lable at ScienceDirect

Teaching and Teacher Education 26 (2010) 979e986

Contents lists avai

Teaching and Teacher Education

journal homepage: www.elsevier .com/locate/ tate

Collaborative research for assessment for learning

Beverley Cooper*, Bronwen CowieSchool of Education, University of Waikato, Private Bag 3105, Hamilton, New Zealand

a r t i c l e i n f o

Article history:Received 18 July 2008Received in revised form6 August 2009Accepted 28 October 2009

Keywords:Assessment for learningCollaborative researchSecondary schoolPolicy

* Corresponding author at: Tel.: þ64 7 8384382; faE-mail addresses: bcooper@waikato.ac.nz (B. Co

(B. Cowie).

0742-051X/$ e see front matter � 2009 Elsevier Ltd.doi:10.1016/j.tate.2009.10.040

a b s t r a c t

This paper reports on a collaborative research study between three secondary teachers and twouniversity researchers, initiated by the teachers, into the practice and impacts of assessment for learningin science, geography and history classes. The research provides insights into how teachers andresearchers can collaborate to develop a research and practice agenda. The study illustrates the need fora dynamic interconnection between the personal, micro or school-related level and the macro or policylevel factors. The findings highlight the importance of external support, shared teacher knowledge andbeliefs, professional experimentation, and shared reflection on student responses to classroominnovations.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Teachers are being encouraged to engage with researchevidence and to participate in research studies related to theirclassroom practice with the aim to raise student achievement.Following the review by Black and Wiliam (1998), there has beena worldwide focus on the teacher use of assessment for learningpractices to improve student achievement. This paper reports oncollaborative research between a group of secondary teachers anduniversity researchers, initiated by the teachers, into the practiceand impacts of assessment for learning in chemistry, geography andhistory classes. It provides insights into one way that teachers anduniversity researchers might work collaboratively to developa research and practice agenda that is driven by the teachers ratherthan the researchers. The findings highlight the contribution ofexternal support, shared teacher knowledge and beliefs, systematicprofessional experimentation, and shared reflection on studentresponses to classroom innovations if changes to teacher's thinkingand practice are to be sustained.

1.1. Defining AfL

It is now widely recognised that Assessment for Learning (AfL),has an important role to play in enhancing student learning andachievement (Black & Wiliam, 1998). Formative assessment is theprocess used by teachers and students to recognise and respond to

x: þ64 7 8384977.oper), bcowie@waikato.ac.nz

All rights reserved.

student learning in order to enhance that learning, during thelearning (Cowie & Bell, 1999). Typically it revolves around strategiessuch as: wait time; clarifying and sharing learning intentions andcriteria for success; comment-only marking; orchestrating class-roomdiscussions using questioning andother learning tasks to elicitevidence of student understanding; providing timely feedback andpeer and self-assessment (Black&Wiliam,1998, 2009). Teachers canplan to elicit and act on information about class and individualprogress (Cowie & Bell, 1999) but the quality of interactive feedbackis a critical feature in determining the quality of learning activity,and is therefore a central feature of assessment for learning peda-gogy. Effective feedback is non evaluative, supportive, timely andspecific in providing information about howa learner might modifytheir thinking or behaviour to improve their learning (Shute, 2008).

1.2. Teacher assessment for learning practice: a personal challenge

Teacher implementation of AfL is not without its challenges andhas been the focus of numerous research and professional devel-opment initiatives internationally. One line of research focuses onexploring teacher personal knowledge and beliefs. Sato, Coffey, andMoorby (2005) working in the United States point to the role ofteacher personal beliefs and background as an influence on how, oreven if, teachers explore how AfL practices might be incorporatedinto their wider responsibilities. Black and colleagues provideevidence that different teachers adopt and adapt different practicesconsistent with their understandings of effective teaching (Black,Harrison, Lee, Marshall, & Wiliam, 2003). They argue that teacherbeliefs about their own learning and student learning and agencyunderpins the way AfL becomes embedded in classroom practice.

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986980

The promotion of learner autonomy lies at the heart of AfL and ifteachers don't appreciate this it appears that AfL can be imple-mented as a series of ritualised teaching strategies and hence losesmuch of its efficacy (Lee & Wiliam, 2005; Marshall & Drummond,2006; Sadler, 1989). Changing teacher assessment practice is notsimply a matter of increasing teachers' assessment literacy throughprofessional development about activities such as wait time itrequires conceptual shifts for all stakeholders (Tierny, 2006, p. 259).

