Enhancing the enactment of assessment for learning principles during school placement: preservice teachers as practitioner researchers within a learning community

ABSTRACT Enacting Assessment for Learning (AfL) is challenging for teachers and even more for Preservice Teachers (PSTs). The support for those attempting to enact it has been considered insufficient and/or inadequate. This study aims to examine how positioning preservice teachers as practitioner researchers while belonging to a learning community can facilitate PSTs’ enactment of AfL during their school placement. It draws on the triangulation of different qualitative data sources, e.g. focus group, researcher’s field notes, testimonial survey, and school placement report, inquiring the practice of eight PSTs and the interaction within a learning community supported by a researcher, university supervisor and cooperating teachers. Despite belonging to the same learning community, PSTs struggle differently in embedding assessment in the learning experiences, including students, and facilitating students’ understanding of feedback and the teaching-learning process. Teacher education programmes are prompted to consider how best to support PSTs enacting AfL practices.


Introduction
Teachers tend to struggle and resist changing their assessment practices (DeLuca et al. 2018), even more so when considering Assessment for Learning (AfL) given that AfL introduces multiple assessment practices (Moura et al. 2021).These multiple practices include using evidence from assessment to improve students' decisions and address their needs, having them regulate and actively construct their own learning (Baird et al. 2017;Coombs et al. 2018).While there is an acknowledgement that embedding AfL in the teaching-learning process is challenging for (preservice) teachers (Birenbaum et al. 2015), little has been done to consider practical ways to support Preservice Teachers (PSTs) to overcome such a challenge (Brevik, Blikstad-Balas, and Engelien 2017;Kleij, Joy Cumming, and Looney 2018).Supporting teachers' use of AfL is considered a key aspect to the success of AfL implementation (Qin and Yi 2021;Schildkamp et al. 2020) given that PSTs are, unsurprisingly, influenced by practicing teachers' conceptions and intentions (Shannon, Smith, and Dana 2016;Yan et al. 2021).More research on PSTs' conceptions of assessment is necessary (Wang, Kao, and Lin 2010) considering it is a key aspect when planning and implementing Assessment (for Learning) (Lutovac and Assunção Flores 2022).On the study of Moura et al. (2023), anticipatory socialisation resulted in PSTs struggling to change their assessment conceptions and practices due to low assessment literacy levels and tendency to replicate what they experienced as students.
The opportunity to belong to a learning community (LC) that supports members to persevere when facing problems, that encourages and enriches exposure to different perspectives and experiences, may positively influence learning.Creating a willingness for (preservice) teachers to reflect on, and be constructively critical about, their AfL practices is essential to increasing the likelihood of being more successful using AfL (Heitink et al. 2016).Conscious that encouraging (preservice) teachers to reflect solely on their knowledge can lead them to a 'closed circle of information', it is desirable to encourage (preservice) teachers to consider their knowledge and practices in the context of the wider literature, and researching their practices in an attempt to improve and transform them (Heissenberger and Matischek-Jauk 2020).
This study aims to explore the extent to which PSTs researching their own practices while involved in a LC (that includes PSTs, a researcher, a university supervisor and cooperating teachers) enhances PSTs' enactment of AfL during school placement.This study addresses the following research questions: 1) To what extent does being a member of a LC facilitate PSTs' thinking, planning and practices of AfL?; and 2) In what ways does the positioning of PSTs as practitioner researchers facilitate PSTs' enactment of AfL.
The authors believe this study is important as it positions preservice teachers as practitioner researchers in a LC with all those actively involved in the education of future teachers.It is anticipated that the support provided in the LC as well as PSTs undertaking the role of practitioner researchers will contribute to an understanding of the extent to which both (practitioner researcher and LC) contribute individually and/or collectively to enable PSTs' learning.This study also fills gaps pointed in a recent literature review (Rutten 2021) regarding teacher education and practitioner research like research of practice as a continuous process, and study designs that link practical research with empirical changes on PSTs' thinking and practices.

