Exploring student and demonstrator perspectives of the teaching and learning provided by demonstrators during undergraduate laboratory practical sessions

demonstrators who are teaching on these programmes. His current research endeavours to determine the role of nematode assemblages in novel and sustainable agricultural systems. In addition, as a qualified science teacher and Fellow of the Higher Education Academy his education research and interventions aim to improve teaching by improving assessment, critical thinking and by enhancing the quality of teaching delivered by GTAs. Abstract Graduate teaching assistants (GTAs) such as demonstrators are integral to higher education teaching. However, undergraduate student perceptions on the impact of demonstrators to their teaching and learning is understudied. Similarly, demonstrator perceptions on the quality and value of their teaching also remains largely unexplored. We addressed both students’ and demonstrators’ perceptions of demonstrator teaching and learning within undergraduate Biology practical sessions at a UK research intensive university. We combine quantitative and qualitative data to explore students’ ideas around where their learning and feedback comes from, and the effectiveness of demonstrators during their practical sessions. This was paired with analogous data from demonstrators derived from the same concepts, thereby helping to assess their dynamic. Most students considered demonstrators to be important for teaching and learning in the laboratory setting by delivering high quality pedagogy, creating a positive learning environment and by being their primary source of feedback. Conversely, a small number of students raised issues regarding demonstrator consistency and lack of knowledge when compared with a lecturer. Students largely considered demonstrators to have sufficient knowledge and to be more approachable than lecturers. Demonstrators also recognised their valuable contribution to teaching and learning and largely mirrored the ideas undergraduate students had about effective demonstrators. Many demonstrators believe they need to be allocated more paid time to fully prepare for teaching and maximise their potential. Finally, we reflect on lessons learnt from both students and demonstrators regarding how demonstrators can improve their teaching and how universities can support this.

laboratory and practical teaching on the Biology and Zoology BSc programmes and the management and training of GTA demonstrators who are teaching on these programmes. His current research endeavours to determine the role of nematode assemblages in novel and sustainable agricultural systems. In addition, as a qualified science teacher and Fellow of the Higher Education Academy his education research and interventions aim to improve teaching by improving assessment, critical thinking and by enhancing the quality of teaching delivered by GTAs.

Abstract
Graduate teaching assistants (GTAs) such as demonstrators are integral to higher education teaching. However, undergraduate student perceptions on the impact of demonstrators to their teaching and learning is understudied. Similarly, demonstrator perceptions on the quality and value of their teaching also remains largely unexplored. We addressed both students' and demonstrators' perceptions of demonstrator teaching and learning within undergraduate Biology practical sessions at a UK research intensive university. We combine quantitative and qualitative data to explore students' ideas around where their learning and feedback comes from, and the effectiveness of demonstrators during their practical sessions. This was paired with analogous data from demonstrators derived from the same concepts, thereby helping to assess their dynamic. Most students considered demonstrators to be important for teaching and learning in the laboratory setting by delivering high quality pedagogy, creating a positive learning environment and by being their primary source of feedback. Conversely, a small number of students raised issues regarding demonstrator consistency and lack of knowledge when compared with a lecturer. Students largely considered demonstrators to have sufficient knowledge and to be more approachable than lecturers. Demonstrators also recognised their valuable contribution to teaching and learning and largely mirrored the ideas undergraduate students had about effective demonstrators. Many demonstrators believe they need to be allocated more paid time to fully prepare for teaching and maximise their potential. Finally, we reflect on lessons learnt from both students and demonstrators regarding how demonstrators can improve their teaching and how universities can support this.

