Bartłomiejczyk, 2007). While completing their reflective writing, trainees were able to consult the scripts of the role plays if they wished to assist their memory. To ensure appropriate conditions for reflective writing, trainees were asked to go into a classroom next door in order to be able to complete the task without being disturbed. Five of the students, however, decided to sit at the back of the classroom and complete their reflection there. Finally, trainee interpreters were given the option of reflective writing to be anonymous or could include their name if they wished.
In addition to peer feedback and self-assessment, feedback was also completed by the teacher, and trainee interpreters' performances were voice recorded for the purpose of teacher feedback. To avoid increasing the trainees' anxiety it was decided not to use video recording. While producing their reflective writings, the trainees were not presented with audio recordings, the practice which they were used to during almost every simultaneous interpreting class. Such a decision was motivated by the fact that in their future practice as public service interpreters in the institutions such as police, court or city halls, it will be rather impossible for them to record their own performance for the purposes of reflective practice. Yet, it is hoped that as professionals they will still reflect on the assignments, areas they could improve and decisions they made. This will be their reflection-on-action (Schön, 1983), a skill which is so vital for practice professions. After the sessions, trainee interpreters were able to consult peer feedback as well as received a detailed teacher feedback. Also, in an informal whole class discussion, trainees were asked about the usefulness of peer feedback and self-evaluation with regards to their learning.
Qualitative data collection methods used in the study are subject to certain limitations, as observed in existent literature. The limitations of observations and reflective practice are commented upon below in the context of the current study Observations can be considered intrusive to a certain extent (Creswell, 2014;Flick, 2015) and consequently, likely to be stressful for the subjects under observation. It is worth noting that in the observations carried out for the purpose of the study to obtain peer feedback, the observers were the colleagues with whom trainees were familiar with, and thus some rapport had already been established. It could then be argued that this allowed for the alleviation of stress levels and anxiety to a certain extent.
Another commonly mentioned limitation of observations is potential bias and subjectivity (Creswell, 2014). In other words, in the context of the pilot study, the fact that observers were acquainted with their colleagues could possibly affect the quality of their peer feedback. As far as reflective writing is concerned, trainees could avoid disclosing certain details about their own performance, particularly the negative ones, or quite the contrary, be too selfcritical (Black & Wiliam, 1998). Yet, it is important to stress that subjectivity seems an inherent element of qualitative research and certain steps can, and should, be taken to minimize it. In the current pilot study, the observers providing peer feedback were asked to apply and focus on the assessment criteria for the subject, which they were familiar with and which were used during prior peer assessments. What is more, they were instructed to provide such a constructive feedback that they themselves would wish to receive from their peers. Trainees engaged in reflective writing were asked to provide honest comments about their own performance, and were ensured that their reflections would only be shared with the teacher for the purpose of the study. In addition, "to balance out the subjective influences of individuals," investigator triangulation was applied (Flick, 2015, p. 218;Flick, 2018). This means that trainees obtained feedback from their peers and the teacher 2 as well as reflected on their own performance. This way, they were able to compare three types of feedback and also obtain a comprehensive perspective.
Finally, the quality of observation is considered to depend on the skill of the observer "to observe, document, and interpret what has been observed" (Kawulich, 2005, p. 6;Creswell, 2014), while the quality of self-evaluation-on the skill to reflect in detail. These aspects seem to be particularly true when they are not conducted by professional researchers, that is, in the context of classroom peer observation and self-evaluation. A viable solution to minimise this problem is to train students in providing constructive peer feedback and self-evaluation and ensure they become a constant element of the classes. Such practices were encouraged by the author of the study during simultaneous interpreting classes. However, as stated previously, it is vital to bear in mind that, students' knowledge and understanding of the assessment criteria for the course are inherent elements of successful training (Fowler, 2007;Su, 2019;Y. Lee, 2005; Z. Lee, 2015).