The challenges of enacting assessment for learning
The use of AfL (i.e.assessment embedded in the teaching-learning process) requires changes to the more traditional teaching-learning process, i.e. using assessment evidence to support students' learning, using student-centred approaches with the students themselves having an active role in the learning process (Baird et al. 2017;Coombs et al. 2018).AfL is any assessment carried out primarily with the intention of promoting students' learning (Wiliam 2018).Socio-constructivist learning theories (Vygotsky 1978) support students acting as co-constructers of their own learning and inform the main intention of AfL.Five key principles underpin AfL: (1) clarifying and sharing learning intentions and criteria for success; (2) engineering effective classroom discussions, questions, and learning tasks that elicit evidence of learning; (3) providing feedback that moves learners forward; (4) activating students as instructional resources for one another; and (5) activating students as the owners of their own learning (Wiliam and Thompson 2007, 64).
The use of AfL to improve student's learning has occasioned new challenges to teacher's practices (Allal 2020).The response of teachers to such challenges is important, as there is evidence that each different enaction of AfL produces dissimilar effects on students' learning (Schildkamp et al. 2020;Wiliam 2018).While each AfL principle aligns with several assessment techniques, the underlying principles are summed up as 'tight but loose' (Wiliam and Leahy 2015), which allow teachers to enact AfL in what they consider the most appropriate way for their specific context.(Preservice) teachers using AfL must therefore be critical and reflective about which strategies they use, the impact they have on their students' learning, and how to better use assessment evidence to improve said learning (Heitink et al. 2016).
Considering the challenges and changes in the teaching-learning process demanded by the enactment of AfL, there are suggestions that the AfL support offered to (preservice) teachers has been insufficient and/or inadequate (Kleij, Joy Cumming, and Looney 2018;Moura et al. 2021).Often, this support has been characterised as too theoretical, with a limited direct relationship to teachers' classroom challenges and a lack of follow-up /continuous support with teachers on their AfL practices (Brevik, Blikstad-Balas, and Engelien 2017;Kleij, Joy Cumming, and Looney 2018).

Practitioner research
Schools and universities have differing visions and scope with respect to research activity.Research is not necessarily viewed in schools as a support to the improvement of teachers' practices (Henning, Petker, and Petersen 2015).As such, schools generally concern themselves solely with the practical application while the theoretical components are left to university teacher education programmes (Admiraal et al. 2017).
As university supervisors facilitate PST's access to literature, the onus should be on the PSTs to subsequently transform their practices through undertaking independent research (Heissenberger and Matischek-Jauk 2020;Hilton and Hilton 2017).Practitioner research can be considered as one means of research that can support such a professional development (Rutten 2021).PSTs' investigation of their practices through action research practices has noted benefits to their professional development (Ulvik and Riese 2016), with a warning that in order to develop an inquiring perspective PSTs require time and space.
Practitioner research is commonly considered a 'transformative approach' that empowers PSTs lacking researching skills, criticism and ability to reflect (Henning, Petker, and Petersen 2015).In practitioner research, (preservice) teachers are responsible for constructing knowledge as a consequence of the systematic analysis of their own practices and continually questioning their own thinking (Heissenberger and Matischek-Jauk 2020).
While studies explicitly using the keywords 'PSTs' and 'practitioners' researchers' are uncommon in the literature, some empirical studies do highlight the benefits on PSTs' current (as PSTs) and future (as teachers) learning when involved in practitioner research.Some PSTs have acknowledged the importance of practitioner research approaches on their work as teachers while other PSTs have considered that action-research helped them to improve their skill set and mindset in teaching and learning (Davis, Clayton, and Broome 2018;Smith and Sela 2005).There is evidence that having PSTs researching their own practices leads them to identify and solve problems, improve their teaching, and cast a critical and reflective view on the teaching-learning process (Kennedy-Clark et al. 2018).Practitioner research emphasises a practice based on reflection that enables to produce knowledge specific to the context in which is applied and the transformation of the own practice (Heissenberger-Lehofer and Krammer 2023).This process entails researching one's own practice (e.g.analysis of classroom evidence), solving problems and enhancing the learning (e.g.adjustment of teaching strategies to attend students' needs) (Yuan, Yang, and Stapleton 2020).
There is a need for more empirical studies demonstrating the impact of practitioner research on pre-service teachers' thinking and teaching.Most current studies focus only on describing the characteristics of PSTs acting as practitioner researchers (Rutten 2021).To help PSTs become researchers of their own practices, it is crucial to provide them with support and collaboration.Learning communities that bring universities and schools closer can facilitate this process (Ribaeus, Enochsson, and Löfdahl Hultman 2020).
Learning community Smith and Sela (2005) highlighted the importance of teachers' support to help PSTs surpass difficulties they might experience when undertaking school placement.Such difficulties can include PSTs being exposed to specific approaches to teaching in the school with which they were previously unfamiliar and/or receiving no support from school teachers when enacting teaching practices uncommon in schools (Qin and Yi 2021;Ribaeus, Enochsson, and Löfdahl Hultman 2020).Listening, participating and interacting with other PSTs has been found to help PSTs' learning, as well as fostering the growth of the LC (Cardoso, Batista, and Graça 2014).Further evidence suggests that PSTs considered research collaboration and mentor support important to start valuing (action) research and developing their skills as researchers (Davis, Clayton, and Broome 2018).
In the aforementioned studies, learning communities were a consistently important contribution to PSTs' development, encouraging them to be more critical and reflective about their learning.The interaction between learning communities and PSTs proved to be mutually beneficial, with the former progressing as a result of the latter's learning.For this reason, learning communities may be beneficial in a school placement context to support PSTs' practices while promoting mentors'/teachers' professional development (Skerrett and Williamson 2015).Subsequently, it is envisaged that learning communities are not effective by themselves, i.e. to be effective, teachers and PSTs need to learn how best to work with each other (Skerrett and Williamson 2015).
In a LC, learning happens in interaction between mentors-mentees-mentees (Cardoso, Batista, and Graça 2014) which implies learning from different members of the community.The degree of PSTs' success in enacting any pedagogical approach is related to the alignment between mentors' and PSTs' conceptions, and with PSTs' practices (Qin and Yi 2021).

