A Bottom Up View of the Needs of Prospective Teachers

by Peter Watkins

University of Portsmouth (Portsmouth, UK)



This article considers the provision of short pre-service initial teacher education courses, such as the Cambridge ESOL CELTA (Certificate in English Language Teaching to Adults) course. Its starting point is the experiences of the prospective teachers who undertake the courses. A series of interviews were carried out and the key points of the discussion are presented and analysed here. Tentative conclusions are drawn about the needs of these teachers and suggestions of what might constitute “best practice” in the field are made. This includes the centrality of the teaching practice (TP) sequence, the deliberate reduction of hierarchies through language choices, and supporting reflection on previous teaching when planning, as well as in the traditional “feedback” phase.



The aim of this research was to examine the experience of prospective teachers when they undertake short, pre-service teacher education courses, such as the Cambridge ESOL CELTA (Certificate in English Language Teaching to Adults). The focus on participant experience is particularly relevant as the courses are often characterized as being both demanding and stressful. By focusing on participants it takes a “bottom up” view of teacher education, in that it considers the needs of prospective teachers and best practice in providing such courses, in the light of the experiences of those who follow the courses. It should be noted that this is not to contest the validity of more top-down approaches that identify prospective teachers’ needs (such as Richards, 1998) but instead offers a different starting point from which potentially different priorities may emerge.

I will begin by outlining the background of the research and the type of course under consideration.


There are many types of initial teacher education courses available, depending on the context in which someone intends to work. This paper looks at short pre-service courses with around 120 hours of input and six hours of assessed TP. The best known is the Cambridge ESOL CELTA course and for this alone there are currently over 286 approved centres around the world, with around 12,000 candidates following CELTA courses each year. Wright (2010, p.272) comments as follows: “The CELTA has gained a good reputation for providing beginning English language teachers with a sound basic grounding in teaching and content knowledge.”

The Trinity College London Certificate in TESOL is another well-known example of such a programme. Both these courses are highly regarded largely because of their very practical nature, with a large part of assessment based on the teaching practice component. They both offer a generic introduction to ELT, in the sense that no specific eventual teaching context is envisaged. The British Council recognises both the Cambridge ESOL and Trinity College qualifications, making them an ideal starting point for someone at the beginning of their teaching career who wishes to enter the private sector. Senior (2006, p. 38) points out, “[a] CELTA Certificate opens doors: a quick surf on the Internet reveals the number of jobs available to an English speaker with an English teaching certificate and a willingness to travel.”

While these short courses have gained a good reputation, their intensive nature has some drawbacks. Not only does the content have to be carefully selected to prioritize those aspects of practice considered to be most important, but also the intensive nature can lead to participants finding courses stressful, as Senior (2006, p.40-41) reports in her study. Indeed, centres draw candidates’ attention to potential stress at the application stage. Through researching participants’ experiences of these courses it may be possible to reduce the impact of any negative affective factors.

All of the interviewees in this study had done either a CELTA or Trinity Certificate course.

There were 14 interviewees in total and they had varying degrees of post-course experience. The respondents gave their informed consent to their comments being used for publication. They are not, of course, identified by name but are sometimes labelled using a letter for ease of reference. Not all are quoted directly, but their views are represented. One point to note is that nearly all the respondents were positive about their experiences overall, often comparing their initial training experiences favourably with other training experiences.


Using routines

Intensive CELTA courses tend to start TP from the second, or even in some cases, the first day. Here is the experience of one candidate (J):

I had to teach the second day of the course. I was so nervous–couldn’t eat–nothing.

 The teacher [trainer] just said ‘copy me’–and he showed me [before the lesson] what  to do–what to say–what to write on the board and when–everything. I kept running  through it in my head and [in the lesson] did what he had done. It turned out OK.

That first lesson was a big, big step.

It is clear that this prospective teacher found it useful to copy his supervisor and learned enough to survive his first lesson through a largely behaviourist approach. He watched, practised (or at least mentally rehearsed) and then performed the lesson.

Many new teachers like the security of having a model to copy, as A, a recent trainee, testifies when being asked about her strategies for dealing with grammar lessons during her course:

We were taught to follow a PPP [Presentation, Practice, Production] method. I  thought it was really good to learn one way quite thoroughly. We practised it loads.  They mentioned TBL [Task Based Learning], but most of us just wanted to do one  thing we could rely on so I think we all stuck with that [PPP].

Leaving aside the merits or otherwise of the choice of method, it is clear that this trainee valued the apparent simplicity of having a single method to follow, which was learned thoroughly through repetition (“we practised it loads”). Of course, there is a negative aspect to this as well because if one method is extensively practised, it leaves little time for others to be considered. Trainee A says that TBL was only mentioned so it may be that she would favour PPP even in a context where a stronger form of communicative language teaching was appropriate.