1.3. Teacher assessment for learning practice: shaped by the context

A second line of research, typically informed by socioculturalviews of learning, has provided a more contextual explanation ofhow and why teachers might implement assessment for learning intheir classrooms (see for example Black & Wiliam, 2006, 2009;Gipps, 1999; Pryor & Crossouard, 2007). School case studies of theimplementation of assessment for learning across schools in Jersey(Jones & Webb, 2006) highlight the impact of variations in-schoolpriorities and cultures for change, along with the role and impor-tance of leadership and trust when school communities areresponding to changes, whether imposed externally or developedinternally. This said, when teachers reflected on and analysedexamples of formative assessment that worked, and exampleswhen it had not, they were able refine and develop their practice.Teachers involved in the Assessment Is for Learning programme inScotland were also influenced by contextual factors (Hayward,Priestley, & Young, 2004). They appreciated taking research ideasinto the classroom and were generally comfortable with the ideasand learner-centred classroom strategies that underpinned theformative assessment project. Nevertheless, they found the adop-tion of such approaches to be problematic because of the need toprepare children for national tests and the time involved. The studyalso highlighted the importance of credible leadership and regulardialogue with other professionals in scaffolding innovation. Whenthe principal had a real and demonstrable commitment to theinnovation they were powerful contextual catalysts for theteachers. Most commonly the teachers valued support fromcolleagues and the opportunity to talk with teachers from otherschools. Local context clearly impacts on teacher AfL understandingand implementation.

Carless (2005) explicates the influence of the broader context, inconjunction with more local factors, on teacher and school change.He proposed that teacher practice is shaped by a nested set offactors at the personal, school and political level. Carless describedhow the personal level encompasses teacher understandings ofprinciples and practices of assessment for learning and thecongruence of these with their existing beliefs. The next “micro”level of the change environment is that of school support, schoolculture and school resourcing. This is nested within a “macro” levelof external factors such as policy, infrastructure and externalagencies. In this study we highlight the interactive influence ofpersonal, school and political factors on teacher assessmentpractice.

1.4. Collaborating to investigate and enhance teacher assessmentfor learning

Teachers prefer information that is immediately and contextu-ally relevant. Much of the research on assessment for learninginvolves collaboration through university school partnerships(Black & Wiliam, 1998, 2009; Sato et al., 2005; Tierny, 2006). Thework of Torrance and Pryor (2001) is a rich example of this. Theyworked collaboratively with teachers within an action researchframe to investigate and develop formative classroom assessmentin primary schools by building on their own research. It was

important that the teachers wanted to change their practice(Priestly, 2005) and so the study beganwith the teachers examiningtheir own practice, particularly the way they questioned and gavefeedback to students. This was followed by classroom experimen-tation and careful reflection on practice. The collaborative studythat is the focus of this paper differs from the initiatives previouslydescribed as it was initiated by a teacher rather than us asresearchers.

1.5. The research goal and question

At the beginning of the study the overarching research questionwas negotiated between the teachers and university researchers. Thequestion was: What do teachers see as the impact of their use ofassessment for learning on student learning and student motivationand willingness to take more responsibility for their learning?

2. Research design

This study occurred in a large city high school of approximately1500 students aged 13e18 years old, situated in a high socioeco-nomic area. The school is coeducational and classes are organisedinto 25e30 students of mixed ability. Colin an experienced head ofdepartment became interested in AfL work carried out by theScottish Executive Education Department designed to empower thelearner and help encourage independent learning methods duringa staff sabbatical. He was particularly interested in self-evaluationby students of their own learning needs and their participation innegotiating personal learning targets to improve their achieve-ment. At around the same time Colin's school management carriedout a strategic planning exercise that identified that studentperformance in the recent national examinations did notadequately reflect the calibre of the student body when comparednationally and set raising achievement in external qualifications asa school wide strategic goal. Colin considered assessment forlearning had the potential to support this goal and with the supportof his principal approached the authors of this paper.

2.1. The participants

Colin convinced some colleagues, also Heads of Department,that developing AfL practices would contribute to their obligationsin relation to the strategic plan by raising student achievement. Heinvited the authors of this paper to present an overview of AfLprinciples to interested staff. After the presentation, the groupstrategised how the teachers and researchers could work together.Six teachers expressed interest in being involved in a research anddevelopment project that investigated the impact of AfL on theirclassroom practice and student learning in years 11e13 (15e18 yearolds), these being external examination classes. Three teacherschose to participate in the full study; the other teachers withdrewdue to other commitments. Each teacher selected a class to workwith that they felt would benefit from using AfL strategies tosupport learning. Colin selected a year 12 Chemistry class (16/17year olds), Simon a year 11 History class (15/16 year olds) andCampbell a year 11 Geography class.

2.2. The context for teacher assessment practice

Assessment for Learning has been of interest in New Zealandsince the late 1980s (Department of Education, 1989) but to date,there has been no research that documents teacher use of AfLstrategies across a range of curriculum subjects in New Zealandsecondary classrooms. AfL was mentioned explicitly in the NewZealand Curriculum Framework policy document (Ministry of

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986 981

Education, 1993) and implied within the supporting nationalcurriculum statements. It is emphasised in the revised nationalcurriculum statement released in 2007 (Ministry of Education,2007) and in recent policy documents (Absolum, Flockton, Hattie,Hipkins, & Reid, 2009; Ministry of Education, 2009). The NewZealand Teachers Council Graduating Teachers Standards (2007)implemented in 2009 also emphasise the role of assessment andstipulate that graduating teachers should be able to: use evidenceto promote learning; systematically and critically engage withevidence to reflect on and refine their practice; and gather, analyseand use assessment information to improve learning and informplanning.