Methodology
This study used a practitioner research approach (Heissenberger and Matischek-Jauk 2020;Shannon, Smith, and Dana 2016) which affords participants the opportunity to research and reflect on their practices.This approach aims to narrow the theory-practice gap, empowering PSTs to research and transform their own practices, as well as address their teaching needs.

Participants and context
Eight PSTs from a two-year Physical Education Master's degree programme from a Portuguese Public University took part in the study.Upon successful completion of this Master's degree, PSTs are qualified to teach physical education in primary and secondary schools.The first year of the programme takes place at the university and focuses on three main areas: Sports Sciences, Education Sciences, and Didactics.Assessment is a theoretical topic integrated in the 'Sports General Didactics' module and focuses mainly on assessment content and assessing students' performance.PSTs have some practical experiences of assessment in the 'Sports Specifics Didactics' courses, with a focus on teaching how to assess technical-tactical content of different sports.During the second year of the programme, PSTs spend the majority of their time on school placement (from Tuesday to Friday) with their cooperating teacher.On Mondays, PSTs attend the university to meet with their supervisor, discuss their planning, challenges and teaching (strategies), as well as how best to develop as teachers.
During school placement, PSTs are placed in groups of three or four in primary and post-primary schools.Each Preservice Teacher (PST) is responsible for teaching one of the cooperating teacher's classes during the entire year and at least, one teaching unit in a class at a different educational level.The cooperating teacher will be surveying the PSTs performance throughout the year.The university supervisor observes three lessons throughout the whole year and carries the most weight in supporting PSTs' school placement report.On completing the school placement, PSTs provide a written school placement report that captures their school teaching experience.PSTs are then asked to present and defend this document publicly in the university.This presentation determines the success, or lack thereof, of their school placement.
The eight PSTs were purposively selected (Patton 2002) according to (i) their commitment, predisposition and interest in joining the pedagogical project, (ii) being supervised by the same university supervisor, and (iii) undertaking school placement in schools where cooperating teachers were interested in being involved in the project and had already been collaborating with the university for more than 10 years.These PSTs (five women and three men) had no professional experience as teachers.Six of the PSTs had completed their undergraduate programme at the same university as the Master's programme.
All participants (i.e.PSTs, cooperating teachers, PSTs' students, and the students' parents) signed an informed consent form to participate in the study.The study purpose and design were explained to all participants, and they were clearly told they could choose to leave the study at any time without any consequences.University and school authorities in which the research was conducted granted ethical approval to the study.A numeric code (e.g.PST1, PST2) was assigned to each PST to preserve anonymity.

