Lesson study outcomes: a theoretical model

John Paul Mynott (Central Primary School Watford, Watford, UK)

International Journal for Lesson and Learning Studies

ISSN: 2046-8253

Publication date: 8 April 2019

Abstract

Purpose

Lesson study (LS) research is disadvantaged by a lack of clarity surrounding the potential outcomes an LS cycle can produce for participant learning. The purpose of this paper is to set out a model of the potential outcomes an LS cycle can achieve. The model identifies the limitations that can occur in LS groups and how these limitations impact on the overall outcomes for participants.

Design/methodology/approach

Case studies are used to exemplify the different outcomes in the model taken from five years of LS work in a primary school in England. The case studies shape the four different outcomes of the model, defining and contextualising the attributes and characteristics of each outcome.

Findings

The model presented indicates that there are four key outcomes for LS cycles, with the most common outcome being a form of limited learning. The paper explores the limitations of time, collaboration and expertise to articulate how each of these limiting factors has a bearing on the overall outcome for an LS cycle.

Research limitations/implications

The model is currently based on a singular educational setting. This means that each outcome needs further exploration through wider LS work in order to clarify and refine the outcome model.

Practical implications

The outcome model will support the development of a shared vocabulary for discussing LS cycles. By articulating where on the outcome model an LS is, it is possible for researchers to discuss how to reduce the impact of limitations and other challenges to LS, enabling research to develop a more evaluation-led approach to using LS.

Originality/value

The outcome model supports LS researchers in articulating the outcomes of their LS cycles with a shared vocabulary. It addresses understudied areas of LS research, namely failed and dysfunctional LS cycles and identifies that while an LS can bring the potential for participant learning, the cycle outcomes are the starting point for participant change.

Keywords

Citation

Mynott, J.P. (2019), "Lesson study outcomes: a theoretical model", International Journal for Lesson and Learning Studies, Vol. 8 No. 2, pp. 117-134. https://doi.org/10.1108/IJLLS-08-2018-0057

Download as .RIS

Publisher

:

Emerald Publishing Limited

Copyright © 2019, Emerald Publishing Limited


1. Introduction

Lesson study (LS), as an emergent method of professional development, has an underdeveloped research base. This research base is particularly insufficient in considering the impact and evaluation of LS cycles (Xu and Pedder, 2015; Godfrey et al., 2018; Seleznyov, 2018). Xu and Pedder (2015) found only nine articles addressing teacher and/or student learning and Godfrey et al. (2018) and Seleznyov (2018) both suggest that little progress has been made in addressing a fundamental limitation of current LS scholarship. LS outcomes and evaluations need to be more clearly detailed, discussed and developed. Godfrey et al. (2018) have set out their model of evaluation-led LS, a welcome development and this article addresses LS outcomes.

This paper is a theoretical paper based on the author’s experience of LS since 2012 in a primary school in England. The author is a school-based leader and researcher and all the examples provided in the discussion are from LS work conducted in the author’s school. Case studies will be used to exemplify and elaborate the development of Mynott’s (2017) LS outcome model. Mynott (2017) proposed that there are four broad, key outcomes for LS cycles; depicted in Figure 1.

This paper aims to further refine Mynott’s (2017) outcome model so that it can be a lens, which can be used by LS practitioners, participants and researchers, to analyse and explore LS cycles. In the research and creation of this paper an LS outcome framework was developed. This framework (Table AI) further details the characteristics of each outcome level. As a result, the outcome model is hoped to enable reflection on LS cycles, and help identify their characteristics, which in turn may strengthen future LS work.

Each outcome, within the model, is termed as a type of dissonance, with a variety of limitations from absence and dysfunction to limited and rich learning dissonance. The use of the term dissonance relates to the moments of learning identified in previous LS research, (Pella, 2011; Dudley, 2013; Mynott, 2017) but Mynott’s outcome model suggests that moments of learning, in LS, are actually only potential moments of participant learning. It is only, if these moments of dissonance are sustained and acted upon, within the LS and beyond, that they are enabled to be integrated into individual participant practice and/or organisational changes. Figure 2 shows the processes that lead to potential participant learning within LS work.

Actual participant learning occurs after LS cycles are completed and would be evidenced in the future work of each participant and/or within their organisation. Guskey (2000) presents five levels of professional development impact and LS cycles would only allow Level 1 participant reactions to occur during the LS cycle. Level 2 participant learning, in Guskey’s model, can be likened to the notion of potential learning, just presented, as it is only through the movement through the later levels, of Guskey’s model: organisational change and pupil outcomes, that participant learning becomes active and impactful. Using Guskey’s (2000) sequence of measuring impact of professional development it is only when potential moments of learning cascade into impacting in these wider areas, beyond the initial LS, that they may lead to actual learning. Therefore, it can only be considered potential learning until it develops into future practice.

Little research has been conducted beyond LS cycles and so the overall impact of LS to enable actual participant learning is, as yet, unknown. To clarify this point further, cognitive dissonance research suggests that an individual might know something through a moment of cognitive dissonance (Festinger, 1957) but awareness does not always result in a change in behaviour. In a participant’s case, they might have a moment of realisation about their pedagogic practice, but this may not change their work. Aronson (1999) suggests that individuals can find consonance with a discordant thought using other justifications, such as “I do not have the time to do it that way” or “everyone else is doing it this way”. Indeed, cognitive research on coffee choice, indicated that while people knew fair-trade coffee was more ethical, they did not select it at the counter when they needed to ask for it (Chatzidakis et al., 2007). Thus, ease became the excuse to undertake an action that was discordant with their knowledge base. Consequently, due to the lack of LS studies concentrating on the impact of LS, it is not yet possible to talk about actual participant learning from LS, so this article will talk about potential moments of teacher learning and how these occur in different ways within LS.

The rest of this paper will describe each outcome level on Figure 1, identifying the limitations and characteristics of each outcome level.