Currently the New Zealand Ministry of Education employs Assessto Learn facilitators who provide in-school professional developmentand support for assessment for learning practices. Many New Zea-land teachers, particularly in the primary sector, are now familiarwith the key aspects of AfL. The development of new standards basedhigh stakes assessment for the senior secondary school (NationalCertificate in Educational Achievement, [NCEA]) has encouragedsecondary teachers to review their assessment practices. Studentscan gain each standard at different levels of achievement: achieved,achieved with merit or achieved with excellence. To achieve atexcellence level students need to be able answer questions thatinvolve more complex higher order thinking, such as synthesiseinformation and/or apply information to new contexts. There isa focus on teaching for understanding rather than recall to enablestudents to achieve at this level. National teacher professionaldevelopment for the implementation of the NCEA, focussed on thepurposes of different types of assessment and this included theeffectiveness of assessment for learning. New Zealand teachers havebeen encouraged to engage with research through the publication ofa series of syntheses of international research (see for example Alton-Lee, 2003). Research projects funded within the New ZealandMinistry of Education's Teaching and Learning Research Initiative(Ministry of Education, 2003) are expected to include teachers asactive participants in the research process. This has not beenwithoutchallenge for both teachers and researchers (Haigh & Dixon, 2007).

At a school level the recognition from the partnership schoolthat exemplary teachers should be rewarded with scholarships tosupport professional development opportunities was a factor inthis study. Colin's visit to Scotland confirmed that AfL principlesaligned with his beliefs about students needing to take responsi-bility for their own learning and that this would fit with the schoolstrategic goal of raising achievement in the senior school. Inci-dentally, the institutions he visited were drawing on AfL work fromthe University of Waikato. He knew the Waikato researchers anddecided to approach them to further his understanding. Hispresentation to school staff describing his sabbatical experienceand his enthusiasm and vision were pivotal in recruiting hiscolleagues to participate in the project.

2.3. The collaborative research and development process

The project was framed as a research and development initiativeto meet the needs of both the teachers and the universityresearchers. Throughout the project we were ever mindful that itwas a privilege to be invited to collaborate with a cross curriculagroup of teachers in a secondary school. We were careful toacknowledge the professional commitment of the teachers and toensure that the teachers maintained control of the overall directionof theprojectwhilst stillmeeting thedemandsof a rigorous researchprocess. The teachers were named as co-researchers in the univer-sity ethical consent process. The teachers then gained ethicalapproval for the study fromtheir school principal and the students inthe class they had selected to work within the study. As well as

exploring their AfL practices the teachers developed an under-standing of the requirements of research in terms of systematic datacollection, deadlines, funding constraints, analysis, reportingprocesses and the need for the dissemination of findings to others.

During our collaboration we took cognisance of Clarke andHollingsworth's (2002) model for professional growth. In this theyproposed it is important to take into account four interconnectedchange domains: external sources of information, teacher knowl-edge and beliefs, professional experimentation and salient class-room or student outcomes resulting from experimentation (Clarke& Hollingsworth, 2002). In line with this, we contributed researchinformation to discussions about theory and practice and theteachers explored different teaching and learning approaches andmonitored the impact of these.

There were two parts to the data collection. The three teachersand two researchers were involved in a series of meetings facili-tated by the university researchers, to develop a shared under-standing of the principles of AfL and the research process. Theyshared and critiqued potential AfL strategies (some sourced by theresearchers and others from the teachers themselves) and followedthis with classroom trials of selected strategies. The differencebetween research and informal classroom trialling was exploredand the teachers were provided with a template to assist them todevelop their individual research plans. The teachers decided tofocus on the achievement of one of their examination classes.Teachers developed individual research questions and plans withthe assistance of the researchers, that focussed on an aspect of AfLthat they believed would support the selected classes to take moreresponsibility for their own learning and thus raise their achieve-ment. Timelines were negotiated with each teacher. These includedschedules for regular meetings, the development of individualteacher research plans, collection of data, document collection bythe teachers and the researchers, classroom observations, studentsurveys and interviews and the final report of their findings.

The university researchers kept in regular contact with theteachers by phone, email and through face-to-face meetings. Theresearchers operated an open door policy and the teachers wereencouraged to call in at the end of the day or make contact byphone or email to discuss what they had been trying out. Theteachers prepared a final report of their study using a template weprovided. We also provided assistancewith the final report writing.During this time the researchers were involved in practicum visitsfor cohorts of students and made informal contact with theteachers if they were in the school. Group meetings occurred at theuniversity or in the school depending on the needs of the teachers.Some times these were held after school hours and on occasionsrelease time from classes was funded to ensure teachers couldcommit to the project.

In the classroom trialling part of the data collection the teachersintroduced the students to the new strategies, collected studentwork samples, recorded their own reflections and observations,conducted student surveys, and interviewed their students aboutthe effectiveness of the new strategies. The student work samples,reflection diaries, student work samples and surveys were sharedwith the researchers. The university researchers carried out someclassroom observations and the teachers also visited each othersclassrooms and conducted informal observation and talked tostudents. The teachers shared the same workspace and oftencollaborated informally and shared successes and failures, whichthey found invaluable.