Study design
The intervention study consisted of three cycles over a period of four months (Table 1) with PSTs attending university seminars on Mondays and undergoing school placement from Tuesday to Friday.The pre-intervention intended to understand PSTs' conceptions of assessment.For this to happen, seminars had a positive environment where PSTs felt comfortable and safe to share, question, and discuss their assessment conceptions, experiences, and act as co-constructers of their learning.The first cycle was intended to understand how PSTs share learning objectives with their students, identify aligned learning tasks and share success criteria.The second cycle focused on PSTs providing feedback to their students.The third cycle focused on students' use of, and involvement in, assessment tasks.Each PST was formally observed three times by the researcher.
Seminars allowed the LC, researcher, university supervisor, four cooperating teachers and eight PSTs, to meet every two weeks and share planning, practices, experiences, and reflections PSTs had throughout the school placement.Seminars were outlined around AfL principles (Wiliam and Thompson 2007) to support PSTs' practices.The researcher and PSTs' university supervisor (acting as facilitators) created a 'space' where PSTs felt safe to learn, interact and share, acting as active constructers of their own learning and researchers of their own practices (practitioner researchers).The LC developed itself throughout the process with the increasing feeling of belonging to it, feeling comfortable to share their ideas and experiences, and PSTs understanding the community was created for supporting them.

Focus group interviews
Two focus groups took place at the end of cycle two and three.The eight PSTs were involved in both focus groups, with the researcher and PSTs' university supervisor acting as facilitators of the discussion.Cooperating teachers did not attended the focus group given that the goal of the focus groups was to capture PSTs' changes to, and construction of, knowledge.Researcher and university supervisor intervention was minimal, being there just to facilitate the discussion.Each focus group lasted, on average, 70 minutes with both exploring (i) how interaction in the LC supported PSTs to overcome problems and constraints while planning and enacting AfL, and (ii) the meaning and value attributed to the LC by PSTs.Examples of questions posed include, 'What concerns do you have with respect to helping students progress in their learning?','How do you design assessment tasks?' and 'How has the LC supported your practices?'.Interviews were audio recorded and transcribed verbatim before being returned to PSTs to read and approve the accuracy of their respective transcripts.

Researcher observation and field notes
The researcher collected field notes each time he observed PSTs' classes to capture PSTs' enactment of AfL.After each observation, the researcher and PST considered the strengths and aspects to improve with respect to the use of AfL and students' learning.
PSTs received the observation sheet and lesson plan form, both with comments, on the day following the observation.The observation sheet had been created by the researcher and PSTs' university supervisor.All involved in the study agreed that the observation sheet was open to changes at any stage, if appropriate.The observation sheet filled by the researcher comprised three main areas -planning, organisation/ management, and students' involvement in the learning process.These areas were assessed using a Likert scale (1 (unreached) to 4 (fully achieved) and a 'not applicable' option.There was also space to write qualitative comments on the use of AfL principles (Wiliam and Thompson 2007), with three comment boxes for prompting suggestions on how to improve the use of AfL, the degree of students' achievement of the learning goals, and the role of students and PST in the lesson.

Testimonial survey
Upon completion of the study, a survey with eight open-ended questions was emailed to each PST.The goal of the eight questions was to prompt reflection, so none of them required a direct answer.Survey questions focused on (i) how PSTs planned and enacted AfL, (ii) how they provided feedback to students to move their learning forward, and (iii) how they included students in the process.PSTs were also asked for feedback on how their involvement in a LC contributed to their learning about AfL and for suggestions on improving the effectiveness of the LC.All PSTs completed the survey, with six choosing to answer the specific questions that were posed and two choosing to reflect freely.
Regardless of what option PSTs chose, survey responses were approximately five pages long.

School placement report
PSTs complete a school placement report at the end of their second year of the Master's to capture their teaching experience during school placement.A specific section in the report prompts PSTs to describe the experience of planning and enacting the teachinglearning and assessment process.It was anticipated that the report would capture additional information about PSTs' conceptions of assessment that may have been overlooked during the school placement.Each substantial report included contextualisation of school context, PSTs' students, examples of annual planning, teaching units, lesson plans, reflections, a log diary, and involvement in the practitioner research study carried out throughout school placement.

