This summary briefly sets out the background, rationale and methods used to conduct this systematic review. The results are outlined in relation to the design, content, methodology and context of the studies involved. The summary then outlines the findings in relation to the review questions and concludes with implications for practitioners and policy-makers.

Background

Our concern as a Review Group is to help inform practical choices made by those who choose continuing professional development (CPD) activities and those who plan them. This review, comparing the impact of collaborative and individually oriented CPD, builds upon an initial review which only explored the impact of collaborative CPD on teaching and learning. We have been encouraged by the evidence of connections between sustained, collaborative CPD and positive benefits for teachers and pupils, and by the extensive interest in the first review from national and local agencies concerned with providing or facilitating CPD for teachers. We aimed in the second review to see if non-collaborative (individually oriented) CPD was capable of similar impact to collaborative CPD. By updating the searches for studies of collaborative CPD, the second review also created an opportunity to test and amend the first review findings.

The second review has been supported by the Department for Education and Skills (DfES) in two ways: through the national CPD strategy and via registration with the Evidence for Policy and Practice Information and Co-ordinating Centre (EPPI Centre), based at the Institute of Education, University of London. The General Teaching Council (GTC), in accordance with the Council's CPD policies and strategy, also supports the Group, as does the National Union of Teachers (NUT). As with the first EPPI Centre review, the Centre for the Use of Research and Evidence in Education (CUREE) has also made a significant financial contribution.

Aims

Our aim was systematically to review the literature on CPD in order to discover evidence about the impact on teaching and learning of individually oriented and sustained CPD interventions and to compare this with evidence about the impact on teaching and learning of sustained, collaborative CPD (based on the findings of the first review). We also aimed to apply the findings of the first review to any further studies of collaborative CPD identified in the updated searches to explore the nature of collaboration further.

Definitions

For the purposes of this review, 'collaborative CPD' refers to programmes where there were specific plans to encourage and enable shared learning and support between at least two teacher colleagues on a sustained basis. 'Individually oriented CPD' refers to programmes where there were no explicit plans for the use of collaboration as a significant learning strategy and/or no activities explicitly designed to support/sustain such collaboration. 'Sustained CPD' refers to programmes that were designed to continue for at least twelve weeks or one term. The review includes those studies of CPD which report evidence of impact, either positive or negative, on teaching and learning.

Review questions

The overarching question for the review is as follows:

How do (1) collaborative and sustained CPD and (2) sustained but not collaborative CPD affect teaching and learning?

This is followed by the sub-question:

(3) How do the findings from (1) and (2) compare?

Our first CPD review addressed the first component of the over-arching question considering the impact of collaborative and sustained CPD. This second review focused on the second component of the main question. It aimed to identify studies that have investigated sustained but not collaborative CPD and to compare the findings from the first and second reviews. The searches of the first review were also updated and additional collaborative studies were identified. Therefore the findings of the first review were applied to any other recent studies identified in the second review to explore further the nature of collaboration.

Rationale

Like the first review, this second review grew out of a genuine interest on the part of the Review Group, driven by practitioner concerns and shaped within a policy context of continued devolution of CPD resources to teachers and schools. Much of the literature relating to teaching as a research and evidence-informed profession (Cordingley and Bell, 2002) emphasises the professional development benefits of sustained professional dialogue between teachers and of experimenting with and adapting new approaches - as does the theoretical and analytical literature about CPD. For example, Joyce and Showers (1988; 2001) give considerable emphasis to experimentation and coaching over time. Our approach also drew on the teacher development literature, including Hargreaves' (1993) model of the way in which teachers are able to extend aspects of practice cumulatively and Rich's (1993) work on the learning of beginning and expert teachers.

The literature on CPD outcomes (including Harland and Kinder (1997), Joyce and Showers (1988, 2001) and Day (1999)) convinced us that the question would identify studies likely to produce findings of interest to practitioners and policy-makers.