3. Findings

The findings are presented in two parts. In the first part theteachers' experiences are presented as stories of change, one from

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986982

each teacher involved in the project. We chose to report the findingsin this way to maintain the integrity of each teacher's project. Thesecond part of the findings is the story of the whole teamcollaboration.

3.1. Colin's experience

Colin's research focus was on the impact that personal learningplans had on student motivation and achievement in his year 12chemistry class (16 and 17 year olds). He also wanted to establish ifthere was any relationship between the students' self-assessedapparent understanding of topic specific learning outcomes andunderstanding as evidenced by topic test scores. Colin explained:

.ideas such as encouraging ‘self evaluation by pupils of theirown needs’, ‘negotiating personal learning targets’, ‘empower-ing the learner’, ‘helping with the development of independentlearningmethods’, as well as' improving pupil attainment’. I wasquite impressedwith these ideas and thought that it would haveapplication in the designing of a personal learning programmefor my Year 12 Chemistry students.

He chose a year 12 class because the school perceived the overallyear 12-examination results to be somewhat disappointing and theschool strategic goal was to raise their achievement. He consideredyear 12 to be significant in students' social, emotional and intel-lectual development, all the more so because it was the second ofthe three years that students would participate in external exam-inations as described below.

The other reason I chose Year 12 was that I believe that year 12in secondary school, is an interesting year in terms of studentdevelopment, socially, emotionally, and academically. It is sortof a transition year that the enthusiasm they had for the externalexam system in Year 11 is now been tempered by their results,and they know that they still got another year to prove them-selves as year 13's, . it is also year of massive social andemotional development in my view.

Colin developed personal learning plans with each of hisstudents, this process aimed at helping his students to individuallyplan ahead and to takemore responsibility for their own learning toimprove their achievement. He developed templates to guide himin his student questioning. He interviewed each student twice andsurveyed him or her three times during the year to establish theirpersonal goals, and to provide the students with time to monitortheir achievement against their goals and to assess their ownprogress in understanding the specific learning outcomes for theirchemistry course. He scheduled interviews during quiet classroomtimes and on one occasion while on a bus during a field trip. Oneach occasion he asked for feedback on his questionnaires andinterview sheets from the researchers and he made modificationswhere necessary.

He reported that the ‘snippets of information’ divulged by thestudents gave him a better understanding of their motivations forstudying chemistry and their expectations of him and the course.He commented that he had gained insight into the way thestudents worked and developed relationships with a greaternumber of the students.

I know the students a bit better. I think that these year 12students are quite fragile and they do need someone to take aninterest in them and to be on at them in a positivewaye Are youup to the mark? How are those goals going? In the backgroundthey gave to me, some of them told me about their interests andI could talk to them about how is the springboard diving going?Or your ballet level 8 how is that doing? It is just means that we

have a personal understanding of them and perhaps they realisethat someone is interested in them.

He also developed a better understanding of each individual stu-dent's approach to learning and how they did, or did not, individuallymonitor their learning and how he might assist them with this.Overall, he was surprised at the students' lack of awareness of theirown learning processes and the nebulousnature of their future plans.

I was astounded to find that the majority of them do not selfevaluate. Apparently, they do not reflect much on their learning.It is as if they live very much in the present with little or noconcern about the future. I found that quite staggering. They sitthe test on Friday; get results backon theMondayand then that islocked away in the past. They didn't really think about howwellor how poorly they had done andwhat they could do to improve

Colin also gained some insights into the effectiveness of histeaching through the students' self-monitoring of their achieve-ment against the specific learning outcomes for their chemistrycourse. Colin asked them to “traffic light” their perceived under-standings after they finished each topic of his course and again afterthey had sat the topic test. He was then able to collate this infor-mation and assess where he should put emphasis in revisionprogrammes.

I have clear understanding of how they [the students] operate.Their attitudes needed to focus on better learning and under-standing and the SLO [Specific learning outcome] sheets that Igot from them have indicated to me, weakness in my teachingthat I can perhaps sort out for next year.

Colin observed that students became more realistic with theirself-assessment of their understanding of chemistry concepts asthe year progressed.

Initially students were happy to ‘green light’ SLOs when inactual fact an “amber light”was appropriate. As the programmeevolved the students became more conservative in their ‘Trafficlighting’.

He described the process of working with his students over timeto develop their self-assessment skills as one of the “most invigo-rating and satisfying” of his career because it provided an outlet forhis creativity and “opened up a new perspective on good teachingpractice.” It confirmed to him that teaching was not just aboutcontent and that it was important to give students opportunitiesand processes to take responsibility for their own learning and toreflect on and review their progress.

Most of the students who participated in the personalisedlearning programme interviews, surveys and traffic lighting of theSLOs, felt they were useful in keeping them focussed on learning.The students viewed the ongoing nature of the interviews withColin as important. The following is a representative studentcommentary on the personalised learning programme.