Data analysis
A deductive-inductive analysis procedure was used across all data, moving back and forth between the aim of the study, the data, and AfL principles.Data analysis used a three component flow process: data condensation, data display, and conclusion drawing/ verification (Miles, Huberman, and Saldaña 2014), with the researcher oscillating between the three components.Data analysis began after the first observation conducted by the researcher, which informed the design of the following seminars.Data was constantly triangulated across the different sources used to inform one another.For example, participant observation and field notes provided data useful (e.g.material do discuss) to seminars and focus group interviews.Similarly, seminars and focus group interviews provided valuable data in and of themselves as well as with respect to participant observation and field notes (e.g.compare what they say with what they do).Analysis of survey testimonies and school placement reports took place at the end of the study.
Data collected from different methods were first read and examined incident by incident, highlighting each relevant extract in the text.In the codification process (data condensation), these data were re-read by two of the authors.Initial codes, underpinned on the theoretical framework or generated from the data, were refined and merged by proximity in a second coding stage through a constant comparison process (data condensation and data display interaction with one another).For example, in relation to the use of the AfL principle, 'providing feedback that moves learners forward', the codes 'too vague feedback' and 'description of students' performance, without providing information to improvement' were aggregated into a broader code of 'ineffective feedback' (Figure 1).
The final phase of analysis involved data triangulation across the different sources, taking into consideration the data, AfL principles, practitioner research and LC.For example, comparing AfL principles with the enactment of AfL in the lessons observed by the researcher, together with the ideas imparted in seminars and focus group interview discussions.The analysis led to the following themes: i) teaching and learning as an aligned process; ii) facilitating students' understanding of the teaching-learning process; iii) moving from transmission towards dialogical feedback; and iv) the process of including students as assessors.

EUROPEAN JOURNAL OF TEACHER EDUCATION
In analysing the PSTs' school placement reports the researcher looked for unprompted references related to 'assessment', 'learning', and 'teacher (PST) and students' role in the teaching-learning process'.

Results
Results were organised according to the four themes derived from the data analysis which captures PSTs' change of conceptions, and enactment, of AfL.

From assessment as an 'add-on' to embedded in the teaching-learning process
In the beginning, PSTs were solely concerned with their class management skills.However, as a consequence of researching their practices and the LC support, PSTs slowly began to understand the importance of adjusting the teaching-learning process to address students' needs and changing their consideration and enactment of planning towards a more learning-oriented perspective.The feeling of needing to deliver the lesson exactly as they had intended dissipated: In the beginning (of the school placement), I was only focused on the content I had to teach.Only after the first two teaching units, I realised I was not addressing students' needs, but was already too late to make any changes.Reflecting and researching about my teaching made me consider what we did on previous lessons and how students reacted to that, how successful they were.This helped me changing from a simple perspective when planning (what I had to teach) towards a more complex (what and how to teach to help students learn).The challenges faced while teaching also provoked changes on the way I used to plan.(PST3 School placement report) [Dialogue after researcher's observation] Researcher: I noticed that you chose not to do the final learning task of the lesson plan (which was more difficult compared to the one, students were doing).Why did you make that decision?PST2: Students were already finding it difficult to do the task they were doing.I preferred helping them to have some success on that task than just move to the next one, just because it was planned.(Field notes) PSTs reported researching their own practices as central to understanding their teaching.They also noted support from the LC led them to better understand the role of assessment and the teaching-learning process.This change led PSTs to change the way they previously understood and enacted assessment, as a separate component, to that of an embedded component of the teaching-learning process: [There are any changes on the way you plan?] PST2: In the first teaching units, I planned, taught and assessed at the end what students had learned.Now, the planning is open to changes throughout the process, considering what students can and cannot do.PST1: I feel the same.Now, I observe what students fail to do, I listen to them and we work on their weaknesses in the following lessons.PST4: In my case, when I noticed students were not progressing in their learning, I thought about the methodology and strategies I was using to readjust the teaching-learning process.I had to readapt all of that.(Focus group2)