Analysis
The study places itself within the qualitative research paradigm and thus qualitative approaches were applied in the data analysis. All of the trainees' comments, from peer feedback as well as reflection, were coded (initial coding) according to a skill or behaviour they described. Next, a thematic analysis was conducted and codes were aggregated into five themes (Creswell, 2014). Lastly, there followed counting the number of occurrences of the codes and themes, that is, qualitative data transformation (Creswell & Plano Clark, 2007) or "quantitizing" (Sandelowski, 2003;Sandelowski et al., 2009).
The themes, which corresponded to the assessment criteria categories, were: presentation and delivery, accuracy and fidelity, interpreting skills and strategies, interpersonal competence and non-verbal communication. One of the categories not mentioned by the trainees, yet listed in the assessment criteria for the subject, was cultural awareness. The first three categories are in line with Wu's (2010) interpreting categories for simultaneous interpreting. The final two categories included skills particularly essential for public service interpreting, that is, interpersonal competence and non-verbal communication (Toledano Buendia & Aguilera Avila, 2017; Van den Bogaerde et al., 2016). Each of the five categories comprised of the following subcategories: -presentation and delivery: pronunciation, intonation, clear voice, pace/tempo, fluency, seriousness/composure, professional behaviour; -accuracy and fidelity: content accuracy, omissions, additions, grammar (correctness), minor errors/ slips, terminology; -interpreting skills and strategies: asking for clarification, asking for repetition, asking for explanation, comprehension, code switching, memory, focus and concentration, note taking, reaction/responsiveness, stress management; -interpersonal competence: rapport/attitude towards participants, communication skills, decision making, trust, empathy/concern; -non-verbal communication: eye contact, gestures, body language, facial expressions. All trainees completed and submitted peer feedback and reflection. The great majority of trainees did not seem to experience difficulties with providing constructive peer feedback and overall a good quality of feedback can be reported. Peer feedback did not contain any "personal" or irrelevant comments and all trainees' comments were related to the actual performance of their peers. The majority of comments were "moderately elaborate" (see Table 1), that is, "specifying the problems and their locations, or illustrating the problems with examples" (Su, 2019, p. 181). There were only three instances when peer feedback seemed rather vague and "the least elaborate" (Su, 2019, p. 181) or similar comments were repeated for several students. An overall good quality of peer feedback could indicate that trainees are well aware of assessment criteria applied in public service interpreting classes and key concepts in public service interpreting (Fowler, 2007). The comments in reflective writing appeared natural, rather detailed and reflecting trainees' feelings immediately after their performance. The comments were also longer than those for peer feedback. This can be attributed to a greater amount of time devoted to writing them, which on average was 20 minutes, while in terms of peer feedback it was 6-7 minutes. Table 1 contains examples of trainee interpreters' peer feedback and reflection. The symbol '-----' indicates that no comments were given. With regards to the language chosen, 30 out of 40 trainees (75%) chose English to provide their peer feedback and 38 out of 40 (95%) used English in their reflection. As stated previously, in both cases they were informed that either Polish or English was acceptable.
Given that the aim of the study was to observe whether trainee interpreters display tendency to be critical and report more negative than positive aspects of their own and their colleagues' performance, Figure 2 illustrates the distribution of positive and negative comments with respect to peer feedback and reflection. To understand what aspects of their own and their peers' performance trainee interpreters perceive as strengths and weaknesses, as well as explore possible discrepancies between peer feedback and reflection, the distribution of the five categories obtained in the qualitative analysis in the context of peer feedback and reflection is displayed in Figures 3 and 4.

Examples of strengths and weaknesses from peer feedback (PF) and reflection (R)
Category

Positive (strengths) Negative (weaknesses)
Presentation and delivery (PF) She is calm and speaks clearly. Good pace and fluent.
(PF) She can express emotions well. An expressive interpretation.
(R) I believe it was a good interpretation, almost no repetitions, a smooth flow of conversation was maintained.
(R) Quick reactions, there were no long pauses in my translation.
(PF) His voice was really monotonous.
(R) I had some problems with being serious.
(R) Flow of speech. I got confused and there was an unnecessary pause in the conversation.

Accuracy and fidelity
(PF) She does not omit anything. Really close to the original.
(PF) A really good knowledge of terminology (R) I didn't add anything, I was not trying to alter the message (R) I didn't have any problems with terminology. I was prepared.