I think the programmewas useful throughout the year as it keptme focussed on my original goals. The “traffic lighting” of SLOshelped in showing my strengths and weaknesses in each topicand this helped me with my study techniques. It was effectivewhen therewere interviews throughout theyearbecause itmademe realise howmucheffort I had toput in andhowmuch revisingI had to do to obtain the goals I had set from the beginning of theyear. Also traffic lightingof the SLOs before andafter showedhowmuch I actually understood and what need to be revisited.

Some students acknowledged that the personalised learningprogramme would have a long-term impact on their approach tostudy as one remarked:

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986 983

The programme didn't really help me take more responsibilitybut it did emphasise the fact that I will have to put a lot moreeffort into my subjects THROUGHOUT the year and not try andcramp everything into my head at the very end. I have alreadymade the decision to set aside three to four hours a day toreview things throughout the year of 2007. I will stick to thisdecision because I have learned it is rewarding and I will mostcertainly get results.

3.2. Campbell's experience

Campbell explored if the use of learning intentions and studentnegotiated success criteria could improve the questioning skills ofboth the teacher and the students in a year 11 geography class.Campbell considered that student achievement in external exam-inations was closely linked to their ability to answer high orderquestions because ‘excellence’ questions in NCEA fell into thiscategory. He also felt that if the students understood the learningintentions they would be able to formulate more sophisticatedquestions and develop a deeper understanding of ideas to producequality answers to examination questions. He discussed theformulation of learning intentions with the researchers and hiscolleagues and the researchers contributed professional develop-ment and resources in this area both individually and in the jointmeetings. Campbell outlined and wrote on the white board hislearning intentions for a lesson and negotiated success criteria withstudents. He noted that thorough planning of his lessons wasimportant in this process.

My planning became vital so that learning intentions could bedeveloped for each lesson and series of lessons. A thoroughknowledge of the topics covered was essential when I wasnegotiating success criteria with the students.

Campbell's focus on learning intentions had a positive impact onhis classroom practice and the students' ability to answer highorder questions. He felt that by identifying specifically what wouldbe learned in each lesson or sequence of lessons it conveyed to thestudent exactly what was to be accomplished.

Because the success criteria were negotiated with the studentsthey had ownership of the tasks and were more likely tocomplete activities due to them being able to see where the taskwas taking them. The activities became more relevant as thestudent was able to check their own progress.

Campbell worked on his questioning skills and reported that:

I have developed several new strategies that I have been able totrial and anecdotally I have noticed that my students havebenefited by; providing better answers in more detail, morestudents are now involved in answering and discussions,student confidence in their own ability [to answer questions]has increased. I have noted that my planning and preparationhas also improved and that I am trying to be more innovative inthe activities that I do.

Campbell worked to support his students to pose questions aboutthe key ideas of a lesson/unit to themselves and their peers. Studentswere encouraged to develop their questioning techniques throughthe use of a question dice strategy. Initially students developedquestions using simple What?, Where?, Who?, Why? When? andWhich? prompts as the basis for their question. As students becamemore competentwith creating their own questions theywere able todevelop questions that promoted higher order thinking such asanalysis and evaluation. This generated productive discussionsbetween the students because they were expected to have someunderstanding of the answers to the questions they asked.

By using the question dice students not only had to come upwith the question, they also had to know what the answer wasand be able to communicate the question effectively as well asdecipher the answer

The researchers' observations of Campbell's class providedevidence that students were engaged in lively and critical dialoguewhen set tasks involving peer questioning. Students were positiveabout the AfL strategies. They felt they developed a fuller under-standing of an idea because they had to explain to others and werechallenged to extend their thinking

The learning intention helped to clarify what was to be learnt ineach lesson.They [the questions] made me think about the answers andwhat we had learnt.They were informative. When we learn things, we may under-stand it but not know how to explain it, discussing differentaspects allows us to put things into our own words and alsointeract with our classmates.It also helped when I got a questionwrong. I was able to listen tothe correct answer so I still learnt when I got a question wrong.

Overall, Campbell felt that the use of learning intentions andquestioning enabled him to make judgements about studentprogress with more confidence and to make adjustments to histeaching program to better cater for learner needs.

3.3. Simon's experience

Simon chose to investigate the effect of using feedbackcomments rather than grades in his year 11 History class (15e16year olds). His aim was to provide structured and specific feedbackto students to improve their essay writing skills while using histime efficiently. Simon devised a feedback sheet to structure hiscomments and ensure he covered key skills that he felt needed tobe addressed by the students if they were to achievemore highly inexaminations. The feedback sheet included sections on essaystructure, content and style along with a section where Simonnoted what the student could do next time to improve their essay.There was no space for a grade.

When they first received comment-only feedback Simon'sstudents pressurised him to give them grades. Simon was able toresist this pressure and justify this action to his students because ofthe insights he gained from student responses to his feedback.From their comments Simon identified that many of his studentshad failed to understand fundamental concepts and the vocabularyessential for good history writing. He explained, “I wrote ‘Gener-alisations need work.’ A student asked what a generalisation was.”By monitoring student responses to his feedback he came toappreciate that his students needed specific feedback informationto help them improve their essay writing. Some typical studentcomments were:

It tells me what to improve but not how to improve it.I don't understand generalisation and how to add it in.I would like to have known what I should have put in certainplaces instead of what I put.