Introduction
Graduate Teaching Assistants (GTAs) are postgraduate students who are enrolled in either Master's or PhD programmes and teach students on a part-time basis. In recent years, universities across the UK have relied on GTAs to help educate the increasing number of students attending university (Beaton et al., 2013). GTAs are important in relieving teaching pressure from staff, as well as supporting students with their learning (Ramos, 2001). Furthermore, when compared with other staff members, GTAs are usually in more direct contact with undergraduate students and can be influential in enhancing a student's learning experience (Huffmyer & Lemus, 2019).
The role of a GTA can be varied; for example, GTAs are asked to lead seminars, mentor students, demonstrate in laboratories and mark assessments.
In STEM (science, technology, engineering, and mathematics) subjects a student's experience in the laboratory positively impacts their interest in the degree they are studying (Benjamin, 2002;Jaeger, 2008). Thus, as GTAs are a vital source of teaching in practical sessions, it is likely that their relationships with undergraduates have an important and sometimes unmentioned part to play in their academic success. However, due to their own research commitments, teaching part-time can feel stressful and unmanageable for GTAs (Park, 2002;Rao et al., 2021). This feeling of stress is not aided by their complicated status within universities, as they are neither full-time students nor permanent staff (Compton & Tran, 2017). GTAs have less time for their own studies and are not valued as legitimate teaching staff, thereby occupying a highly demanding niche in the university (Muzaka, 2009).
Previous research has primarily focused on the role of GTAs and their experiences of teaching (Elliott & Marie, 2021;Muzaka, 2009;Park 2002;Park & Ramos, 2002;Ryan, 2014) and have highlighted the need for continual training and better pay to improve GTA confidence and morale (Chiu & Corrigan, 2019;Prieto & Meyers, 1999;Young & Bippus, 2008). Training is crucial to improving GTA teaching ability, which could potentially enhance student learning. A study by Park (2002) lists the positives and negatives of the GTA experience for both the GTAs themselves and for students. Encouragingly, it states that GTAs offer a diverse way of teaching and approachability which in turn benefits the student (Park, 2002). It also states that GTAs gain invaluable teaching experience and the opportunity to develop key skills such as time management and communicationessentially kick-starting their career in academia (McCready & Vecsey, 2013;Park, 2002). On the other hand, a GTA's lack of teaching experience and subject knowledge may hinder student learning and consequently complicate matters (Park, 2002). Muzaka (2009) highlights that most students recognise a GTA's knowledge as being very specific relative to academic staff, which is unhelpful for their learning. Perhaps this is exacerbated by a situation where GTAs often feel overworked and stressed due to insufficient time to simultaneously carry out their own research and prepare for teaching (Park, 2002). GTA confidence and authority are further areas that have been recognised as lacking by students and GTAs themselves, however GTA confidence has been shown to improve significantly with adequate training (Chiu & Corrigan, 2019;Kendall & Schussler, 2012;Muzaka, 2009;Young & Bippus, 2008). Nonetheless, some universities provide little to no training for GTAs. A study by Prieto & Meyers (1999) has reported that 30% of GTAs in their sample received no supervision or training. If this statistic reflects the picture today, improper GTA training could be negatively affecting both the student's learning and the GTA's teaching experience.
Student expectations of GTAs and their role within teaching may also influence the learning and teaching experience of students and GTAs. A study by Ryan (2014) exploring the role of GTAs in sciences surveyed students on what they thought made an effective GTA. The results of this study concluded that three themes -'knowledge, communication and affective' -were key to being a successful GTA. In essence, students believed that to be an effective GTA a good understanding of the topic, the ability to explain ideas well and engagement were fundamental. These attributes develop with training, practise and of course time, something that GTAs already struggle to find. Thus, living up to student expectations whilst being new to teaching and in some cases unprepared can make GTAs feel anxious about their teaching role and unappreciated by students (Elliott & Marie, 2021). Anxiety surrounding their role can once again damage GTA confidence and subsequently enthusiasm whilst teaching. Ultimately, this impacts a student's learning experience and their perceptions of GTAs.
The relationship between students and demonstrators inevitably works both ways (i.e., the opinions and actions of students towards GTAs affects GTA performance and vice versa). Studies that provide an insight into the student-GTA dynamic are limited (Golish, 1999;Kendall & Schussler, 2012;Nasser-Abu Alhija & Fresko, 2018;Park, 2002) and instead most studies have thus far focused on the pros and cons of the GTA experience (for example: Elliott & Marie, 2021, Muzaka, 2009Park & Ramos, 2002;Pezzella, 2014;Ryan, 2014). Thus, the purpose of this research paper is to evaluate both the students' and GTAs' perceptions of the teaching delivered by demonstrators (GTAs) in laboratory practical sessions, using surveys tailored to each party. We aim to address the following questions: The research takes place at a researchintensive Russel group university, where demonstrators are paid to support laboratory teaching, run tutorials, and mark student work. Newly appointed demonstrators must complete four hours of paid mandatory general training (which includes some pedagogic theory such as Bloom's taxonomy). Demonstrators are paid to attend preparatory briefings for all practical sessions and marking they are involved in but are not typically paid for any other preparation they do.