(PF) Sometimes she does not translate short answers. (*trainee's underlining) (PF) His choice of vocabulary in Polish, e.g. when speaking to the patient he used "dehydracja" which would be too difficult to understand.
(R) I lacked specialised vocabulary, I had to ask the doctor to explain and repeat many times.
(R) The fact that he did not know one term caused serious communication problems.

Interpreting skills and strategies
(PF) He wasn't afraid to ask for repetition once -when he didn't hear something.
(PF) Good that she asked for clarification and didn't try to make things up.
(R) I was concentrated on my task, not on the fact that I feel stressed (R) I translated fluently, I had no problems with memory (PF) She used omissions if she did not know the the terminology.
(PF) She could take notes but she didn't. She missed some information because of that (R) I was stressed and I know that I had to start sentences a few times to make them sound natural in Polish and to make the conversation understandable for both sides. (PF) A really poor eye contact which makes the whole situation overly formal.
(PF) He was using a lot of hand gestures.
(R) Too many facial expressions.
(R) I did not look at the other participants often.

Results and Discussion
Data comparing the distribution of positive (strengths) and negative (weaknesses) aspects of trainee interpreters' performance by means of reflection and peer feedback is displayed in Figure 2. As can be seen from the figure, it is apparent that these two methods of providing feedback yielded significantly different results As far as reflection is concerned, there was a tendency among trainee interpreters for reporting more negative (58%) rather than positive aspects of their performance, which is in line with the findings of Bartłomiejczyk (2007Bartłomiejczyk ( , 2009 who, in her study on simultaneous interpreting, also noted a negative trend in students' self-assessment (56.2%). However, Z. Lee (2015) in her study of student logbooks as a form of reflective practice, reported approximately equal numbers of segments coded for positive performance and negative performance. The positive ones accounted for 757 of all 1655 segments coded (about 45%), while the negative ones for 792 (about 47%). This differs to some extent from the findings presented here.
On the contrary, a positive trend can be observed within peer feedback. The majority of trainee interpreters identified more strengths (62%) than weaknesses (38%) in their colleagues' performance. These results are in contradiction to Bartłomiejczyk (2009) who reported a negative trend in the case of peer feedback (57.7%). The tendency for being critical when providing peer feedback has also been mentioned by Fowler (2007, p. 256), who claims that "there is always the temptation for students to focus on the negative, rather than the positive, aspects of an interpreter's production." Such results would suggest that public service interpreting trainees display a tendency to be less critical of their colleagues than conference interpreting trainees (Bartłomiejczyk, 2009), while appear to be more self-critical The quantitative data illustrating the distribution of the five categories obtained in the qualitative thematic analysis in the context of peer feedback and reflection is presented in Figures 3 and 4. The former figure presents positive aspects (strengths) of trainee interpreters' performance, while the latter negative ones (weaknesses).
The analysis of the responses revealed that trainee interpreters tend to pay attention to multiple aspects of the interpreter's behaviour and appear to be aware of the complexity of the public interpreter's role in the interaction. This was reflected in their numerous and diverse comments, which are distributed across almost all of the categories, as is demonstrated in Figures 3 and 4 (apart from one noticeable exception in the case of reflection, where none of the trainees focused on interpersonal competence while reflecting on their own weaknesses). Such findings indicate that trainees consider paralinguistic aspects an integral component of public service interpreting apart from the linguistic ones -the most significant ones for interpreters. This in turn, would imply that some of them recognise a public service interpreter as a visible agent (Angelelli, 2004) and co-participant in the interaction (Roy, 2000;Wadensjö, 1998) with a dynamic "role-space" (Llewelyn-Jones, & Lee, 2014), and not merely a conduit or a channel. The multifaceted character of the comments also implies a diversified type of feedback received by the trainees from the teacher, an aspect of teacher feedback suggested by Bartłomiejczyk (2007).  The results of the study show that there are discrepancies between peer feedback and reflection in the distribution of the categories in terms of positive aspects (strengths) and negative aspects (weakness) of the trainee interpreters' performance. First, positive aspects of both peer feedback and reflection will be commented upon, followed by the negative ones.