Simon's experiencewas consistent with research elsewhere that“Students are wedded to idea of grades” and that changing class-room practices requires teachers to renegotiate with studentsclassroom norms and expectations (see for example the work ofBlack et al., 2003). He recognised, “ Getting students to use thefeedback to improve their work is the key thing.” Although he hadread about the need for descriptive feedback in the literature it wasonly through the monitoring of his own student responses that he

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986984

came to appreciate the nature of descriptive feedback that wasuseful to his students.

Simonmodified his feedback template to take into account whathe had learned from student comments. He expanded the space forhis teacher comment and added a section for the student tocomment. This had two prompts: I think I improved on. and nexttime I plan to.. Simon used the revised form to give the studentsdetailed feedback. He then asked the students to complete thestudent comment section. Some student comments were ‘precep-tive’ and reflective, indicating the students had analysed Simon'sfeedback. Simon added an overall grade of achieved (A), merit (M)and excellence (E) to the sheet once the students had added theircomments. The students were asked to transfer their commentunder “Next time I plan to.” to the top of their next essay tocontinue a dialogue between himself and the student about theirachievement and progress.

Simon reported that students were positive about this process.The students considered that they were better informed aboutwhat was expected andmore in control of their learning. Theyweremore aware of what they needed to do. Simon noticed a change instudent attitudes and believed they became more focussed on howto improve their essay writing. Essay writing skills improvedparticularly for the “struggling” students. Simon commented thathe could not specify if there had been an improvement inachievement for this particular group of students but he consideredthat their experience of learning History was more positive. He sawthat developing self-assessment skills was a long-term process asthey needed to develop both skills and the vocabulary of reflection.He also found the student feedback to him very useful to refine histeaching focus. He saw the next stage was to use peer discussion tocomplement the interpretation of his comments. He reported hisinvolvement in the project had been “successful and rewarding”and it had reinforced the priority that he accorded to learning andteaching over externally driven summative assessment.

3.4. The team's collaborate research story

There were many factors that contributed to the success of thisresearch project including the support for the collaborationbetween the teachers and researchers from their institutions andthe professional interest and commitment of the teachers.

The teachers considered the involvement of the universityresearchers and the explicit negotiated and structured timelineswere essential to the success of the project. External support andaccountability were both crucial for maintaining a focus by theteachers on the development of their AfL practices as Colinexplained.

The involvement of external ‘moderators’/‘mentors’/ as ‘supportpeople’ has been essential. Time lines have been set and adheredto. Milestones have been placed and reached. Without this, itwould have been very difficult in the business of the school day/week/term to keep at it.

Teachers developing their own research questions with thesupport of the researchers meant they personalised the research tomeet the learning needs of their selected class in the curriculumarea being researched. The funding obtained by the researchersgave the teachers opportunities for classroom release time todevelop and discuss their evolving understanding of AfL practices.Apart from the agreed after school joint meeting times the teachersnegotiated the use of release time with the school managementwithin the school day to meet their research focus. Joint meetingsbegan with teachers reporting back on and discussing their prac-tice, impacts on student learning and engagement, and theirunderstanding of AfL. The teachers built on and critiqued each

other's ideas and often provided additional instances and examplesof similar practices and student responses. The researchersprovided examples from their experiences and suggested addi-tional ideas and readings that they thought would enhance theteachers’ understandings. The cross curricula nature of the teamwas important in this because it enabled cross-fertilisation of ideasas teachers “fed off the positives they had experienced.” They feltthey gained a more comprehensive understanding of assessmentfor learning by being able to “bounce ideas and get different pointsof view”, including those of the university researchers. Simonremarked that:

By meeting and presenting as a result of this research project Ibelieve it has stimulated me to look into formative assessmentin amore structuredway. The project has allowedme to developstrategies and trial these with classes. Having time off to reflecton what I have done has enabled me to modify and analyse thestrategies to ensure the benefits are realised. By working withColin and Simon, I have been able to feed off the positives thatthey have experienced. Because each of us is doing a differentarea, I believe we have all gained a greater idea of a lot moreaspects of formative assessment than working as individuals. Italso gives us the chance to bounce ideas and get different pointsof view.

Discussion in team meetings centred on how, when and towhat effect students had engaged in AfL processes. This dialogueassisted in the development of a shared understanding of theunderlying principle of AfL as that of students taking moreresponsibility for their learning and assessment. The discussionalways converged on the value of students taking responsibilityfor their own learning no matter how divergent the discussionand debate became.

The teachers noted how the location of their workspaces facil-itated the informal sharing of ideas and findings from their class-room research. They remarked how much they enjoyed learningfrom each other and that the value of cross-fertilisation of ideasfrom different curriculum areas.

The teachers experienced significant success with the strategiesthey trialled and were keen to share this with others. The teacherspresented their findings at three national conferences supported bythe university researchers. One conference focussed on AfL and theother two were subject specific. They valued the opportunity tointeract with fellow researchers and teachers and were delightedby the positive feedback they received after their presentations.