For the second review, we felt it important to examine CPD from an individually oriented perspective in order to consider the distinctive contributions and/or advantages of both collaborative and individually oriented CPD. In addition to this comparative exploration of the relative impacts of collaborative and individually oriented CPD, by updating the searches for collaborative CPD and applying the findings of the first review to any additional collaborative studies identified, this review was also able to build on the findings from the first review which linked collaborative (and sustained) CPD interventions with positive changes in teacher attitudes and behaviours, and with beneficial pupil outcomes. We wanted to develop the evidence base for our knowledge and understanding of CPD processes, and their impact on teaching and learning.

We were also keen to explore the nature and relative importance of collaboration more systematically. Our intention, by the time a third review is complete, is to have begun to develop a more detailed understanding of the constituents of effective collaboration, based on the evidence from the studies in the three reviews.

Methods

Identifying, describing and appraising studies

For practical reasons, this review focused on studies published after 1991 that were reported in English, although no geographical limits were set. We wanted to engage the interest of both primary and secondary practitioners, so the review included studies that involved teachers of the 5-16 age group. While this excluded further education (FE) and sixth-form college practitioners, it did not exclude those who teach within the 11-18 age range. The studies had to have a focus on teaching and learning and outline the explicit learning objectives of the CPD.

Methods of identifying studies for the systematic map and in-depth review comprised:

• a systematic search of the literature, using electronic databases, handsearching key journals, word of mouth, citations and websites
• the application of a set of initial inclusion criteria to the titles and abstracts thus uncovered
• retrieval of full reports, to which the criteria were re-applied to see if they were suitable for inclusion in the mapping stage of the review
• keywording all the included reports with EPPI Centre core keywords, such as type of study, type of setting, age and curriculum focus, as well as a number of review-specific keywords to distinguish finer detail between types of intervention, teachers and processes
• the application of a second, narrower set of inclusion criteria to the keyworded reports, to ensure that only studies which contain data about the impact of the CPD on pupils were retained for in-depth review
• using EPPI Centre tools and data-extraction software to extract data from the studies and to assess the weight of evidence they helped to provide for answering the review-specific question.

Results

The process of identifying studies for the second review revealed a number of studies of collaborative CPD that had not been identified in the first review, most of which had been published since the completion of the first review. These studies were included in the systematic map and, where applicable, in the in-depth review, thus enabling us to build on the findings of the first review.

Mapping of all included studies

We sifted 5,505 titles and abstracts systematically, reviewed 223 full text reports and identified 81 studies as relevant to the review, and keyworded them in order to create a systematic map of the literature. Of these, 26 studies were considered to focus on individually oriented CPD, while 55 studies focused on collaborative CPD. These 55 studies were in addition to 72 studies that were identified in the systematic map of the first review.

In-depth review process

Seventeen studies met a second set of inclusion criteria and data was extracted from them. Fourteen of these were studies of collaborative CPD and three were studies of individually oriented CPD. The 14 studies of collaborative CPD were in addition to those (N = 17) identified in the first review.

In-depth review findings

As in the first review, the majority of studies reviewed came from the USA (N = 10). Three were from Canada and one from New Zealand. One study took place in each of the UK, Taiwan and China. The educational settings in which the studies took place were predominantly primary (N = 13) followed by secondary (N = 8). Many of the studies took place in more than one educational setting. In terms of curriculum context, cross-curricular studies featured the most strongly (N = 5), with literacy (first languages) the next most common (N = 3), followed by mathematics (N = 2).

Of the 17 data-extracted studies, 14 were coded as collaborative and three as individual. Eleven studies were researcher-manipulated evaluations (nine collaborative and two individual), and six studies were coded as evaluations of naturally occurring interventions (five collaborative and one individual).

Weight of evidence

Of the 17 studies that met the inclusion criteria for the in-depth review, two were judged to have low/medium weight of evidence (WoE) in relation to whether the study findings could be trusted in answering the study questions and one study was judged to have low weight of evidence in relation to its relevance to the review question. Therefore, these three studies were data-extracted, but as a consequence of their low WoE, were not included in the synthesis. Two studies were found to have high WOE and the rest were assessed as medium.