2. The outcome model

The range of possible LS outcomes is modelled in Figure 1. Mynott’s (2017) doctoral research provided the foundations of this model and case studies from that research and subsequent LS cycles, undertaken since 2016 (Mynott, 2018a, b; Mynott et al., 2018) further exemplify the model. Researcher reflections will provide insights into each outcome and describe how LS proffers moments of potential teacher learning, or not, at each outcome level.

The outcomes of the model are distinguishable by the variance of the following limiting factors:

  • time (both physical time and degree of commitment);

  • collaboration (space to collaborate, skills to collaborate, desire to collaborate and understanding of professional conflict); and

  • expertise (pedagogical, content, subject and LS method knowledge). (Mynott, 2017)

Each limiting factor is a continuum and as such there can be variations within each outcome. This is particularly important at Outcome 3, which will be further sub-divided, in the subsequent discussion.

The outcomes are based on dissonance within LS leading to moments of potential learning – Figure 2. This model focusses on dissonance a stimulus as learning – a spark-point. It then furthers this by defining how the treatment of the initial dissonance can then impact on how that moment can be extinguished, deflected or grown. Therefore, dissonance is an overarching term because as previously stated learning within LS is harder to establish. As such, dissonance is the most accessible way to discuss how potential learning in LS may be generated. This is particularly important as the body of literature in LS does not sufficiently provide information on how learning occurs within and after LS. Therefore, the theoretical model focusses on what can currently be described, based on examples from LS in a school in the UK.

This paper aims to start a rich conversation about potential learning outcomes in LS and addresses outcomes which are missing, due to being perceived as less successful. Outcomes 1 and 2 are examples of these rarely discussed outcomes, and might be considered failed LS cycles if they do not offer any moments of potential teacher learning. Outcome 3 is the most discussed outcome in the current published literature, with most of the limited articles published (Xu and Pedder, 2015) being able to be placed at Outcome 3. Currently, Outcome 4 is potentially a theoretical outcome, as will be discussed later, due to the need for it to be a perfect LS cycle. A perfect study cycle in the sense it does not have any limiting factors, which from the author’s experience of schools and LS, is unlikely to be possible at this time.

2.1 Findings and analysis

For each case study, audio recordings were made of all review sessions and analysis followed a pattern of initial listening, transcribing. In the 2013–2014 case study year this initial listening and transcription developed themes. Affirmation, conflict (dissonance), description and sharing expertise (Mynott, 2017) were themes added to the ones presented by Dudley (2013) and Pella (2011). As the LS cycles developed, the analysis built on Mynott’s (2017) and became more focussed under the themes of collaboration, expertise and time with dissonance points encompassed throughout each of these themes. The principles of applied thematic analysis (Guest et al., 2012) were applied to the initial transcriptions and these themes were then regrouped under the outcome model categories and this provided the framework for the detail provided on the characteristics identifiable at each outcome level. Further details on the treatment of research data, at the transcription stage, can be found in Mynott (2017) for case studies undertaken prior to 2016 and in Mynott et al. (2018) and Mynott (2018a) for case studies undertaken since 2016.

The regrouped transcription themes were then reworded to form descriptions of characteristics for each outcome level and formed an outcome framework. This framework was then used to position the case studies within the outcomes. A copy of the finalised framework is placed in Table AI. To position each case study, the transcriptions and audios were revisited, for all the examples within the discussion and other case studies undertaken within the same setting. The framework was highlighted after this revisiting. The highlighted frameworks are included alongside the case studies to exemplify the position each case study has within the framework. At times, as had been originally noted by Mynott (2017), some case studies had review sessions which belonged to different outcome levels. One of these is included in the subsequent discussion as it is useful to exemplify two different parts of the outcome model.

As part of this reviewing process, it became evident that there were a number of outcome examples at Outcome Level 3 and the amount of dissonance and subsequent potential learning within these examples varied considerably. As a result, Outcome 3 was sub-divided into three further sub-categories which will be explored in the discussion. These sub-divisions have been added to the final version of the outcome framework in Table AI. The additional section, defining the sub-divisions, was generated through the analysis as a means to provide further clarity on the diversity found in Outcome 3.

3. Case studies

Case studies will be used to exemplify each level of the outcome model. These case studies have been selected as they usefully exemplify the outcome levels. The case studies used hold a duality in this paper as they both exemplify the outcome levels and show how the framework might be used to position LS cycles within outcome model. The case studies will be further contextualised as they are introduced. These studies are lesson studies undertaken in a primary school in England, between 2012 and 2018.

The LS models used are from two LS methods. The first is LS method as described by Dudley (2014) and Stepanek et al. (2007), which is based on the establishment of the research lesson, its planning, teaching and review. The second is an extended preparation model (Mynott, 2017, 2018b) which extends the initial phase, preparation, of Dudley (2014) and Stepanek et al.’s (2007) cycle to provide a longer preparation time prior to the first LS lesson in the cycle. This extended preparation form of LS is still being refined and the cases given here are taken from two full cycles of extended preparation LS.

Throughout the establishment of the outcome model, there is an evolution of the author’s LS work and while many mistakes were made in early implementations of LS within the case study setting (Mynott, 2017), these have been refined and developed in subsequent work (Mynott, 2018b). These mistakes and problems demonstrate the variety of outcomes in the model as are useful for others undertaking LS work. In each case used, all participants have been anonymised and consent has been sought for their involvement in this publication. The researcher’s jottings are based on extracts from research journals which provide further development of established points and are a means of providing the reader with a greater sense of detail for the limitations and challenges in each case.

4. Outcome model

4.1 Outcome 1: absence of dissonance

In 2013–2014, four teachers were engaged in an LS cycle exploring pupil talk. Jasmine (senior leader), Libby (experienced teacher), Misha (experienced teacher) and Hanna (inexperienced teacher) were the participants. They completed two LS lessons in their cycle, but analysis of their reviews indicates that they maintain high levels of affirmation and ego-support (Mynott, 2017). As a result, they did not focus their enquiry and avoided any conflict or difference of view within the LS team and can be defined using Achinstein (2002) as conflict-avoidant:

Misha arrived to review the lesson […].