Ethical Approval
Ethical approval for the student and demonstrator specific questionnaires were given on the 26/04/2022 and 06/06/2022 respectively by the Faculty of Life Sciences and Faculty of Science Research Ethics Committee at the University of Bristol.

Survey data collection
To evaluate the student-demonstrator dynamic, we developed two online survey questionnaires. The first was aimed at capturing undergraduate students' perceptions of demonstrators and their performance during laboratory practical sessions. The other was curated for demonstrators and addressed their perceptions of their own teaching and learning experiences. Thus, congruent questions relating to teaching and learning and the studentdemonstrator dynamic were asked in both questionnaires in order to compare student and demonstrator perspectives. Data collection took place in the 2021/2022 academic year.
The completion of both the student and demonstrator questionnaires was voluntary and answered by undergraduate students in years ranging from 1-4 and demonstrators from the School of Biological Sciences. The student-specific survey was completed by 117 individuals (~22 % response rate), whilst the demonstratorspecific survey was completed by 26 individuals (a 27% response rate). Survey questions are listed below in Table 1.

Qualitative analysis
Thematic analyses were used to explore the perceptions, values and ideas held by students and demonstrators regarding the role, ability and importance of GTA demonstrators in teaching and learning during practical sessions.
Thematic analyses of the survey questions were achieved by assigning codes which captured the important ideas in the data, which would then be used to derive the themes. The data was independently coded by the co-authors using an inductive and latent approach focussing on the implicit meaning behind student responses according to the methods used by Braun & Clark (2006, 2012. These codes were then compared in order to validate the key themes. Codes which were uncommon or irrelevant were discounted for the final representations of the data.

Quantitative analysis
All figures were created in RStudio version 3.6.1 (R Core Team, 2019) using the R package, ggplot2 (Wickham, 2016).