When it comes to strengths (Figure 3), in the case of peer feedback the comments are distributed evenly among the first three categories, that is, presentation and delivery (25%), interpreting skills and strategies (24%) and accuracy and fidelity (23%). However, there is a noticeable difference with regards to the first two categories within reflection. There are almost half as many comments related to presentation and delivery (10%), yet it yielded a higher score than in the study conducted by Bartłomiejczyk (2009) where it scored merely 4.5%. This indicates that public service interpreting trainees tend to pay slightly more attention to their own presentation and delivery than conference interpreting trainees (simultaneous interpreting). This could be attributed to the fact that they are not seated in the booth, but are visible to the primary participants, thus are aware of the fact that anything they do can be immediately noticed by them.
Nearly half of the total comments from reflection focused on interpreting skills and strategies (46%), which is almost double in comparison with peer feedback. This was in some part due to the fact that stress management (15%) was one of the subcategories, and it is widely known that interpreting is considered a highly stressful activity (Adams, 2017;Toledano Buendia, & Aguilera Avila, 2017). It is also worth noting that memory, focus, and concentration con- stituted 11% of the comments on interpreting skills and strategies. Appropriate stress management and good memory are vital skills for interpreters, and without the second one in particular, interpretation would hardly be possible. This implies that trainee interpreters recognise the importance of these skills in their work and almost a quarter feel confident about them. The remaining two categories, that is, interpersonal competence and nonverbal communication produced quite similar results, with the overall number of responses being slightly higher for peer feedback than reflection in both categories. It is especially worth noting that interpersonal competence yielded a relatively high score, especially within peer feedback (19%). Trainee interpreters considered it almost as fundamental as presentation and delivery, accuracy and fidelity and interpreting skills and strategies, that is, the categories with the highest scores. Such results point to the trainees' greater awareness of the importance of interpersonal skills in public service interpreting (Toledano Buendia & Aguilera Avila, 2017;Van den Bogaerde et al., 2016). Finally, least attention was devoted to non-verbal communication in both peer feedback (9%) and reflection (7%). A similar trend has also been noted in terms of weaknesses. It is rather surprising particularly when it comes to peer feedback, where trainees who provide it become the observers and thus should be able to recognise whether non-verbal behaviour was displayed appropriately or not, and comment on it in a greater detail.
As regards weaknesses (Figure 4), the dominant category within peer feedback was accuracy and fidelity (42%), which yielded the second highest score (38%) in the case of reflection. The tendency for the trainee interpreters to be rather critical and negative especially when reflecting on accuracy and fidelity corroborates the findings of Bartłomiejczyk (2009), who also noted a negative trend within this category in students' self-assessment. In the study by Hartley et al. (2003, p. 10) accuracy was the second most reported aspect by professional interpreters who provided feedback for trainees (the first category was coherence, that is, "making sense, no contradictions"). Such a strong focus on accuracy and fidelity in the current pilot study could be attributed to the fact that these two are considered most important aspects for interpreters whose task, first and foremost, is to provide faithful and accurate translation. What is more, the need for accuracy and fidelity seems particularly crucial in public service interpreting, where the interpreter's translation choices may have severe legal consequences, result in wrongful convictions, withstand scrutiny in court or impact on patient's health or even life, which trainee interpreters seem to be quite aware of The dominant category within reflection was interpreting skills and strategies (39%). As pointed out previously in the case of strengths, this is partly due to the fact that stress management (24%) was one of the subcategories. The results indicate that trainee interpreters consider themselves more stressed or nervous than they appear in the eyes of their peers (only 10% in the case of peer feedback). The difference between peer feedback and reflection implies that trainees are either able to manage stress quite well without fully realizing it, or that it takes the form of facilitative rather than debilitative anxiety (Alpert & Haber, 1960) and thus becomes less noticeable for their peers.