The discussion was affable but at the end of the review there was no commitment to another lesson in the Lesson Study cycle.

No further lessons took place.

Extract 1: summary of lesson two: lesson study group 2013–2014 (Mynott, 2017).

The summary of lesson two typifies the theme of absence in Outcome 1. It is the absence of dissonance that is the key characteristic of this outcome. When an LS cycle does not produce dissonance and subsequent moments of potential participant learning, it will have some or all of the traits listed in Table I:

Continued […]

The group continues to talk about the lesson, focusing in their speech on supporting Libby through positive affirmation of her teaching. There are many smiles, and laughter punctuates the group’s discussion.

Jasmine: That is what was interesting as your focus was on maths and your maths teaching was fabulous and then you go/ went to the role play area.

Extract 2: lesson study group 2013–2014 (Mynott, 2017).

The two extracts taken from the LS Group 2013–2014 (Mynott, 2017) identify a range of key characteristics identified in Table I. These are highlighted in grey in Table I.

As Mynott (2017) identified, this group used affirmation to protect individuals. Jasmine is doing this in extract 2, where she praises Libby’s Maths teaching. This is an example of dissonance reduction as it meant that anything challenging or difficult to say was diminished by ego-reinforcement. This affirmation not only reduced any dissonance that was generated in this LS team but also meant that the group were unable to draw on their expertise to build to any moment of potential learning because they were concerned that such dissonance might be taken personally by Libby (Mynott, 2017). As such they then found little value in their LS work. A fact that is exemplified in their unplanned ending of their cycle after their second lesson – they just did not organise a third lesson (Mynott, 2017).

The principal reason for the example reaching an Outcome 1 for their LS work was in the failure to establish the safe space for professional collaboration. The LS group had not understood the LS protocols (Dudley, 2014; Stepanek et al., 2007) and had not generated a professional space through which they could promote professional conflict without harming an individual’s feelings. Fernandez and Yoshida (2004) showed that LS work in Japan is about self-reflection and critique of a teacher’s own practice within the structure of a collaborative group. This can also be seen in Pella’s (2011) article where a confident and highly qualified teacher goes into a LS with three other equally confident and well-qualified teachers and learns that she is not the top teacher in the LS group, which is a contrast to her normal school experience, where she is considered the best teacher. In both writings (Fernandez and Yoshida, 2004; Pella, 2011) the participant is not avoidant of this reflection process, they want to get better and they realise that their continuing professional development will be a journey. In the Outcome 1 LS group example, the participants are avoidant of this reflective dissonance. As a result, they did not enable Libby to reach the same moment of potential teacher learning, that are discussed in Pella (2011) and Fernandez and Yoshida (2004) and the protocols (Dudley, 2014) for how their group discussions would work were not established. Mynott (2017) identified that this LS team would require addition external facilitation to provide external professional conflict (Achinstein, 2002). This external conflict could then support the team’s establishment of moments of dissonance being generated and sustained towards creating potential moments of participant learning. Facilitation in LS is also under described in the literature and Perry and Boylan (2018) indicate that this is a prevalent challenge to improving facilitation in professional development for professional development facilitators. However, without facilitation and external challenge an LS team might only achieve an Outcome 1 because they may not have the individuals, like Pella (2011) and Fernandez and Yoshida (2004) describe, who are able to take on board professional conflict and use it productively.

In 2017–2018, a proposed Year 4 LS was developed to explore rhythm and echo reading to see if this developed pupils’ expressive reading. Two Year 4 teachers (both experienced) and a senior leader were involved in the initial generation of this LS. The senior leader took the role of facilitating the LS and organised the initial general reading around the theme to be explored:

We had a discussion about developing reading instruction in Year 4 […].

I sent them the reading materials on rhythmic reading, and developing reading instruction and asked them to select a date for the first planning meeting.

No lessons took place.

Extract 3: Summary of an attempted lesson study 2017–2018.

The 2017–2018 LS group, above, focussed on reading in Year 4 but did not start. The group were using the extended preparation model, but the LS did not start as the group could not or did not find the time to collaborate. The senior leader – facilitator was running three LS groups in 2017–2018 and while the other two groups undertook their full cycles, this group did not commence. The reasons given were conflicting schedules and not being able to find the time to work together. In this case there was an avoidance of the potential additional workload an LS cycle might generate. This links to avoidance as a means of generating absence as if the team did not commence their work they did not need to commit to discussion of any dissonance produced. Overcoming the perceived workload vs the benefits of an LS is always a challenge as participants need to be invested in the cycle to benefit from it. Again, a further exploration of facilitation and how a facilitator can explore establishing an LS team with a more reluctant group would be useful for avoiding non-starting or early-ending LS cycles and thus preventing more Outcome 1 outcomes in LS work.

4.2 Outcome 2: dysfunctional dissonance

Outcome 2 is an LS group that has a dysfunctional and imbalanced dissonance profile. This is different from Outcome 1 because within a dysfunctional dissonance cycle, dissonance – which might lead to potential teacher learning – will be present but it is not used collaboratively, safely or purposefully. This can mean that potential moments of learning are generated, but these are likely to go unnoticed or not be acted upon due to the dysfunctional nature of an Outcome 2 group. The characteristics of an Outcome 2 LS group are detailed in Table II.

Dysfunctional groups are not widely written about. Yet, when they occur they can be a particularly damaging group to the individuals in that LS group and to the sustainability of LS within in a school. Mynott (2017) wrote about his experience with a group in 2013–2014 that was dysfunctional and required intervention to safeguard the LS cycle.

In 2013–2014, an LS cycle involving three teachers Camille (inexperienced teacher), Alex (inexperienced teacher) and Teresa (experienced teacher) took place looking at independent work and group work (Mynott, 2017):

Team 2: Lesson 1 Review: October 2013

Camille: I was watching Marli. He didn’t seem to be challenged by this work.

Alex: Didn’t he?

Camille: He did all his work on a mini-whiteboard and then rubbed it all out and starting again in his book. So it might look like he has only done one of two questions but he has completed them all. I think he found them easy.