Perceptions on the teaching and learning provided by demonstrators in a laboratory setting
There were 117 coded student survey responses and 26 coded demonstrator survey responses used for thematic analysis. The undergraduate and demonstrator surveys were completed by 79 females, 37 males and one nonbinary person and 16 females, 9 males and one non-binary person respectively. A visual representation of the themes derived from the combined survey data regarding how both students and demonstrator perceive the teaching and learning provided by demonstrators during laboratory practical sessions can be seen in Fig. 1.
Eight key themes were derived from the thematic analysis for both students and demonstrators regarding the teaching and learning provided by demonstrators. Both datasets derived analogous themes though framed from slightly different perspectives. As the eight themes were commensurate the descriptions of the themes will be given in congruous pairs below: Theme pair 1: Demonstrators are the interactive and engaging teachers that encourage learning/ We provide inspiration, motivation and discussion that help keep students engaged in their learning Undergraduate responses within this theme described how demonstrators engaged and interacted with them in a positive light. This included discussions of topics covered and new interesting facts taught, and often described how the interactions kept them engaged and focussed. A representative comment was: "They help [you] stay present during the practical, being asked a question can sometimes catch you off guard but it's useful to make sure you're staying on task. Being monitored aids any selfdirected learning, its less likely I am going to get distracted." The demonstrator responses reflected student responses, as many saw part of their role as "keeping students engaged", but also to inspire and motivate them. Over 25% of student respondents and 70% of demonstrators made comment relating to this theme. Table 1: Questions asked in both the student and demonstrator online survey questionnaires. Options to questions are indicated in brackets and questions that required a free text response are marked with an asterisk (*).
Figure 1: Thematic Analysis of combined survey data to ascertain how students (left) and demonstrators (right) perceive the teaching and learning provided by demonstrators in the laboratory setting.
Theme pair 2: Demonstrators are inconsistent, sometimes unprepared and/or less knowledgeable than lecturers/ We could be more consistent, prepared and better for teaching and learning if we had more paid preparation time Of the 117 student respondents, 15 undergraduates felt that demonstrators sometimes lacked consistency or preparedness. Some respondents were completely unsatisfied with all demonstrators, whereas most were complementary but cited issues with consistency between the quality of individual demonstrators and their preparedness. Other student responses indicated a preference to have the lecturer deal with their enquiries due to their perceived superior knowledge and teaching quality.
Demonstrator responses in this theme also acknowledged that they were sometimes unprepared, with some comments attributing this to a lack of sufficient paid preparation time for their teaching.
Theme pair 3: Demonstrators provide approachable, accessible, support for learning/ We provide approachable, accessible support for learning Responses in this theme tended to highlight demonstrator accessibility and how much more approachable they were relative to lecturers. Often responses explicitly emphasised how unapproachable the lecturer was or how they perceived the lecturer to be too busy for their questions, for example: "Demonstrators are the least intimidating [to ask for help] and I'm not worried about asking them a stupid question" "They [demonstrators] are more available and approachable than the lecturers, [and] often better at explaining things. They are needed because there is no way the lecturer could answer all the questions people have in practicals." Students also felt that they could confidently double-check their work with demonstrators due to their accessibility and approachability. Demonstrator responses mirrored these ideas, also recognising themselves as more approachable, relatable, and less intimidating. This was a frequently mentioned theme, as 61% of student respondents and 88% of demonstrators attributed comments to it.

Theme pair 4:
Demonstrators deliver high quality teaching that encourages critical thinking and overall learning/ We deliver high quality teaching that encourages critical thing and overall understanding Student responses attributed to this theme talked explicitly about the quality of teaching or the impact demonstrator teaching had on their learning. This included responses praising the quality of explanations from demonstrators and how that enhanced their knowledge and understanding or describing how demonstrators had encouraged them to think critically. Many responses also described how demonstrators guided students towards answers and how this had been beneficial to their learning. Some examples include: "If you were stuck on something, they [the demonstrators] wouldn't just give you the answer, they would unravel it slowly letting you actively search for the solution yourself. I felt this was a very effective way of handling it and is of great benefit to the Lab learning experience." "[Demonstrators] circulate and often ask additional questions promoting critical thinking" Demonstrator responses in this theme mirrored those made by students and often mentioned the importance of guiding and not telling, giving good explanations, developing critical thinking and challenging students. A higher proportion of student respondents (28%) than demonstrators (23%) outlined the delivery of highquality teaching and consequent learning.
Theme Pair 5: Demonstrators are one of the few key sources of feedback/ We provide feedback to students help their progress Of student respondents, 55% made comments contributing to this theme, describing demonstrators as a source of feedback (predominantly verbal) and often as the only source of 1:1 feedback they received regarding their performance in laboratories. It is noteworthy, though not a component of this theme, that many students felt that they did not receive feedback on their laboratory performance or that it was largely generic class feedback.
Demonstrator responses also recognised their role in giving feedback and often listed or described the types of feedback they gave. These included using feedback to guide students to the answer, giving feedback on their use of equipment and giving verbal feedback on the answers in their laboratory workbooks. Many also mentioned giving positive feedback to boost confidence and enhance the learning environment. More demonstrators than students (81% vs 55% respectively) provided comments relating to this theme. procedures and answering questions. Their presence is important because they encourage a certain social dynamic to develop by encouraging conversations about the work we are doing. Also, they talk to everyone. So, even if a student is working alone, they talk to them and make them feel part of the class." Demonstrators' responses in this theme echoed the above responses and some demonstrators also elaborated on the value that a positive learning environment created for students. Only one of the 26 demonstrators commented on the potential of demonstrators to create a positive learning environment relative to the 16 student responses for this theme.