Presentation and delivery received slightly more attention within peer feedback (25%) than reflection (17%). A similar trend was also reported for strengths. Such results are partly consistent with Bartłomiejczyk (2009), who noted significantly more negative comments with regards to peer feedback (26.6%) as opposed to self-assessment (2%). The results of the study concerning the category of presentation and delivery are, however, contrary to the studies by Hartley et al. (2003) and Z. Lee (2015), in which it was rated as the most dominant category in students' self-assessment. Yet, relatively small differences between peer feedback (25%) and reflection (17%) indicate that, quite surprisingly, trainee interpreters themselves were to a certain extent able to recall and reflect on negative aspects of their own performance, not only their peers -the observers.
It is worth noting that significantly less attention has been devoted to nonverbal communication in both peer feedback (10%) and reflection (6%). Low scores for non-verbal communication are similar both for strengths and weaknesses. This finding partly accords with that of Z. Lee (2015), who reported that non-verbal behaviour was mentioned least in her students' consecutive interpreting self-assessment and accounted merely for 9 of all 1655 segments coded. The results from the study indicate that greater awareness needs to be fostered with regards to non-verbal communication and its importance in public service interpreting (Krystallidou, 2017).
Conversely to the data reporting on strengths (Figure 3), interpersonal competence received least attention among trainee interpreters. This, however, could be attributed to the fact that trainee interpreters believed interpersonal competence was displayed appropriately due to the fact that they were familiar with the trainees acting as primary participants in the role plays, and consequently, for example, rapport and trust had already been established. Thus, there was perhaps no need for negative peer feedback or reflection.
Lastly, it is important to note that none of the trainees mentioned cultural awareness skills both in peer feedback and reflection, which were discussed during the course as one of the key competencies for public service interpreters. A highly plausible explanation for this might be that all participants in the role plays were from the same culture, and therefore there were no cultural differences or misunderstandings to be commented upon.

Concluding Remarks
The intent of the pilot study was to examine interpreting quality and investigate differences between peer feedback and self-assessment (reflection) in terms of positive aspects (strengths) and negative aspects (weaknesses) of trainee interpreters' performance in public service interpreter training. The study has shown that there are significant discrepancies between peer-feedback and self-assessment (reflection) and that overall more negative aspects are reported in reflection than in peer feedback. This reflects students' tendency for self-criticism. Thus, it seems reasonable to complement reflective practice with peer feedback, which as the results of the study suggest, tends to be positive overall. In this way, it is likely that a balanced student-conducted feedback could be achieved, thanks to which students might be able to identify some positive aspects of their performance (however small they may be), and not merely weaknesses. Consequently, a more positive attitude and motivation for learning can be fostered.
The findings suggest that both peer feedback and reflection can be considered valuable tools for public service interpreting training. Firstly, it is demonstrated by numerous and diverse comments provided by trainee interpreters focusing on different aspects of the public service interpreter's skills and behaviour, not merely those of a linguistic nature. This would suggest that trainees understand the assessment criteria (Fowler, 2007;Su, 2019;Y. Lee, 2005; Z. Lee, 2015) and are aware of crucial aspects of the public service interpreter's behaviour, and are able to apply them to their own learning. However, as stated before, comments from peer feedback tended to be less elaborate than those in reflective writings. To address this issue, extra training sessions on how to provide effective and more elaborate feedback should be considered in a future study. What is more, the study has shown that peer feedback seems to be particularly beneficial when evaluating interpersonal and non-verbal communication since by acting as observers of the situation, peers are much more likely to notice those aspects. As far as reflective practice is concerned, it would also be possible for trainees to reflect on their own interpersonal and non-verbal communication skills in greater detail provided their performances were video recorded. However, this was not the case in this study and should be considered in further research. When it comes to drawing trainees' attention to the importance of cultural awareness, to which none was drawn both in peer feedback and reflection, a rather effective solution might be to engage some minority languages speakers, such as Erasmus exchange students, as one of the primary participants in the role plays. This may be an enriching experience especially as regards trainees' cultural awareness skills, and also non-verbal behaviour which tends to differ from one culture to another.