Alex: So there is nothing in his book?

[…]

Camille is focused on the performance of one child, Sayed.

Camille detailed, to Alex (less to Teresa) that the work was not sufficiently challenging for Sayed, and thus he did not do anything in his book.

Neither Teresa nor Alex can interrupt this feedback; Camille remains dominant and unyielding throughout the feedback session.

While Alex and Teresa do speak, this is Camille’s review of the lesson and of Sayed’s learning. Any interjection Teresa or Alex make is spoken over.

Alex: Are we done?

We close with a tense atmosphere between the three teachers and no sense of commitment to another lesson.

Extract 4: lesson study group review (Mynott, 2017).

In this example the dysfunction characteristics have been highlighted in Table II.

Observations of this dysfunctional group found aspects of their expertise and collaboration were underdeveloped (Mynott, 2017). As a result, the overall group’s work was dysfunctional. Pella (2011) and Horn and Little (2010) showed that participants who are new to working together generate dissonance as they share their differences, but the example from Mynott (2017) demonstrates that if protocols (Dudley, 2014) are not established then a dysfunctional outcome may occur. Dudley (2014) suggests that everyone should have the right to contribute and these contributions should be accepted and considered. He (Dudley, 2014) does not say that participants need to be agreed with and it is clear that agreement – or affirmation as in Outcome 1 – is not what is needed to avoid dysfunction. However, if you have a group of teachers working in collaboration, the agreement of protocols is likely to insufficient to support a safe space for collaborative thinking. They are going to need additional training. In Extract 4, the group discussion was dominated by one individual who was critically appraising the research lesson’s teacher rather than taking turns (Mynott, 2017). This meant that the focus of the research was instead focussed on the teacher and the observer’s thoughts and opinions. Due to the relentless nature of this feedback the teacher of the research lesson ends this review session by saying “Are we done?” indicating both her anguish and frustration (Mynott, 2017). So, what was it within the establishment of the safe space that went wrong? It seems that, in this case, there was a lack of turn taking, an inexperience of reviewing lessons and an inexperience of providing critique rather than criticism (Mynott, 2017). These elements are all underpinned by the need to establish a greater level of expertise within teachers before commencing LS, training them in core LS skills – observation and feedback.

The reliance on dissonance being generated because of differences between participants new to collaborating with each other is also a form of dysfunctional dissonance, as it means that the potential moments of learning are by chance rather than enquiry. In reality, a reliance on dissonance being generated by differences between individual participants can mean that the focus comes from what is known by the participants, some of which might not be the best or most informed practice. In 2013–2014, when the example LS was undertaken, teaching observations were still being graded and feedback often focussed on the teacher, as this is the culture the participants knew, it is what was replicated in the group’s observation and feedback. This meant that the feedback, as seen in the extract from Mynott (2017), was focused on Alex. By focussing on the individual participant, who taught the research lesson, the group did not establish a joint endeavour through which they could share and explore dissonance, moving these toward moments of potential learning. Instead their individual expertise was focussed on telling Alex where she could have improved (Mynott, 2017). This was not only dysfunctional but highly uncomfortable and risked disrupting this group’s cycle, and could have been a potentially cycle-ending situation (Mynott, 2017).

Experience development, LS protocol training, professional conflict training and facilitation by an expert (Mynott, 2017) are all ways to reduce the risk of dysfunctional LS cycle, but awareness of the complexities of participant interactions are essential to avoid LS cycles relying on the positive dissonance being generated through collaboration.

4.3. Outcome 3: limited learning dissonance

Since Mynott’s (2017) initial research, Outcome 3 has proved to be the most regular and divergent outcome, with distinct subgroups forming linked to how limited the learning dissonance was. It is likely that most LS cycles are likely to be within Outcome 3. This is, in part, because the limited learning spectrum encompasses LS cycles where the smallest amount of potential learning has been generated to LS cycles where potential learning was only slightly limited, preventing Outcome 4: rich learning dissonance. As such, the key characteristics of Outcome 3 are the same as Outcome 4 but one or more characteristic will be limited in some way to impede the generation of potential learning within the overall LS cycle.

To clarify the sub-divisions of Outcome 3 further, Figure 3, presents the three aspects of the limited learning outcome. Each of the three stages 3.1, 3.2 and 3.3 gives more clarity to the overall outcome 3 and how the characteristics listed in Table III might be limited thus effecting the over outcome of potential learning moments in an LS cycle.

Within each of the sub-sections of outcome 3, it is possible to see how LS groups might be more or less effective at generating potential teacher learning opportunities.

4.3.1 Outcome 3.1 high limitations to learning dissonance

The example of Theresa, Camille and Alex, which was a dysfunctional group from its first two LS research lessons, evolved into an Outcome 3.1 group, following facilitator intervention. Yet, despite this transition to Outcome 3 Mynott (2017) wrote about this group plateauing after Lesson 3 in their cycle because of limitations:

Team 2. Lesson 3 Review. February 2014

The group is sitting in Teresa’s classroom at lunchtime, lunch in front of them discussing the lesson that Teresa has just taught to Alex’s class.

All: Lesson Study 3? Yes 3.

Alex introduces the team to their feedback session.

Alex: So, what we planned to look at was how the children had self-involvement or self-interest in their work or if they could use story-mapping through a problem: a word problem.

Teresa: So, we planned the starter so that it linked with what the children were then going to do. We looked at vocabulary. Did that help?

Extract 5: Lesson study review (Mynott, 2017).

Team 2. Lesson 4 Review. March 2014

Alex, Teresa and Camille are sitting together round a small table in the office; they are at ease with each other and ready to discuss their lesson.

Alex begins the talk about the lesson:

Alex: We divided the class into three groups in this lesson. One independent, one guided – with Camille, and one input and left alone with Teresa.

Teresa: I didn’t leave them alone though. I kind of prompted them throughout the lesson. I didn’t help them as such just talked out loud.

Camille: That’s interesting, what do you mean?