Overall, there was a varying amount of detail about the sample in some of the studies, and a disappointing lack of detail, in some cases, about the CPD processes. The study results should be considered in the light of this and of the small number of studies which were retrieved, and the fact that a number of these were small scale.

Conclusions

Individual and collaborative CPD

Overall, the findings were consistent with those from the studies in the first review in relation to the effectiveness of collaborative CPD in bringing about changes in teaching and learning. By contrast, the studies of individually oriented CPD - both in number (only three) and in the relatively low degree of pupil impact (two) - offered only weak evidence of their capacity to influence teacher or pupil change. There was also a suggestion in one of the individually oriented studies, based on their discussion of the literature, that it was, specifically, the absence of applied peer collaboration in the CPD design which was the critical factor in the failure of the programme to achieve its stated aims. When discussing the findings of the collaborative studies, it is the findings of the studies identified in the second review that are described below. The text highlights where these are similar or different from the findings in the first review.

The findings are discussed in more detail below.

Outcomes from studies of individually oriented CPD

PUPILS: Two of the three individual studies found some evidence of modest impact as a result of the CPD intervention. This was focused on behaviours and attitudes rather than learning outcomes. None of the three studies attempted to measure gains in pupil achievement as a result of the attitudinal and behaviour changes they found.

TEACHERS: While two studies found some evidence of changes in teachers' practice and beliefs, one found minimal impact on teacher efficacy, either personal or general.

Outcomes from studies of collaborative CPD

PUPILS: Ten of the 11 collaborative studies were reported to have found some evidence of improvement in pupil learning, accompanied in seven cases by positive changes in either pupil behaviour or their attitudes or both. This is consistent with the patterns of impact in the first review.

TEACHERS: All the studies found links between the CPD and changes in teacher practice, attitudes or beliefs. As in the first review, there was evidence (in six studies) that changes in teachers' classroom behaviours were accompanied by positive changes in attitude to their professional development.

CPD processes and characteristics

Part of the intention of this review was to explore in more detail the characteristics of CPD for which there is medium to high evidence of positive teaching and learning outcomes. The themes and clusters below were identified in the first review and were applied to the collaborative and individually oriented studies identified in the second review.

Themes and clusters

The studies included in the synthesis focused on CPD interventions which displayed similar patterns of activity to those in the first review of collaborative CPD:

• the use of external expertise linked to school-based activity
• observation and reflection (often based on observation)
• an emphasis on peer support, acknowledging individual teachers' starting points and factoring in processes to encourage, extend and structure professional dialogue
• scope for teacher participants to identify their own CPD focus
• processes for sustaining the CPD over time to enable teachers to embed the practices in their own classroom settings (see section entitled 'How long did it last?' under 'Nature of the collaboration', p 8 of the technical report) .

The use of external expertise linked to school-based activity

Individually oriented CPD

All three included studies of individually oriented CPD involved inputs from specialists (the researchers themselves).

Collaborative CPD

As in the studies included in the first review, the extent and nature of the partnerships between 'experts' and teachers varied. All the interventions involved the use of specialist expertise initially. This ranged from an initial two-week 'instructional institute' to a two-and-a-half day in-service session. Following these intense initial inputs, the external/expert input was also sustained throughout the life of the intervention in all but one of the collaborative studies. We had wondered whether a comparison of individual CPD and collaborative CPD might show how far collaboration is able to promote an effective, cheaper, more accessible and closer-to-school alternative to external input. In the event, we found little evidence that the inputs from the external experts were any less intensive or sustained in the collaborative than the individually oriented CPD. However, we do not have enough studies of individually oriented CPD to make comparisons about the relative detailed external inputs meaningful. On the evidence from the collaborative studies alone, as in the first review, it appears to be a combination of external expertise and peer support which delivers the desired outcomes of the collaborative CPD. However, in the absence of a detailed comparison of the individual interventions, it is not possible to assess the relative weight of any of these characteristics in isolation.