Teaching-learning process: from teacher to student-centred
In the beginning, PSTs were solely focused on having students physically active.Gradually, however, they started questioning the extent to which relying solely on being physically active could be detrimental to students' awareness of their learning: [Dialogue after researcher's observation] Researcher: How did you view students' engagement?PST4: Not great . . .Researcher: Is that just your perception or is related with the questions students were doing?PST4: Both, but mainly the questions.Researcher: What did they ask you?PST4: Why they were doing that . . .how they are going to be assessed . . .Researcher: I think these are great questions.Have you thought about them before?What have you told them?PST4: I just told them to do the sequence, be focused on the task I gave them and we would discuss that later.PSTs acknowledged that even if they shared learning goals and criteria with students, there was no guarantee that students would understand or work towards them.Questioning their practice and their students, self-analysing their instruction and sharing experiences with the LC helped PSTs address such difficulties: I remember that in the beginning (of the school placement) I was just worried about sharing goals and criteria.But when teaching, I realised that students do not always understand what we say.I reflected and also your feedback (from the researcher, university supervisor and cooperating teacher) led me to start observing if students are just doing the learning tasks or if they are focused on the goals and criteria I shared, asking students for the necessity for further clarity around my instruction, asking one or two students to explain the expectations in their own words, and also encourage students to check their learning for themselves.(PST7 Testimonial survey)

From transmission towards more dialogical feedback
While feedback was valued by PSTs, the quality of their own feedback to students was weak.PSTs tended to provide feedback that was too vague, one-directional and misaligned with students' needs and success criteria: [Dialogue after researcher's observation] Researcher: Do you remember any feedback you gave during the class?PST3: Yes, for example, I asked students to 'move their legs' to prevent the ball to hit the floor.Researcher: How did it work?PST3: Not good . . .The ball kept hitting the floor.Cooperating teacher: What do you think you could have said instead?PST3: I do not know.I thought this would be a 'familiar language' to students like some authors tend to say.Researcher: Why do you want them to move their legs?PST3: To hit the ball.Cooperating teacher: They can move their legs and not have the conditions to hit the ball, or the ball can come towards them.PST3: Maybe I should encourage them to be under the ball to be able to receive it.(Field notes) Aware of PSTs' difficulties in providing feedback that develops their students' learning, PSTs recorded and analysed their instruction.This strategy allowed PSTs, with support from the LC, to compare their feedback with what was stated in the literature as being the hallmarks of quality feedback.This helped PSTs to identify and improve their feedback and, subsequently students' learning: Feedback can be so powerful, can make a big difference on students having success on the task.However, providing accurate feedback is so difficult.Even preparing my feedback I knew my feedback was not clear all the time and it could be improved.PSTs from previous years reported such difficulties in their thesis and how recording their lessons helped them with that.I tried (recording) and I have to admit this improved so much the quality of my feedback.I noticed in the first recordings, my feedback/comments were general, or I just reinforced students, saying 'well done', 'do this'.It is reported in the literature that this type of information does not help students improve their learning.Now I ask students what the learning goal is, I mention the criteria, and what students should improve, I give them 'directions'.(Focus group2) Despite the increase in quality on the feedback provided, the greatest gain that led to an increase in the quality of the feedback provided was the understanding and use of feedback as a dialogue with students.This change resulted of what PSTs perceived as a failure on students' understanding of the transmitted feedback: The other day, after some reading, I was questioning myself, how to use feedback to improve students' learning, but in a way that engages them with that learning, not just follow what we say.I think, building a dialogue with students, for example, why we are doing this exercise?what is this going to result in?, can help students understand better what they are doing, what they have to do to improve and why.This allows us to build a trusting relationship with students, allowing them to understand that there is a reason for what we do and making that reason clear to them.Some of them may identify something we miss.(Focus group2)