The trainees referred to above felt they learned in a fairly mechanistic, behaviourist tradition. Most initial teacher education courses are described as training courses, so there is little surprise that these needs are catered for and have a place. Indeed, learning such routines can be very useful. Thornbury (2006, p.201) says, “Learning to teach is largely a case of acquiring a repertoire of useful, automated routines that can be adapted to different classes, levels and circumstances.”

Learning routines may do more than give new teachers security. It could be argued that they work in a similar way to learners learning prefabricated chunks of language. Not only do they add fluency to the immediate performance, but they also provide data for later analysis, and therefore, promote future development. Of course, there is no definitive way of calculating how many, or which particular, routines a new teacher may need. However, it would seem sensible to start with those that articulate core teaching competencies and are appropriate for the demands of the teaching practice context.


Thoughtful implementation of routines

Of course, when people are thrown into teaching with little preparation, things can go wrong as well as right. Another teacher, R, nearly two years into a teaching career at the time of the interview, recalled his worst experience of the course. The main focus of the lesson was listening, with some vocabulary to be pre-taught:

It just went wrong. It seemed so easy when I looked at it–I knew what to do but it  turned out really boring. I wanted to get the students to say the word first [to elicit the  vocabulary] and it just took so much time. …I got a bit rattled… [I] didn’t have time  to do the listening properly. My teacher [trainer] said the lesson was borderline–but I  knew–it just felt wrong.

When asked why it was so important to elicit the vocabulary, R replied that at the time that was what he believed was the “right” way to teach vocabulary: “It had been hammered on the course that they [the students] said it [the new word] first and then the teacher. I was just sure I had to do it that way.”

The experience of R shows that there is more to learning to teach than following routines. It is important to know which routines may be appropriate at any given time, and also, as Thornbury (cited above) says, to adapt them “to different classes, levels and circumstances.” This suggests that while mechanistic learning has a place in early teacher education, it is not sufficient.

Another example comes from teacher L, interviewed approximately six weeks into her teaching career. She commented that she was doing a lot of extra planning because

I keep needing to find a way of doing a gen sit [generative situation–the method of  language presentation she had used extensively on her course]. The book I’m using  just doesn’t have them so I put them in before doing the book.

As in the case of A above, this suggests that while teacher L may have thoroughly learned one routine, she had left her course over-reliant on one formula and perhaps unaware of other possibilities of language presentation (such as those exploited in the course book she was using). Clearly, this was having a negative impact not only on her workload but most probably in terms of development, too. Rather than spending time reflecting on her early experiences she was perhaps unnecessarily investing a huge amount of time and energy in designing material that fitted a known, comfortable routine.

This argument moves the view of the prospective teacher away from being a mere imitator of others towards Schön’s conception of professionalism (for example, 1983), where a professional (in this case a teacher) is faced with unique situations that training cannot fully anticipate and therefore provide answers to. Instead, the teacher needs to identify challenges, form hypotheses to serve as solutions, and then test those hypotheses. This is the notion of the “reflective practitioner.”

But can intensive initial teacher education courses hope to develop embryonic reflective practitioners? Here are the recollections of the support offered for planning from teacher K, three years on from doing a course:

I remember a workshop lesson [input session]. We all had to sit and work on our  plans for the next [TP] lesson. The tutor went round and kept asking questions…Will  that be too difficult? How can you make it easier? Which students [in the TP class]  will be able to do that? That kind of stuff.

In this encounter, the tutor fills the role of the “more skilled other,” as envisaged in a social constructivist view of learning (for example, Vygotsky 1978). Through using questions, the tutor supported (“scaffolded”) K’s development in thinking of how to adapt learned routines in the light of her knowledge gained from previous teaching. Not only that, but because the support was targeted at the issues that K was herself thinking about, it may well have been more “salient.” Pennington (1996, p.340) remarks, “Attempts to influence teachers’ behaviour will have an impact only in areas where the input is valued and salient to the individual, and where it is congruent with, and interpretable within, the teacher’s own world of thought and action.”

Such saliency of input can perhaps be best achieved at the planning phases of a lesson when the prospective teacher is grappling with a particular issue, making any input in that area highly relevant. Of course, this implies a degree of individual attention that may not always be feasible on a large scale, but earlier course input may become more meaningful at this point, even if it is only briefly referred to or prompted by the trainer. The second part of Pennington’s quotation suggests that our acceptance and understanding of the routines learned in early training will be influenced by our existing beliefs and values. Here is an example, again from teacher K: “I know I’m a good language learner – I speak four languages. And I know that grammar is important. It’s how I learnt…So yes – grammar is [a] really important part of teaching.”