Teresa: Well I knew they could do it from the carpet, so I was saying things like, can 44 be a multiple of 5 if we know multiples of 5 end in 0 and 5?

Extract 6: Lesson Study Review (Mynott, 2017).

The group received facilitation to stem the dysfunctional nature of their work but did not receive further facilitation, beyond this intervention, which meant that limitations in subject and pedagogic expertise prevented them from moving beyond the thinking they reached through their natural differences in experience (Mynott, 2017). In this example the limiting factor was expertise. The group had insufficient expertise in the area they were researching – group work and peer support – to go beyond the dissonance generated by their own differences. Expertise could have been developed by further reading, inviting in experts or refocusing on the available resources. The group not knowing that these were options also forms part of the limitation of expertise in this group, and for me as the facilitator at the time of this LS. Addressing the limitations of this LS cycle led to the development of the extended preparation LS model (Mynott, 2017). This extended period of preparation supports expertise development, a model seen in other recent studies (Lewis and Perry, 2017), and means that the participants are less likely to be limited in the same way as the example above.

4.3.2 Outcome 3.2: moderate limitations to learning dissonance

The case example of an Outcome 3.2 LS group was an extended preparation LS (Mynott, 2017, 2018b) which aimed at avoiding the plateauing seen in the above 3.1 example. H, D and J were all experienced teachers and this was the first time the three had worked together in an LS, but all three had been involved in LS cycles previously. J acted as an internal facilitator within this study and would research the readings for the group. J also involved other teachers from the school in the study at different points to gain their expertise. This LS, in 2016–2017, focussed on French and Geography in Year 4. < Je vais à l’école > was a topic about exploring the locality around the school and acquiring French vocabulary for locations in town. This extended preparation LS began preparations in the autumn term 2016 with a range of participants (diverse in both their experience and expertise linked to the subjects involved), and worked towards an implementation in the summer term 2017. The participants engaged with reading and developed their mapping and geographical skills, while talking about how these might help support and develop the recall of vocabulary in French:

H – But have we, maybe we don’t simplify it next time but we spend longer, get them to focus on the vocabulary for longer. We could give them the chance to do that.

D – You would think that. They should know that they are a time connective as there is a comma after it. Like in my head that is what makes sense. I can almost just include it as it is telling me something like after or later. But I think there is just so much to look at.

H – So you think we should simplify. Take away the time connectives.

D – I would say so.

H – Because it is just one extra thing to think about.

J – Yeah because I suppose if they couldn’t get <gare> from that instruction. Then something like this is the train station is needed.

15:00

J – This is the train station.

D – I think choose the word that is train station.

J – <une gare> but if they didn’t get <gare> from that then maybe we need to think about simplifying the sentences.

J – Um

D – See that this one. Number 4. Um is quite. Um. Like I turned right and I see the pyramid. It’s not a real pyramid it is a shop. Technically my children should be able to translate all of that. Because like turn is pretty much the same.

J – I felt like that was the most successful one though.

D – Yeah

J – They could really see the pyramid. I don’t think they got turn right.

Extract 7: Extended Preparation Lesson Study 1: Lesson 1 2016–2017.

Ultimately, while this group had facilitation and support from external expertise the group’s work plateaued during the cycle because the LS evolved to require additional subject and pedagogical expertise in the form of retrieval and cognitive science. The team, including the facilitator, did not have this additional expertise nor sufficient time within the LS window to obtain and comprehend further research, meaning that time and expertise limited this LS cycle.

Time was a significant limiting factor because this study was tracking a French unit of work taught in two Year 4 classrooms. This structure meant the LS happened in two sessions each week and while the team were able to review the lessons and feed that into their re-teaching to the parallel group they were not then able to also read and integrate the information about memory, retrieval and cognitive science into their work and thus this then impacted on their ability to follow the thread of their learning, which indicated strongly that vocabulary retrieval would be an important aspect of the students’ vocabulary flexibility. As the group had evidenced that they could obtain and explore evidence and expertise in their preparation work their use of expertise was limited by the time to gain further knowledge, and while their lack of expertise limited their discussions, with more time they would have been able to have acquired the relevant learning to further their discussions. Overall, this LS cycle is not as limited as the 3.1 example because despite the limitations of time and expertise, the LS group generated frequent potential teacher learning moments, as seen in the grey highlights in the extract, and it was their capitalising on some of these that was more challenging due to the limitations of time to further their expertise.

4.3.3 Outcome 3.3: low limitations to learning dissonance

An LS conducted in 2017–2018 exemplifies outcome 3.3. Mynott et al. (2018) researched consonant clusters using an extended preparation LS model (Mynott, 2017). M, A and J were all experienced teachers and group was formed around a mutual interest to support improvements to phonics instruction in Year 1. The group undertook three research lessons after a term of extended preparation (Mynott et al., 2018). In this LS, J took the role of internal facilitator but M and A also contributed to the group’s research, gathering of literature and further enquiry research.

The transcripts of this study indicate that the group were able to draw expertise effectively from beyond the group and use this to generate opportunities for potential teacher learning throughout the LS cycle. Within this LS cycle there were fluctuations in the effectiveness of the facilitation to promote potential learning and ultimately there was a need to further develop expertise about morphology and how this related to the phonology of consonant clusters:

J: And at the moment we are still focusing on the initial clusters, um. With that there are still some who are not blending the cluster first.

M: But compared again to last week, I think that 90% of them are.

J: Yes, was it P who didn’t do that.

M: I was one who did not do that and then even later on when I had the second phonics intervention some of them would not do it. I would be like “oh but I see something”.

04:00

M: Then they know that I did.

A: (jokes) they know there is one then.

M: He was very quiet today. There are a few of them and I do not know why that happens. I mean we want them to blend it because we are focused on that.

A: Looking at the worksheets, lots of them I learn, I feel learn the pattern. So, for these words they will do box, dot, dot. Box, dot, dot. They know it is the same. It’s the same with the splits, box, split and dot in the middle of the word, but they are not reading them. They all know that the first two letters will be a cluster. So, they will always put a box. So, when we are saying that they are doing the boxes correctly, are they? Or have they just learnt […]

M: That it is there always.