From teacher to teacher and students assessment
In the beginning of the study, PSTs' lesson plans conveyed a disconnection between (self or peer) assessment and the teaching-learning process.In addition, it was apparent that students' absence of experience with assessment, not to mention a lack of understanding as to why they were being encouraged to be involved with it resulted in them encountering difficulties, and lacking interest or meaning when using self and peer assessment: PST6: (. ..) when I started using self and peer assessment, most students would not care about it, I could feel it.They would assess because I asked them to, but I cannot say I would trust their assessments.PST4: Throughout the entire year I tried to use self and peer assessment.Several studies report advantages and I considered it could be important for students to understand their performance, their difficulties, but it was disappointing . . .students could not assess . . .They just described what happened.(Focus group1) Although PSTs recognised the importance of self and peer assessment in the teachinglearning process, they struggled to make it meaningful to students.Researching their practices and realising the need of clarifying to students why they were assessing them and their peers, as well as how gaining a skill set on assessing themselves and their peers could contribute to their, and their peers', learning: PST4: I think throughout the process I became more critical about the way I prepared and used self and peer assessment.I think that influenced students, helping them to be more critical too when doing these tasks.PST2: That is why when I ask students to assess, I complete or assess their assessment, 'I agree with that, but not with this' or 'I saw it this way, but that is a fair point'.It is like assessing their assessment.For me, this needs to be done and is also important to giving them valuable information about their learning and their assessment.It can perhaps even help them improve their assessment skills.(Focus group2) PSTs also noted the importance of being constructively critical as to 'what' and 'how' they use assessment strategies/techniques.Although all PSTs agreed on the benefits of using self and peer assessment to students' learning, PSTs enacted self and peer assessment differently depending on their contexts: Although written evidence could be important, I have tried it several times and I think it does not work with my students.When self and peer assessment is oral, my students interact, give comment/feedback to others in the moment.When this is written, usually it is more difficult to them to say something and I feel they just write something but without giving too much value to it.(Focus group2) In our group (PSTs and cooperating teacher) we created some questionnaires so students could assess themselves and be aware of their learning.These questionnaires allowed students to self-assess throughout the process so they could see their progress.I also assessed their progress and added comments (in the questionnaire).I think this helped students to see what I thought about their progress, their learning, while they could also compare my assessment with theirs and improve their assessment skills.(PST1 School placement report) Those PSTs and students who were more autonomous and constructively critical were the first to attempt changes to their practices and the most effective in the incorporation of AfL.

Discussion
The difficulties shared by PSTs in adapting and adopting AfL in their teaching are not surprising, considering the substantial number of studies reporting the challenges of using AfL (Brevik, Blikstad-Balas, and Engelien 2017;Yan et al. 2021).Initial difficulties encountered by the PSTs in this study are clearly related to their initial lack of exposure, and therefore knowledge of, AfL.This led PSTs to an instrumental use of assessment at the expense of what is intended as the essence of AfL.That is, the use of assessment to improve students' learning (Wiliam 2018) can only happen when assessment is embedded in the teaching-learning process (Allal 2020;Coombs et al. 2018).Gradually, PSTs began to realise that enacting AfL requires more than just changing assessment techniques (Moura et al. 2021).Although difficult, it is possible to change PSTs' conceptions, and practices of assessment.
The gap between PSTs' use of AfL in their practice and the benefits of AfL (as highlighted in the literature), helped PSTs understand the need to investigate their practices, be constructively critical of such practices and reflect on how to improve them.As the study progressed, PSTs' critical stance extended to research.PSTs began to enhance their exposure to research to improve their teaching, reflecting and being able to recognise that what other researchers do or recommend may not work in specific contexts.This contributed to improving the limited research skills and increasing their proclivity to reflect on their teaching, two aspects that tend to be problematic for PSTs (Henning, Petker, and Petersen 2015).
The critical, reflective and researcher-like stance was supported by being a member of a LC, where PSTs could share, compare and acknowledge differences between their practices, even within the same school, when using 'the same assessment ideas'.This reinforced the notion that teachers must find their own ways of putting AfL principles into practice (Wiliam and Leahy 2015).A critical stance towards research and practices helped PSTs adapt and learn from their teaching, with the LC acting as a supportive and crucial element in challenging themselves and looking for ways to improve their practices.This underscores the importance of a supportive context in encouraging PSTs not to give up on their initial attempts to change their practice(s) (Ribaeus, Enochsson, and Löfdahl Hultman 2020;Yan et al. 2021).
The LC built around these PSTs provided them with support on their practices.This was essential to them accepting the difficulties they faced in successfully adopting the principles of AfL (Qin and Yi 2021;Schildkamp et al. 2020).Positioning PSTs as being responsible for their own learning, and positioning facilitators (the researcher, university supervisor and cooperating teacher) as collaborators, enhanced PSTs' engagement with the LC and their own professional development.It was apparent that in encouraging PSTs to consider data from their own practices that this matched a key aspect of practitioner research literature (teachers improving their own practices) (Heissenberger and Matischek-Jauk 2020;Hilton and Hilton 2017).
Researching their own practices helped PSTs better understand the teaching-learning process, be more accepting of changes to their practices, and become more constructively critical about their teaching and research.It was evident that the PSTs became more self-critical throughout the study as a consequence of their involvement in researching, teaching and belonging to a LC.This, in turn, encouraged PSTs to move from considering assessment as an 'add-on' towards an embedded process (Baird et al. 2017).Initially, PSTs tended to solely describe the events that happened in their classes and it was not until further on in the study that they became more concerned in reflecting about the impact of their teaching on students' learning.The 'safe space' and facilitators involved in the LC provided support and feedback to PSTs throughout the process of debriefing AfL practices experienced during school placement.Literature conveys that current support for PSTs to express their thoughts, be responsible for their learning, have ongoing support while enacting AfL, and experience membership of a LC is lacking.Leading PSTs to question their conceptions, to transform their future thoughts and practices might happen in this active construction of their learning and collaboration with others within learning communities.Receiving feedback in a constructive way (e.g. during the meetings, discussions, planning) provide an example to PSTs on how to provide feedback to their students.Discussing and sharing AfL strategies allowed PSTs to extend their understanding on the potential enactment of AfL, and consider how AfL might work before enacting it in their lessons.
PSTs learned to become part of the community, with their involvement in the community, being essential to develop it.Like mentioned by Skerrett and Williamson (2015), the interaction between learning communities and PSTs proved to be mutually beneficial.The different contexts in which PSTs worked, and subsequent challenges, helped PSTs understand the importance of being practitioner researchers and reflect about what they learn and do.This study showed 1) that bringing together PSTs as practitioner researchers while being members of a LC can beneficial to their learning; 2) the involvement of all stakeholders (PSTs, schoolteachers, teacher educators) in a non-hierarchical LC where all can learn from each other; and 3) the ongoing support and study design allowed transferability of learning from the LC about aspects other than AfL.Indeed, there is a need for more studies that capture the extent to which PSTs' involvement in practitioner research results in changes to their practice (Rutten 2021).