The assessment conference we attended as part of our researchproject was a real highlight. In particular, it was great to meetwith others trying to do similar things and we really appreciatedthe supportive comments about the practical and school-lednature of our efforts. It gave us a real boost.

One teacher began a professional enquiry group within theschool to assist other staff learn about and work with AfL strategiesand all three shared their research with their departments.

4. Conclusion and significance of the study

The study provides an example of how teachers can drivea research agenda for their own purposes in collaboration withuniversity researchers. The impact of a single motivated teacherand the official endorsement by the school of the teacher's capacityto drive the initiative should not be underestimated in the changeprocess (Priestly & Syme, 2005). As researchers it was rewarding tosee the professional growth of the three teachers during the projectas they developed a shared understanding of AfL principles. It was

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986 985

a privilege to work with a group of teachers who were committedto enhancing their practice through a research and developmentprocess. The study highlights the importance of external support,involvement and input, shared teacher knowledge and beliefs,professional experimentation, and shared reflection on studentresponses to classroom innovations (Clarke & Hollingsworth,2002). The involvement of the university researchers served asa catalyst to motivate the teachers to systematically experimentand reflect on their classroom practice. The project became a jointenterprise with its momentum sustained by regular contact withthe researchers. The university researchers contributed ideas aboutAfL that extended the teachers' vocabulary assisting them inmaking explicit their tacit knowledge when analysing and sharingthe impact of changes to their practice on student learning.

This study demonstrates that collaborative research can besuccessful when there is a dynamic, mutually respectful interactionand interconnection between personal, micro or school-relatedlevel andmacro or policy level factors (Carless, 2005). The impact ofpersonal factors intersecting with school/micro factors is veryevident in this study. Initially the motivation to be involved in theproject was embedded in the ‘micro level’ where the internalsupport from the school for professional development provided anopportunity and incentive (Spillane, 1999) for Colin to extend hispractice. It was the congruence between his beliefs and hisunderstanding of the principles of AfL that sustained Colin's activeinitiation of the project. The school's strategic focus on raisingexamination achievement (a micro level factor) was a response tothe high stakes testing environment (a macro level factor). Externalmandates in the forms of expectations and accountabilities havea role to play in embedding good practice (Swaffield & MacBeath,2006) but it was Colin's vision and action of linking the personal,micro and macro environments that led to wider teacher involve-ment in AfL innovation. Formal support and recognition by theprincipal of the school for the teachers' involvement in the projectat the micro level was significant as was the university researchers'ability to guide the teachers through the planning and develop-ment of research questions and plans and provide some resourcing.Also important was the teachers' perception that AfL could meetschool strategic goals.

Teacher understanding and valuing of the underlying principlesof an educational innovation is essential. It was notable that in thisstudy, students taking responsibility for their own learning(Marshall & Drummond, 2006), was a guiding theme for teacherdecision-making and critical reflection. As the teachers involvedbuilt up their knowledge of the principles and practices of AfL therewas an obvious congruence of their personal beliefs about teachingand learning and the underpinnings of AfL. However it is importantthat teachers pursue areas of personal professional relevance.Teachers working as co-researchers on their own research ques-tion, within the framework of an overall project question, is aneffective approach to collaborative research. The three teacherscontextualised and investigated different aspects of AfL, which theyperceived, would enhance their classroom practice and interaction.Colin was convinced that individual learning plans would supportstudents learning in the long-term and enable him to developbetter relationships with students. Campbell recognised that heneeded to bemore focussed in his planning and saw the refinementof his ability to write learning outcomes and negotiate successcriteria would assist both himself and the students to focus onhigher order thinking and Simon perceived that improving hiswritten feedback in a way that it was not too time consuming wasthe key to lifting student achievement in his history class.

In collaborative research all participants benefit from sharing ofideas, concrete examples of successful and unsuccessful practiceand drawing on each other's expertise (Black & Wiliam, 1998;

Fullan, 2005). Emotional, social and intellectual support, areimportant factors in sustaining classroom experimentation.Working with colleagues from different curriculum areas providesaccess to different perspectives and constructive feedback andsupport for innovation, as does the involvement of universityresearchers. Teachers gain motivation through the affirmation oftheir peers whom they respect professionally. Classroom experi-mentation and dialogue enables teachers to get beneath the surfacefeatures of AfL strategies to appreciate its implications for theirteaching (Spillane, 1999). A spiral of discussion and classroomexperimentation can develop a sense of obligation to each otherand to a research project, which can motivate teachers to explorechanges in their practice (Spillane, 1999). As a result of their focuson AfL the teachers considered they had greater insights into theirstudents' understanding of the content, the ways individualstudents approached learning, and teaching strategies thatcontributed to better student engagement with their own learningand assessment. The teachers observed that students were moreengaged in their learning and self-assessment, which was rein-forced by student written, or verbal comments. As researchers webenefited from robust discussion with classroom practitionersabout the intersection between theory and practice. The jointdisseminating of findings at teacher conferences enhanced theprofile of the research findings and confirmed the value of collab-orative projects. The university researchers and teachers havecontinued to collaborate as the teachers have moved to work withcolleagues within their departments to develop and extend theirAfL practices. The next phase of the research process is to establishif AfL as an educational innovation can be sustained and diffusedacross departments in the school and to identify the factors thatassist in embedding AfL in teachers' practice.