Theme pair 7:
Demonstrators provide support with equipment and with simple queries when self-directed learning does not yield results/ We provide training and support with equipment and techniques and answer question when students have not been able to figure them out with self-directed learning Responses from students in this theme were predominantly from students who described their learning preferences as self-directed or self-motivated and typically described a demonstrator's value in terms of showing them how to use new equipment or in answering questions and queries when they could not do it themselves. Comments were not typically critical of demonstrators but focused less on the teaching led and/or enhanced by demonstrators and saw them as more of a support resource to supplement their chosen self-directed learning style. Demonstrator responses described their role as a demonstrator in the same manner. They also emphasised the importance of selfdirected student learning and that demonstrators should avoid micromanaging or interfering excessively. An equal percentage (27%) of student and demonstrator respondents contributed to this theme pair.
Theme pair 8: Demonstrators are knowledgeable and when they do not know the answer they know where to find it/ We are confident in our knowledge but will ask for help from a peer or lecturer or help students find the answer when we do not know the answer Student responses attributed to this theme (in contrast to theme 2) described a satisfaction with the level of knowledge the demonstrators had and were understanding and content with demonstrators looking up the answer or asking for help from a lecturer when needed.
Demonstrators' responses within this theme expressed a confidence that the knowledge they had was sufficient for their role and that they were happy with seeking help or finding answers collaboratively with students. Some responses actually highlighted how not knowing the answer might be comforting for students and looking up answers together might enhance learning. More than half of respondents from both the demonstrator and students surveys contributed comments towards this theme pair.

What makes an excellent demonstrator?
The same 117 coded student survey responses and 26 coded demonstrator survey responses as described above were used for a second thematic analysis. A visual representation of the themes derived from the combined survey data exploring perceptions regarding optimal demonstrator performance can be seen in Fig. 2.
The themes regarding demonstrator best practice were also commensurate and have therefore been grouped into 6 different theme pairs when described below: Theme Pair 1: Demonstrators build relationships, have fun, and start discussions/ We inspire, motivate, start discussion, and add new perspectives This theme included student responses that suggested the best demonstrators started discussions around scientific topics, careers, further study, or even friendly small talk. Encompassed in these responses was the idea of relationship building. Responses suggested students found this a fun part of laboratories and that it improved the learning environment.
Demonstrator responses also focused upon relationship building but were more angled towards improving student engagement and motivation. Demonstrators felt that they brought extra knowledge and perspectives as part of these discussions. A higher proportion of demonstrators compared to students (50% vs 21% respectively) identified this theme as a characteristic of the best demonstrators.
Theme Pair 2: Demonstrators are warm, kind, empathetic, understanding, caring and compassionate/ We are warm, kind, empathetic, understanding, caring and compassionate Student responses in this theme described the best demonstrators as those displaying traits that helped support students emotionally during their learning or made the demonstrator an approachable source of help. These responses often chastised demonstrators who were aloof, unapproachable, or condescending and said that this made them hard or undesirable to approach for help. An illustrative quote is shown below: "Being friendly, honest on their understanding of the subject. Not judgemental. Patient. Being friendly, and patient helped me to be relax in the lab and ask for help whenever I need which helped me a lot to understand the content." Demonstrator responses in this theme also described the importance of these traits in improving the learning environment and allowing students to comfortably ask for help. Student and demonstrator responses in this theme were equally common with 76% of both cohorts valuing these character traits.