J: Can we throw some red herrings in?

M: Yeah and also next week’s first slide needs to have words that don’t have clusters.

Extract 8: Phonics Lesson Study Cycle 2018.

This initial extract shows the teachers discussing their LS cycle. It is in the fourth minute of this review that they start to untangle whether pupils are learning how to recognise and read consonant clusters or if they are pattern spotting without understanding the phonetic codes:

J: I know but I just suggesting that with what we are doing here that it is useful because then the split is then in between the two ps. So, it is when we have two consonant sounds next to each other, that’s when it splits. There is a reason, the reason it doesn’t split before nd is because there aren’t two consonant sounds there, there is only one.

09:00

M: So, a last cluster is considered as one consonant sound?

J: So, there are two here, and only one here. This is a diagraph so we split before. So, there is only one here so technically we would be we should split before, but we have added ing as a suffix. So that changes the flow of the word and the ing is going to be the consonant in itself. Whereas up here we would have to one here after this g because the -on is a diagraph but it is a French pronunciation diagraph.

Extract 9: Phonics Lesson Study Cycle 2018.

The group is drawing on the expertise of its facilitator (J) who helps them to develop and further their understanding of both the LS process but also in this example how the word chosen for this particular element of the LS did not work as well as a different choice due to its complex etymology. Mynott et al. (2018) and Mynott (2018a) provide further detail on how the learning of this group was shaped and developed throughout their extended preparation LS cycle. There are many potential learning moments from task development, consideration in detail of pupil misconceptions and the need to continue to read and research throughout the full LS cycle (Mynott et al., 2018).

So why is this LS outcome 3.3 and not outcome 4? Ultimately there are variations in the LS cycle that means while there were multiple opportunities for potential teacher learning throughout the LS these moments fluctuate, due to the facilitator’s ability to continue to develop the subject expertise (Mynott, 2018b), and thus the LS outcome is not Outcome 4 – rich learning.

4.4 Outcome 4: rich learning dissonance

There are no current case studies which demonstrate an Outcome 4 LS. The difference between an Outcome 3.3 and Outcome 4 would be that the element of limitation is removed, so the potential learning opportunities are not inhibited in any way. In 3.2 and 3.3’s examples limitations were found in time, expertise and facilitation; 3.2 was more limited because the LS cycle took the participants into new expertise fields that they did not have as much understanding of. In 3.3 the same happened but the expertise within the group and the facilitator helped continue to pursue the enquiry meaning that the group were able to use their time to focus on their enquiry building in their expertise. An Outcome 4 would have the same characteristics – see Table IV – as an Outcome 3 but none of these characteristics would be limiting potential learning for participants.

An unlimited version of LS – an Outcome 4 – seems theoretically possible, but with the complexity of schools and the factors affecting LS, as described in the examples in this paper, it seems unlikely an LS will be uninhibited by some element of collaboration, time or expertise.

5. Discussion and conclusion

The contributions of this paper are three-fold: one is that there are different outcomes of LS groups and facilitators need to be aware of where our groups sit on the outcome model during and at the completion of our LS work so we can facilitate and improve outcomes further. Two, more research attention needs to be given to groups that do not reach Outcome 3 and above, so we can learn more about groups which, in the past, have been perceived to have failed. These groups do exist and it is important that their experiences are learnt from, as this will help refine and develop LS. The third contribution is Outcome 4 is a theoretical model of the perfect LS cycle, one with no limitations. While, it is not impossible for an LS group to be unencumbered by any of the limiting factors discussed in this paper, it seems unlikely, even with a team of expert facilitators undertaking LS, that no limiting factor of some kind will occur.

As a result of the above contributions, Mynott’s (2017) model of potential LS outcomes has been revised to include the sub-divisions of Outcome 3 – Figure 4.

Ultimately, potential outcomes are a starting point for further discussion. Mynott (2018b) described how we need to think about LS in longer-term evaluation. Increasingly, an impact distribution map, for an LS cycle, may aid our review of how far an LS has reached over time. For, if in a cycle, teachers find that sequencing a Maths question in a certain way aids their instruction, and this was reviewed a year later and that review found that the teachers in the LS team were still using this method, and then an year later seven more teachers were using this method it might be possible to conclude that actual learning had occurred. Yet, it will still not suggest that an LS cycle transforms the pupil outcomes in a school (Mynott, 2018a, b) and if any belief that data of this sort will be generated, in short term studies, will cause ineffective causation to occur like that seen in Education Endowment Fund (2017) report which found little impact of LS on pupil outcomes. This is due to the immense variables that will also impact on pupil data, outside of LS cycles (Mynott, 2018a, b). Yet, equally, it should not be forgotten that pupil learning will be important to any evaluation of LS (Godfrey et al., 2018). Instead, it is all about keeping each area of evaluation in perspective.

This paper set out the need to explore the potential outcomes of LS in more detail and has shown that there are different outcomes which an LS may reach. It is important to note that this is only a part of the work that needs to be completed to improve our understanding of LS. As while the final outcome model, Figure 4, shows what is possible within an LS cycle, in terms of outcomes, far too little attention has been given to what happens after an LS. Do potential moments of teacher learning turn into actual teacher learning? It is a vital question because unless LS is more analytical of its outcomes it runs the risk of being unable to substantiate, beyond participant perception, that it is transformative for teacher learning.