Observation and reflection

Individually oriented CPD

Only one of the studies of individually oriented CPD involved observing teachers in their classrooms as they attempted to put new knowledge and skills to work.

Collaborative

Like the studies in the first review, observation featured in all the collaborative second review studies. Observation (together with a range of other methods) was the principal means of data collection across the collaborative studies. It was unclear, in three of the studies, whether observation was purely for data-collection purposes or not. In two of the studies, video was used as the principal means of observation. In ten of the studies, it was evident that the observations were used formatively (followed by feedback and discussion), mostly in combination with data collection. In one study, the researchers used observation purely for data-collection purposes.

An emphasis on peer support

Collaborative only

In all the collaborative studies, peer support was a feature of the effective CPD and, in seven of these, peer collaboration was the principal vehicle for professional development. Eight studies, including one of the high WoE studies, were explicit that peer support involved peer observation. This follows the pattern established in the first review.

Since our definition of collaborative CPD specified collaboration between teachers, but not necessarily teachers in the same schools, two studies were included which used CPD peer-support models involving previously trained teachers. In one high WoE study, some teachers were supported within their own schools and some were supported by teachers from other schools. This led the researcher to question whether there was a direct link between the teacher effects he found and the extent and nature of the peer-support teachers enjoyed during the course of the intervention. Three of the nine teachers in the experimental group did not progress as well as the others. Two of these had little or no contact with their peer support teachers. The teachers who made the strongest progress were either peer taught by department heads or had peer (i.e. previously trained) teachers in their own schools. Hence the extent of peer support experienced by the participating teachers, and whether it was in their own schools or not, may have been directly linked to the variations in impact on teacher behaviours. Another study suggested that the absence of peer collaboration among teachers might have been a weakness in the CPD design which helped to explain the disappointing results.

Scope for teacher participants to identify their own CPD focus

Individual

The three individual studies were aimed variously at using SMILE (Science and Math Integrated with Literary Experiences) to develop mathematics learning, changing teacher beliefs about democracy and developing teacher efficacy beliefs. Hence the precise focus of the intervention was determined prior to the start of the CPD, which did not allow scope for teachers to focus on a curriculum area or issue of their own selection.

Collaborative

Of the 11 collaborative studies, as in the first review, the majority (seven) were constructed to give teachers choice within a broad area of curriculum or pedagogy.

Nature of the collaboration

In this review, we wanted to explore the nature of collaboration in CPD more closely than we had been able to do in the first review. As an initial framework, we identified a number of further facets of collaboration which appeared to us from the evidence in the data-extracted studies to be worthy of closer exploration. We used these to ask the following questions of the collaborative studies.

Was the collaboration between teachers and between teachers and experts off-site or in the teachers' own classrooms?

We were interested to note that nine of the collaborative studies in this review associated with positive teaching and learning outcomes involved CPD activities which took place either exclusively in the teachers' own classrooms or in combination with off-site meetings or in-service sessions. In the individual studies, two of the CPD programmes took place off-site and it is unclear from the descriptions in the third the extent of off-site and within-school activities. It appears that CPD based in the learning teachers' classrooms may be linked to positive pupil and teacher outcomes.

Did the collaboration involve experimenting with and adapting/improving different teaching approaches or was it purely reflective/discursive or a combination of both?

In seven studies, including the high WoE study, the collaboration involved experimenting with new approaches. Although only two of the studies were specifically keyworded as action-research based, the majority of the collaborative studies were, in fact, focused on collaborative activities that involved experimenting with new approaches. Teachers were supported in building on what is known, refining approaches, reflecting on evidence of impact, and refining their plans in the light of this.

How long did it last?