Conclusion
Encouraging PSTs to position themselves as practitioner researchers and active constructers of their learning was essential in helping PSTs explore and understand the ways in which AfL is effectively enacted in a bid to improve the teaching-learning process and students' learning.Researching their own practices developed PSTs' teaching, with PSTs becoming more critical about what they do and appreciating that what works in one context may not work in another.Exposure to the differing contexts was heightened and supported through a LC that included the main stakeholders involved in the PSTs' school placement.The LC challenged the PSTs to consider ambitious assessment practices that would focus on improving students' learning, resulting in supporting PSTs to move away from an over-reliance on envisaging sole responsibility and control as a PSTs in a lesson.
Given the importance of school placement as an opportunity for PSTs to enact assessment practices, it is vitally important that they are supported in this endeavour.The establishment of a LC that includes all those involved in support the school placement experience, provides a space in which PSTs can share their experiences, encourage PSTs to research their practices as well as learn from others on how best they might consider changing practices to heighten students' learning experiences.
PSTs valued being involved in a LC and the learning that resulted from their involvement.However, in some cases, PSTs valued some aspects like for example, using self and peer assessment, they struggled to make it meaningful to their students.Even belonging to the same LC, PSTs had different experiences due to the school context they were in, and their capacity to act as active constructers of their learning.Therefore, teacher education programmes need to seriously consider how best to provide a supportive infrastructure that captures the impact of such learning communities on PSTs' practices and, in turn, school students' learning experiences.Teacher education programmes and learning communities need to consider how best to encourage and support PSTs to research and, if possible, transform their practices.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
The work was supported by the Fundação para a Ciência e a Tecnologia [SFRH/BD/137848/2018].

Figure 1 .
Figure 1.Example of codes from one of the themes.
Researcher: Did you?PST4: no . . .Researcher: Do you think you should?PST4: yes, because I felt they were doing but not really understanding what they had to do and why.(Field notes) Although PSTs came to value sharing learning goals and criteria with students, PSTs struggled on how best to do this: PST1: I still fail to explain to the students what they have to achieve.I only share contents, and students find it difficult to clearly understand what the goal is, what they have to be capable of doing.I questioned them many times to see if they understood what they had to learn and how to do it and it was clear to me that they did not understand.PST8: I have the same problem . . .Researcher: What do you intend to do about that?PST2: I think we all find it difficult.It is new for us, something we never did, we have to learn how to do it.In my case, I recorded and listened my lessons to identify what I said and think how I could improve that.I think that helped me to see how clear I was.PST7: I do the same and it helped me too.I also have been trying to prepare what I am going to say and then I listen back to the recording to see if I did.(Focus group1)