Theme Pair 3: Demonstrators guide not tell / We guide not tell
Both student and demonstrator responses attributed to this theme described how the best demonstrators guide students to the answer and do not tell them. A student described this as: "[the best demonstrators don't] just tell you the answer but to try and help you get there yourself, so you understand the reasoning behind the answer better" There was a key difference between the student and demonstrator responses in that many students mentioned that in addition to the guiding process it was also important in some circumstances to share the actual answer and that failure to do so hindered their learning. Only 12% of student respondents were explicitly attributed to this theme compared with 31% of demonstrators.
Theme Pair 4: Demonstrators are honest when they do not know the answer and get help when they are unsure/ We are honest about what we know and get help when we are unsure The theme pair title was almost verbatim for the responses that students and demonstrators gave regarding the best demonstrators. In addition to many instances describing this as a behaviour of the best demonstrators, a number of student responses highlighted the frustration they felt when demonstrator did not do this and described a lack of honesty as a hinderance to their learning. The majority of demonstrators (81%) explicitly mentioned this honesty compared with 4% of students.
Theme Pair 5: Demonstrators are knowledgeable and well prepared/ We are knowledgeable and well prepared Respondents from both the demonstrator and student survey data highlighted that the best demonstrators were well prepared and knowledgeable. Student responses also highlighted the inverse, criticising poorly prepared demonstrators who were not familiar with the basics of the session. Demonstrators typically did not focus on these negative aspects, though some demonstrator comments highlighted that they only selected to work in sessions where they already had good knowledge. More demonstrators than student respondents (81% vs 50% respectively) stated that the best demonstrators were knowledgeable and well prepared. Theme Pair 6: Demonstrators are interactive, attentive, observant and engaging/ We are interactive, attentive, observant, and engaging Student responses in this theme described the best demonstrators as those who checked in on learning and paid attention to who was struggling and needed support. Demonstrators mentioned that some students would not ask for help unless they engaged with them. Some responses indicated however that excessive interactions or those whilst students were in deep focus were described as harmful to learning. A and B were both ranked questions in the student questionnaire, meaning that students could assign ranks to their answers, i.e., choice number 1 was their first choice and choice number 5 was their last. Demonstrator responses were similar but largely neglected the negative aspects of "too much interactivity" and/or disrupting flow and focus. Demonstrators did mention that they would like more strategies for starting higher quality interactions and for spotting students who need help. Over half of the demonstrator respondents (65%) saw these as traits of the best demonstrators whereas only a third of students mentioned these traits.

Where do students ask for help and who do they perceive their laboratory feedback comes from?
The proportion of responses attributed to different individuals with regards to whom students ask for help in the laboratory, and from whom they perceive most of their laboratory feedback to come from is shown in Figure 3.
Students' first choice when seeking help was to ask their peers (60%) followed by demonstrators (27%). The second choice of help was predominantly the demonstrators (62%). The least popular option (5th choice) was predominantly 'I prefer not to ask' (79%) followed by lecturer (15%).
Students perceived demonstrators to be their primary source of laboratory feedback (64%) with the second most selected option for their first choice being 'I have not received feedback' (20%). The lead demonstrator was most often selected as the second-place option (42%) with other options comparable in frequency. The lecturer was most often selected as a third-place option (43%). Justifications for student choices are summarised in Table 2.

Demonstrator knowledge, importance, and influence
Students and demonstrators were asked to rank the importance of demonstrators to practical sessions and whether they had sufficient knowledge for the role. Their responses can be seen in Figure 4.
Survey questions also explored whether demonstrators were role models and whether their teaching was appreciated. Most demonstrators (73%) believed that their teaching is "often appreciated", with a further 4% "always appreciated" and some (23%) feeling that their teaching is only "sometimes" appreciated. The majority of students considered demonstrators to be some form of role model (Yes 35%, somewhat 53%), with the remaining disagreeing (8%) or uncertain (4%).
On the other hand, 46% of demonstrators considered themselves to be role models, whilst 8% did not and 46% were unsure. Furthermore, conversing with demonstrators has made 73% of students think (strongly or somewhat) about embarking upon a Master's degree or PhD.