Figures

Model of potential teacher learning outcomes in lesson study

Figure 1

Model of potential teacher learning outcomes in lesson study

Potential participant learning

Figure 2

Potential participant learning

Detailed exploration and sub-division of lesson study Outcome 3

Figure 3

Detailed exploration and sub-division of lesson study Outcome 3

A revised model of potential lesson study outcomes

Figure 4

A revised model of potential lesson study outcomes

Outcome 1: absence of dissonance

Dissonance is not present/or very little dissonance is present. If dissonance is generated this is reduced by the Lesson Study group’s interactions
1.1 Time 1.2 Collaboration 1.3 Expertise
The group find it difficult to find time/organise lessons and reviews Affirmation and ego-protection impact on discussions preventing dissonance Expertise not drawn on to generate dissonance/or expertise is not sufficient to generate dissonance
Avoidance is a theme (team or members are unavailable etc.) Individuals may take feedback personally Group does not promote professional conflict, or seeks to reduce it
Lesson Study cycle might end in an unplanned way Meetings are hard to schedule/cease to occur before end of Lesson Study cycle Observation skills are underdeveloped
Time is reduced so that meaningful Lesson Study cycles/reviews are unable to reach points of dissonance Enquiry focus of Lesson Study cycle is poorly defined/non-existent Feedback skills are underdeveloped
Group find it difficult to see value in Lesson Study work
May need external facilitation to support collaboration

Outcome 2: dysfunctional dissonance

Dissonance is present but its impact causes dysfunction and/or disruption to the Lesson Study cycle
2.1 Time 2.2 Collaboration 2.3 Expertise
Planning does not always include all participants Insecure or dominant egos affect collaboration Conflict leads to dissonance but this is not professional or helpful
Participants arrive late/leave early May focus on an individual and their perceived failings/successes Expertise is drawn on but dysfunctional approach means it is not used effectively
Lesson Study cycle might end in an unplanned way Safe space for professional discussion is not established Observation skills are underdeveloped
Time is used unproductively by focusing on specific or individuals not related to Lesson Study’s enquiry focus Lesson Study protocols (Dudley, 2014) are not followed Feedback skills are underdeveloped
There is a lack of joint endeavour
May need external facilitation to support collaboration

Outcome 3: limited learning dissonance

Potential moments of participant learning are generated but one or more limitations occurs which impacts on participant learning
3.1 Time 3.2 Collaboration 3.3 Expertise
Time is given to plan, undertake and review Lesson Study work
Lesson Study teams take time to develop expertise and collaboration
Lesson Study teams take time to develop line of enquiry, research question
Joint endeavour exists and focus of work is on “we” not “I”
Focus of work is around enquiry, research question
Team generate and manage dissonance to support generation of moments of potential learning
Expertise is drawn on by groups (experts, research and/or resources may be used)
Expertise in professional conflict supports the generation and support of dissonance to potential learning moments
Internal or external facilitation is used to support Lesson Study cycle

Outcome 4: rich learning dissonance

Potential moments of participant learning are generated as in Outcome 3 and no limitations impact on potential moments of participant learning.
3.1 Time 3.2 Collaboration 3.3 Expertise
Time is given to plan, undertake and review Lesson Study work
Lesson Study teams take time to develop expertise and collaboration
Lesson Study teams take time to develop line of enquiry, research question
Joint endeavour exists and focus of work is on “we” not “I”
Focus of work is around enquiry, research question
Team generate and manage dissonance to support generation of moments of potential learning
Expertise is drawn on by groups (experts, research and/or resources may be used)
Expertise in professional conflict supports the generation and support of dissonance to potential learning moments
Internal or external facilitation is used to support Lesson Study cycle

Detailed lesson study outcome framework

Outcome 1: absence of dissonance
Dissonance is not present/or very little dissonance is present. If dissonance is generated this is reduced by the Lesson Study group’s interactions
1.1 Time 1.2 Collaboration 1.3 Expertise
The group find it difficult to find time/organise lessons and reviews
Avoidance is a theme (team or members are unavailable etc.)
Lesson Study cycle might end in an unplanned way
Time is reduced so that meaningful Lesson Study cycles/reviews are unable to reach points of dissonance
Affirmation and ego-protection impact on discussions preventing dissonance
Individuals may take feedback personally
Meetings are hard to schedule/cease to occur before end of Lesson Study cycle
Enquiry focus of Lesson Study cycle is poorly defined/non-existent
Group find it difficult to see value in Lesson Study work
May need external facilitation to support collaboration
Expertise not drawn on to generate dissonance/or expertise is not sufficient to generate dissonance
Group does not promote professional conflict, or seeks to reduce it
Observation skills are underdeveloped
Feedback skills are underdeveloped
Outcome 2: dysfunctional dissonance
Dissonance is present but its impact causes dysfunction and/or disruption to the Lesson Study cycle
2.1 Time 2.2 Collaboration 2.3 Expertise
Planning does not always include all participants
Participants arrive late/leave early
Lesson Study cycle might end in an unplanned way
Time is used unproductively by focusing on specific or individuals not related to Lesson Study’s enquiry focus
Insecure or dominant egos affect collaboration
May focus on an individual and their perceived failings/successes
Safe space for professional discussion is not established
Lesson Study protocols (Dudley, 2014) are not followed
There is a lack of joint endeavour
May need external facilitation to support collaboration
Conflict leads to dissonance but this is not professional or helpful
Expertise is drawn on but dysfunctional approach means it is not used effectively
Observation skills are underdeveloped
Feedback skills are underdeveloped
Outcome 3: limited learning dissonance
Potential moments of participant learning are generated but one or more limitations occurs which impacts on participant learning
3.1 Time 3.2 Collaboration 3.3 Expertise
Time is given to plan, undertake and review Lesson Study work
Lesson Study teams take time to develop expertise and collaboration
Lesson Study teams take time to develop line of enquiry, research question
Joint endeavour exists and focus of work is on “we” not “I”
Focus of work is around enquiry, research question
Team generate and manage dissonance to support generation of moments of potential learning
Expertise is drawn on by groups (experts, research and/or resources may be used)
Expertise in professional conflict supports the generation and support of dissonance to potential learning moments
Internal or external facilitation is used to support Lesson Study cycle
Outcome 3.1: high limitations to learning dissonance
Learning is largely due to differences between team members Learning does not draw on expertise from beyond the LS team Time/Expertise and or other factors prevent learning from being sustained Facilitation does not occur, or does not impact on teacher learning opportunities
Outcome 3.2: moderate limitations to learning dissonance
Learning draws on knowledge or expertise from beyond the group participants but this is not consistent or has a small impact Learning plateaus or ceases during the LS cycle Time/Expertise and or other factors reduce learning opportunities Facilitation occurs and produces opportunities for some potential teacher learning
Outcome 3.3: low limitations to learning dissonance
Learning draws effectively on knowledge of expertise from beyond the group participants and this extends teacher learning opportunities Learning fluctuates within LS cycle due to a limitation (time/availability) Facilitation occurs to support teacher learning and there are many opportunities of potential teacher learning
Outcome 4: rich learning dissonance
Potential moments of participant learning are generated as in Outcome 3 and no limitations impact on potential moments of participant learning

Appendix

Table AI

References

Achinstein, B. (2002), Community, Diversity, and Conflict among Schoolteachers: The ties that Blind, Teachers College Press, New York, NY.