The interesting thing here is not only does teacher K strongly believe in the efficacy of explicit grammar teaching, but that her belief springs from experiences that predated the course followed. Any attempt to modify this belief will involve reshaping, or perhaps challenging, those existing views. So while some prospective teachers value copying models, they also construct knowledge, interpreting new input through their own schema of what it is to teach.

All the interviewees in this study saw TP as being central to their experience of the course and learning to teach. We will now consider the role of TP in more detail.


The TP experience

The model of reflective practice is most explicitly promoted within both the CELTA and Trinity courses in the TP cycle. Lessons are planned, taught, and reflected on. The reflection phase is usually a group discussion, led by the tutor with as many as five or six trainees present. Although this can be a stressful experience, it is usually highly valued by the prospective teachers. Here are the thoughts of teacher Z, several years on from her course, on being asked which aspects of it she had found most useful:

The TP–no doubt. That was really tough–before the first lesson I was petrified, but  that was where I learnt the most – in the TP and the feedback afterwards too. I just  always tried to do what the tutor told me for the next time and get better. It was so  good.

This emphasises that the experience of TP is a vital part of the course. It is interesting to also note this trainee’s use of “we were shown,” which suggests that she saw her learning as being at least partly based on copying the prescribed techniques. Also, this former trainee characterized her experience as being very positive, despite her initial anxiety.

Feeling some anxiety when performing new skills in front of others is entirely natural, of course, but not all trainees recall their experience with affection. A more recent trainee, again teacher A, who had only just completed her course, had this to say:

The TP was really good. I did learn a lot, I think. And I loved the students. But one of  the trainers just seemed to have it in for me. I don’t know why. I thought I was doing  OK–my lessons seemed like the others [those of the other trainee teachers], but there  was always a problem with mine. I just sat there afterwards in the debrief and made  myself smile and nod. I always knew what was coming though. I just tried not to take  any notice, but it was hard. He always said whatever I did was wrong but then always  passed the lesson. It was weird.

It should be noted that this is the prospective teacher’s perception of the feedback she was given and it is possible that the tutor’s would be quite different.

Teacher S provides a slightly less personal but similar account of the potentially stressful nature of the feedback sequence in TP:

I was OK. My lessons were all OK and I learnt a lot for sure. But there was one  woman who got quite a lot of stick [criticism]. I guess the tutor thought it was  necessary, but it was pretty tough, pretty full on. She [the trainee] was probably the  age of my mum and crying–more than once. It wasn’t nice for sure. It was  embarrassing really, for us, and must have been hell for her. Humiliating, really. I  don’t think she finished the course.

These accounts highlight that there are potentially widely asymmetrical power relationships between tutors and prospective teachers. The tutors are both acknowledged experts in the eyes of the teachers and also their assessors. As Borg (2009, p.165) intimates, this can sometimes lead to a desire to act in a way that will please the tutor, and perhaps this is implicit in the comments of teacher Z:“I just always tried to do what the tutor told me.”

For teacher A, however, it seems to have led to a conscious attempt to create emotional distance between herself and the experience: “I just tried not to take any notice.” She has tried to detach herself from the comments on her lessons rather than allow herself to fully engage the reflection process. It is clear as well that the experiences of the trainee described above by teacher S did not lead to a successful learning experience as she did not complete the course.

The notion of a reflective practitioner has become a dominant model in teacher education. It offers a powerful and persuasive vision of teachers directing their own development. However, not all reflection is equally willing and open and therefore equally useful. In the context of initial teacher education, teachers need to be given time to think about their lessons before being forced to speak about them in order for reflection to be maximally beneficial. In this regard, teacher B commented:

It was hard, I guess, talking about the lessons. I sometimes thought I was guessing  what the tutor had written. It was easier if I was the first one on [the first trainee to  teach that day] because I then had a bit of time to think it through. Otherwise, my  head was all over the place.

Although centres can find it difficult to timetable TP in such a way that there is a substantial break between teaching and feedback, it is crucially important as a way of developing genuine reflection. When trainees do speak about lessons, it is important that this is done in a supportive atmosphere. This can be achieved in part by reducing the power asymmetry in the relationship of trainer and trainee. This may make the negative experiences of the type reported above (teachers A and S) less likely and may have additional beneficial impacts. Coates (2004, p.96) reports research by West (1998) comparing the language used by female and male doctors when talking to patients. West found that when telling patients what they had to do (issuing “directives”), female doctors achieved a higher proportion of compliant responses than their male counterparts. This was ascribed to their use of less direct language. For example, female doctors frequently framed directives as proposals for joint action, as in “Okay? Well let’s make that our plan” rather than through using bare imperatives, which were used more by males. If less direct, and more inclusive, language use is successful in doctor-patient interaction, then there is no obvious reason why it would not be in the feedback sequences that follow teaching practice. If power asymmetry is expressed less explicitly, it may also lead new teachers to be more willing to take risks in their teaching as they will fear criticism less. They may try new routines and adapt others and the outcomes of taking such risks will give more data for reflection and so aid development. However, more research focused on this particular area would be needed to confirm this.