References

Absolum, M., Flockton, L., Hattie, J., Hipkins, R., & Reid, I. (2009). Directions forassessment in New Zealand: Developing students' assessment capabilities.Wellington: Ministry of Education.

Alton-Lee, A. (2003). Best evidence synthesis: Quality teaching for diverse students inschooling. Wellington, NZ: Ministry of Education.

Black, P., Harrison, C., Lee, C., Marshall, B., & Wiliam, D. (2003). Assessment forlearning: Putting it into practice. Maidenhead: Open University.

Black, P., & Wiliam, D. (1998). Assessment and classroom learning. Assessment inEducation, 5(1), 7e74.

Black, P., & Wiliam, D. (2006). Developing a theory of formative assessment. InJ. Gardner (Ed.), Assessment and learning (pp. 81e100). London: Sage Publications.

Black, P., & Wiliam, D. (2009). Developing the theory of formative assessment.Educational Assessment Evaluation and Accountability, 21, 5e31.

Carless, D. (2005). Prospects for the implementation of assessment for learning.Assessment in Education, 12(1), 39e54.

Clarke, D., & Hollingsworth, H. (2002). Elaborating a model of teacher professionalgrowth. Teaching and Teacher Education, 18(8), 947e967.

Cowie, B., & Bell, B. (1999). A model of formative assessment in science education.Assessment in Education, 6(1), 101e116.

Department of Education. (1989). Assessment for better learning: A public discussiondocument. Wellington: Department of Education.

Fullan, M. (2005). Leadership and sustainability: System thinkers in action. ThousandOaks, CA: Corwin Press.

Gipps, C. (1999). Sociocultural aspects to assessment. Review of Research in Educa-tion, 24, 357e394.

Haigh, M., & Dixon, H. (2007). ‘‘Why am I doing these things?”: engaging inclassroom-based inquiry around formative assessment. Journal of In-serviceEducation, 33(3), 359e376.

Hayward, L., Priestley, M., & Young, M. (2004). Ruffling the calm of the ocean floor:merging practice, policy and research in assessment in Scotland. Oxford Reviewof Education, 30(3), 397e415.

Jones, J., & Webb, M. (2006). Assessment for learning (AfL) across the school: A casestudy inwhole school capacity building. InPaper presented at theBritish educationalresearch association annual conference, University ofWarwick, 6e9 September 2006.

Lee, C., & Wiliam, D. (2005). Studying changes in the practice of two teachersdeveloping assessment for learning. Teacher Development, 9(2), 265e283.

Marshall, B., & Drummond, M. (2006). How teachers engage with assessment forlearning: lessons from the classroom. Research Papers in Education, 21(2),133e149.

B. Cooper, B. Cowie / Teaching and Teacher Education 26 (2010) 979e986986

Ministry of Education. (1993). The New Zealand Curriculum Framework. Wellington:Learning Media.

Ministry of Education. (2003). Teaching, learning and research initiative. Retrieved04.08.09, from http://www.tlri.org.nz.

Ministry of Education. (2007). TheNew Zealand Curriculum.Wellington: LearningMedia.Ministry of Education. (2009). Assessment: Teaching and learning. Retrieved

22.07.09, from http://www.tki.org.nz/r/assessment/one/formative_e.php.New Zealand Teachers Council. (2007). Graduating teacher standards: Aotearoa

New Zealand. Retrieved 04.08.09, from http://www.teacherscouncil.govt.nz/education/gts/index.stm.

Priestly, M. (2005). Making the most of the curriculum review: some reflections onsupporting and sustaining change in schools. Scottish Educational Review, 37(1),29e38.

Priestly, M., & Syme, H. (2005). Formative assessment for all: a wholeeschoolapproach to pedagogic change. The Curriculum Journal, 16(2), 475e492.

Pryor, J., & Crossouard, B. (2007). A socio-cultural theorisation of formativeassessment. Oxford Review of Education 1e20.

Sadler, R. (1989). Formative assessment and the design of instructional systems.Instructional Science, 18(2), 119e144.

Sato, M., Coffey, J., & Moorby, S. (2005). Two teachers making assessment forlearning their own. The Curriculum Journal, 16(2), 177e191.

Shute, V. (2008). Focus on formative feedback. Review of Educational Research, 78(1),153e189.

Spillane, J. (1999). External reform initiatives and teachers' efforts to reconstructtheir practice: the mediating role of teachers' zones of enactment. Journal ofCurriculum Studies, 31(2), 143e175.

Swaffield, S., & MacBeath, J. (2006). Embedding learning how to learn in schoolpolicy: the challenge for leadership. Research Papers in Education, 21(2),201e215.

Tierny, R. (2006). Changing practices: influences on classroom assessment. Assess-ment in Education, 13(2), 239e264.

Torrance, H., & Pryor, J. (2001). Developing formative assessment in the classroom:using action research to explore and modify theory. British Educational ResearchJournal, 27(5), 615e631.