Applying both types of student-conducted feedback in the public service interpreting course was a positive experience both for trainees and the teacher, and is very likely to become a permanent feature of the course. There was no formal measurement of trainees' satisfaction, which was performed in Y. Lee's (2005) and Wang and Han's (2013) study. However, in an informal whole-class discussion during a feedback session, trainees considered the experience positive and useful, despite it being time consuming. The measurement of trainees' satisfaction will be taken into account in a further study. Some trainees' comments can be found below: TI1: It was good to write it down because I did not realise all the things.
(about reflection) TI2: I had a general idea that I had problems with this and this but when I started writing it and analysing it question by question, I found out that I had overall more pros and cons than I realised at the beginning. (about reflection) TI2: If we want to improve we should do such things, to ask some questions to ourselves. This is quite helpful, but takes time. (about reflection) TI3: When peers evaluate us, it helps to realise some mistakes and some things that they as a person, who is not a part of the situation, see, how they can actually see things. (about peer feedback) TI4: It was good to get feedback from each possible side. However, it took some time.
From the perspective of the teacher, the application of both types of studentconducted feedback was also a positive experience. It allowed trainees to be actively involved in the learning process, recognise the value of their own work, and most importantly obtain feedback from different perspectives, which was also noted by Fowler (2007). Reflective practice seemed to contribute to their analytical, critical thinking and problem-solving skills and made trainees aware why it is vital to reflect on their own performance. It was visible that the majority of trainees were indeed engaging with it and did not treat it as a "tick off the box" activity. Lastly, it was also hoped that sharing similar difficulties and weaknesses may motivate trainees and build their self-confidence since they were able to realise that their peers also experienced similar problems during their interpretation. However, it is argued that peer feedback and reflective practice should be complemented with teacher feedback to allow students to obtain multidimensional and even more comprehensive perspective. Previous research shows that "students value various types of feedback and appreciate these feedback experiences" (Y. Lee, 2016, p. 166). The combination of student-conducted and teacher feedback is also essential to minimize subjectivity of student-conducted feedback.
The findings in this pilot study are subject to certain limitations. Video recording was not used during simulated role plays. As stated previously, it was decided not to include it in order not to increase students' stress, the level of which, as reported in some reflections, seemed already quite high. In a further study, however, video recording shall be taken into account due to its pedagogical value. As suggested by Coffey (2014, p. 86), video recording can be considered "a means by which a teaching episode can be captured more permanently to be used as a point of reference for reflection" and also "the best way of encouraging self-awareness and critical thinking in the student" (Fowler, 2007, p. 258). However, a voluntary participation in the study will be suggested, since it is likely that some students may not wish to be video recorded.
Next, in order to allow for a detailed reflection and to allow more time for completing it, trainees may be asked to complete their reflective writing the same day at home while watching the recordings. To obtain a more complete account from peer feedback, apart from describing strengths and weaknesses of their colleagues, trainees could also be asked to evaluate them, that is, provide possible reasons of their peers' behaviour and offer solutions. Such practice was advocated in Su's study (2019) on peer review in simultaneous interpreting training. This would correspond to the evaluation stage in Gibbs' reflective cycle (Gibbs, 1988) applied in the reflection.
Finally, to learn about trainees' experiences of student-conducted feedback and enable them to voice their opinion of their role as "a feedback giver and receiver" (Wang & Han, 2013), a questionnaire should be administered and/or qualitative interviews conducted, as suggested by Y. Lee (2005). Such feedback on feedback will allow the teacher to obtain a more comprehensive perspective on the usefulness of student-generated feedback as "it is necessary for teachers and students to engage in dialogue on feedback and make optimal use of it" (Lee, 2018, p. 167, emphasis added K. H.).
The need for a multidimensional view of feedback seems unquestionable due to its benefits, as illustrated by previous research and the current pilot study. Incorporating multidimensional feedback into public service interpreter training can promote "a classroom culture of questioning and deep thinking, in which pupils learn from shared discussions with teachers and peers" (Black & Wiliam, 1998, p. 9). Trainee interpreters need to be able to critically reflect on their own performance, respond and act on feedback received from their clients when they become professional interpreters. This, however, can only be achieved through their active participation in the learning process and experience of student-conducted feedback during their training.