We can see that prospective teachers find TP both central to their learning experience and daunting. Here are another former trainee’s written notes on how her experience was managed, and it gives an insight into one way of alleviating the worst aspects of a potentially stressful situation:

Daunting task of standing up in front of a class was made easier by short (20 minutes)  sessions at first building up to 45 minutes. The feedback from these sessions was  invaluable. I think the course should include as many obs [observed teaching  practice] as you can squeeze in as I found this the most useful.

The simple solution of using short introductory teaching experiences (20 minutes, in this case), which call for the trainee to draw on fewer routines, is seen as having alleviated her anxiety to some extent. Also, shorter teaching episodes obviously constitute a relatively smaller percentage of assessed teaching and are therefore less crucial to the final outcome of the course. This may also help to reduce stress.



So what have we learned from looking at the experience of prospective teachers? Much of it is very positive. “It was simply brilliant” was the comment of one teacher, which was echoed by nearly all the others, even where they expressed reservations over particular aspects of their training. They feel that their courses are a genuinely useful preparation and that much of what they identify as being important reflects current practice. Nearly of all of them see the TP sequence as the most influential part of the course, and when asked to recall experiences from the course, it is these sequences that remain vibrant in their minds. Ideally courses need to maximize this experience, and this could be achieved by tailoring input sessions to follow on from TP, or at least allowing time to analyse critical incidents which arise from teaching.

The respondents in this study also valued having models to copy, and as discussed above, these can be useful. However, they are not the only means by which teachers learn, and it is important that the need to question and adapt routines is stressed. Indeed, it is this questioning that will lead to reflective practitioners emerging from their training and early experience. Perhaps routines need to be examined and constructed in their constituent parts, so that the elements of each can be used in a variety of situations. Rather than leaving a course confident of being able to implement a generative situation presentation phase, perhaps prospective teachers would benefit from seeing themselves as equipped to draw attention to meaning and form, wherever this becomes necessary. In addition, because prospective teachers come to courses with beliefs about learning and teaching in place, it is useful to discuss how new techniques fit into existing beliefs.

Another point to note is that reflection is appropriate not only after a lesson, but also when planning for the next lesson. At this stage the prospective teacher has made less personal investment in the teaching process than they have after they have performed the lesson. The nature of assessed TP also means that the tutor may be seen as being more judgmental after teaching. This may make the reflection more open, objective and less emotional than can be the case immediately after lessons. Courses could work towards this by having planning workshops, for example. In addition, reflection is likely to be more open and honest when the power hierarchy of trainee and trainer is reduced. The language choices made by the trainer are very important here. It is probably true that TP will inevitably be daunting to some extent. However, by starting with very short lessons, or splitting a single lesson amongst three or four prospective teachers, it may become less so.



Borg, S. (2009). Language teacher cognition. In A. Burns, & J.C. Richards (Eds.),The Cambridge guide to second language teacher education (pp.163-171). New York: Cambridge University Press.

CELTA Centre FAQs. Retrieved March 19, 2011, from the Cambridge ESOL website: http://www.cambridgeesol.org/exams/exams-info/faqs/celta-centres.html#4 Coates, J. (2004). Women, men and language. Harlow: Pearson Education.

Pennington, M.C. (1996). When input becomes intake: Tracing the sources of teachers’    attitude change. In D. Freeman, & J. C. Richards (Eds.), Teacher learning in language teaching (pp. 320-348). New York: Cambridge University Press.

Richards, J.C. (1998). Beyond training. Cambridge: Cambridge University Press.

Schön, D. (1983). The reflective practitioner: How professionals think in action. London: Temple Smith.

Senior, R. (2006). The experience of language teaching. Cambridge: Cambridge University  Press.

Thornbury, S. (2006). An A-Z of ELT. Oxford: Macmillan.

Vygotsky, L. (1978). Mind in society. Cambridge, Massachusetts: Harvard University Press. Wright, T. (2010). Second language teacher education: Review of recent research on practice. Language Teaching, 43(3), 259-296.


About the Author

Peter Watkins is a Senior Lecturer at the University of Portsmouth, where he has worked extensively on both pre-service and in-service teacher education programmes. In addition to a number of articles, he has written Learning to Teach English (Delta Publishing, 2005) and is the co-author (with Scott Thornbury) of The CELTA Course Trainee Book and The CELTA Course Trainer’s Manual (Cambridge University Press, 2007).

Leave a Reply