Discussion
Students and demonstrators are largely aligned in their perceptions regarding the teaching and learning provided by demonstrators in the laboratory setting. Similar to findings by Muzaka (2009) andPark (2002), our research showed that students and demonstrators recognised demonstrators to be an engaging and approachable teaching force. We also found that students perceived demonstrators to be supportive and builders of important relationships with students, helping to create a positive learning environment. Demonstrators were also recognised as predominantly good quality educators who help build knowledge and develop critical thinking. However, both parties agreed that sometimes demonstrators are underprepared and their quality inconsistent.
Underprepared demonstrators could therefore be hindering student learning and undermining faith in the demonstrator team with poor explanations or incorrect advice. A possible solution, suggested by demonstrators in this study, would be to provide more sufficiently paid preparation time prior to teaching. It is understandable why some demonstrators, who are already busy with their own research, feel begrudged to prepare sufficiently when they are perhaps unfairly remunerated. Students and demonstrators further agree that the best demonstrators are personable (rather than condescending or unapproachable), honest, knowledgeable, well prepared, observant, and able to initiate discussions. They also believe that demonstrators should help guide students rather than provide answers immediately. However, students have highlighted that demonstrators should share answers with good explanations if a student is unable to be guided to the solution, and that failure to do so can damage the student-demonstrator dynamic. Students further highlighted that demonstrators should know when to ask questions and when not to interrupt e.g., when a student is focused. However, from my past experience of working as a GTA I sympathise, as it is sometimes difficult to know when a student needs help, especially if they are shy. Thus, the introduction of a traffic light system could be beneficial in ensuring students get the level of interactivity they want from demonstrators. However, this does run the risk that some students may choose to avoid interactivity completely and in turn miss out on key learning experiences.
We can describe the studentdemonstrator dynamic as one that sits somewhere between student-peer and student-lecturer, as students consider demonstrators to be high quality educators and role models, but also as relatable individuals who are more approachable and accessible than lecturers.
In the laboratory, demonstrators are the most recognised source of feedback and are students' second choice when asking for help (after their peers). Demonstrators thereby play an important role in student learning as they put students at ease, are more knowledgeable than student peers and provide a key source of feedback. In the UK higher education sector where feedback is often poorly scored in the National Student Survey (Williams et al., 2008), this opportunity to impact student learning should be shared more widely with all staff. However, what we do not know is the quality of demonstrator feedback and so perhaps more should be done to train, support and measure the feedback given to students. Demonstrators also stated that they would benefit from strategies for identifying students that need help and for initiating better quality discussion. Thus, training in these areas should be included in their training programmes. This could be more generic training regarding student body language, or session specific training which points out specific signs in students' experimental set up or workbooks as indicators of a student who has not grasped the concepts fully. Moreover, conversations between students and demonstrators appear to inspire students to consider postgraduate study; this supports similar findings by Park (2002). With many students considering demonstrators to be role models, it is perhaps unsurprising that they would want to explore these topics with them. However, despite positive student perceptions, demonstrators themselves do not always feel appreciated by students. Thus, to boost demonstrator morale and maintain a positive learning environment it would be beneficial for students to regularly vocalise their appreciation.
To further ensure demonstrators feel more valued, students should have the option to consistently leave positive (or negative) feedback for an individual(s).

Concluding remarks
This research provides a blueprint for the pedagogic and personal skills required for excellence as a practical demonstrator.
Students recognise demonstrators to be approachable, relatable and talented educators who provide feedback and create a positive learning environment but would like more consistency in demonstrator performance. Demonstrators do not always feel appreciated by students, or fully prepared for teaching and suggest that they would benefit from more paid preparation time and training to perform to their potential. Universities should do more to recognise and maximise demonstrator potential and improve their teaching experience by ensuring that demonstrators feel valued and providing suitable access to training in the areas of feedback, identifying which students need help (and when) and relationship building.