Aronson, E. (1999), “Dissonance, hypocrisy, and the self-concept”, in Harmon-Jones, E. and Mills, J. (Eds), Cognitive Dissonance: Progress on a Pivotal Theory in Social Psychology, American Psychological Association, Washington, DC, pp. 103-126.

Chatzidakis, A., Hibbert, S. and Smith, A.P. (2007), “Why people don’t take their concerns about fair trade to the supermarket: the role of neutralisation”, Journal of Business Ethics, Vol. 74 No. 1, pp. 89-100.

Dudley, P. (2013), “Teacher learning in lesson study: what interaction-level discourse analysis revealed about how teachers utilised imagination, tacit knowledge of teaching and fresh evidence of pupils learning, to develop practice knowledge and so enhances their pupils’ learning”, Teaching and Teacher Education, Vol. 34, pp. 107-121.

Dudley, P. (2014), Lesson Study: A Handbook, Lesson Study UK (LSUK), Cambridge, available at: http://lessonstudy.co.uk/lesson-study-a-handbook/ (accessed 25 August 2018).

Education Endowment Fund (2017), “Lesson study: evaluation report and executive summary”, November, available at: https://educationendowmentfoundation.org.uk/projects-and-evaluation/projects/lesson-study (accessed 25 August 2018).

Fernandez, C. and Yoshida, M. (2004), Lesson Study: A Japanese Approach to Improving Mathematics Teaching and Learning, Routledge, Oxon.

Festinger, L. (1957), A Theory of Cognitive Dissonance, Stanford University Press, Stanford, CA.

Godfrey, D., Seleznyov, S., Anders, J., Wollaston, N. and Barrera-Pedemonte, F. (2018), “A developmental evaluation approach to lesson study: exploring the impact of lesson study in London schools”, Professional Development in Education, available at: www.tandfonline.com/doi/full/10.1080/19415257.2018.1474488 (accessed 16 January 2019).

Guest, G., MacQueen, K. and Namey, E. (2012), Applied Thematic Analysis, Sage, Los Angeles, CA.

Guskey, T. (2000), Evaluating Professional Development, Corwin Press, Thousand Oaks, CA.

Horn, I.S. and Little, J.W. (2010), “Attending to problems of practice: routines and resources for professional learning in teachers’ workplace interactions”, American Educational Research Journal, Vol. 47 No. 1, pp. 181-217.

Lewis, C. and Perry, R. (2017), “Lesson study to scale up research-based knowledge: a randomized, controlled trial of fractions learning”, Journal for Research in Mathematics Education 2017, Vol. 48 No. 3, pp. 261-299.

Mynott, J. (2017), “A primary head teacher’s exploration of lesson study”, available at: https://uhra.herts.ac.uk/bitstream/handle/2299/18330/14107916%20Mynott%20John%20Final%20Submission.pdf?sequence=1 (accessed 25 August 2018).

Mynott, J. (2018a), “Facilitating the lesson study facilitator: a reflection on expertise in lesson study”, paper presented to The IPDA Annual Conference, Border Crossings: Professional Development in the 21st Century, Aston University, Birmingham, 16-17 November.

Mynott, J. (2018b), “It is easy to be misled…a case for longer term lesson study considerations”, available at: http://lessonstudy.co.uk/2018/04/it-is-easy-to-be-misled-a-case-for-longer-term-lesson-study-considerations/ (accessed 25 August 2018).

Mynott, J., Paalanen, M. and Jaffer, A. (2018), “Does teaching consonant cluster systematically aid decoding IMPACT”, Journal of the Chartered College of Teaching, Vol. 3, pp. 22-25, available at: https://impact.chartered.college/article/mynott-does-teaching-consonant-clusters-systematically-aid-decoding/

Pella, S. (2011), “A situative perspective on developing writing pedagogy in a teacher professional learning community”, Teacher Education Quarterly, Vol. 38 No. 1, pp. 107-125.

Perry, E. and Boylan, M. (2018), “Developing the developers: supporting and research the learning of professional development facilitators”, Professional Development in Education, Vol. 44 No. 2, pp. 254-271.

Seleznyov, S. (2018), “Lesson study: an exploration of its translation beyond Japan”, International Journal for Lesson and Learning Studies, Vol. 7 No. 3, pp. 217-229, available at: https://doi.org/10.1108/IJLLS-04-2018-0020

Stepanek, J., Appel, G., Leong, M., Mangan, M. and Mitchell, M. (2007), Leading Lesson Study: A Practical Guide For Teachers and Facilitators, Corwin Oaks Press, Thousand Oaks, CA.

Xu, H. and Pedder, D. (2015), “Lesson study: an international review of the research”, in Dudley, P. (Ed.), Lesson Study: Professional Learning for Our Time, Routledge, Oxon, pp. 29-58.

Corresponding author

John Paul Mynott can be contacted at: johnmynott@hotmail.com

About the author

John Paul Mynott is Headteacher at Central Primary School, Watford. He has been researching Lesson Study and teacher development since 2012. He completed his Doctorate in Education (EdD) in 2017 at the University of Hertfordshire. He is a full time Headteacher and Researcher.