All the studies had to involve CPD which was sustained over at least 12 weeks or one term. We could find no clear links between any additional length of time for which the CPD was sustained and the degree of impact in relation to teaching and learning. When the intervention was sustained for longer than one term, this did not necessarily mean more impact. This suggests that, even where CPD is sustained over time, it may be the nature of the collaboration between the participating teachers which is critical to the degree of impact on teaching and learning. This is a proposition which we will explore in a third review.

Did it involve groups of teachers, pairs of teachers or other combinations?

In six of the collaborative studies, teachers worked in pairs, although there were opportunities to work in larger groups. It may be that paired or small-group work is a more effective model of collaboration than larger discussion groups.

Was it voluntary?

In all but three of the collaborative studies, teachers were voluntary participants in the collaboration. In the large scale QUILT project (Appalachia, 1994) involving 1,178 teachers from 42 schools in 13 districts, participation was mandatory. This study found that the experimental group, where peers coached each other over a year (compared with two other groups which received initial training but no peer support) changed its practice significantly with demonstrable improvements in pupil learning. The researchers suggested that the long-term 'opportunities for demonstration, practice and feedback' were the key to changing their 'deeply entrenched behaviours'. Hence, although the majority of participation was voluntary, it appears that effective collaboration and peer support may be a powerful way of achieving 'buy-in' from participating teachers.

The answers to these questions have led us to some tentative propositions which will be further tested in the third review. These are as follows:

• Within school, classroom-based CPD may be more effective than off-site CPD even if the latter involves teachers working together.
• Collaboration between teachers which is focused around active experimentation may be more effective in changing practice than reflection and discussion about practice.
• Collaboration may be an effective vehicle for securing teacher commitment and ownership of CPD in cases where it is not possible for the teachers to select a CPD focus of their choice.
• Paired or small-group collaboration may have a greater impact on CPD outcomes than larger groups.

Strengths and limitations of the review

Strengths

A strength of this review, as with the first, is the involvement of a number of user groups in setting and refining the questions, and interpreting and disseminating the findings.

The CPD Review Group believes that it can build on both the findings and experiences of the first and second reviews, specifically as follows:

• The Review Group is at a stage where it can move towards developing a taxonomy of collaboration and make this meaningful and applicable to practitioners and policy-makers.
• The reviews provide the basis from which to continue to unpack the specific processes involved in the CPD intervention and identify those which appear to influence change in teacher practice.
• Due to the significant number of studies which did not pass the criterion of having student impact data, we now intend to look at studies with robust teacher impact data only (i.e. no student data) as the basis for the third review.
• The Review Group is interested in the effect and influence which external and specialist expertise brings to the design and impact of CPD processes.

Limitations

Although we had designed a set of review-specific keywords to enable us to typify and quantify the processes and activities involved in the CPD interventions, we found, with hindsight, that they were unworkable. Consequently, we had to return to the studies themselves to extract these data and we have not mapped them.

The limitations of the studies themselves followed a similar pattern to that which we found in the first review. In particular, we noted the following:

• a varying amount of detail about the sample in some of the studies, with some reviewers noting that they would have liked to have been given more detail about the sample background(s) in order to make the connections between contexts
• a lack of detail, and in some cases, clarity, of the different aims and foci of the studies
• with the overwhelming majority of studies being conducted in the USA, an uncertainty about whether the findings could also apply in other countries
• the possibility of additional fruitful data in a number of PhD theses and other studies not retrieved within our timescale, containing unexplored data
• a relative lack of detail, in some cases, about the CPD processes
• a lack of discussion, in some studies, of the effect on the evidence of using the researchers as part of the CPD intervention
• the small number of studies which were retrieved and the fact that a number of these were small scale

Implications for policy

The Review Group consulted different policy stakeholders in the UK to help identify the main issues highlighted by the review which had implications for policy-makers involved in the following:

• school leadership
• local and national government
• supporting teachers' professional development
• professional and subject representation.

Consultation centred on a seminar at which participants' discussions were based on a detailed summary of the review process and findings together with the implications identified from a similar collaboration for the first CPD review. The Review Group was concerned to recognise that policy-makers themselves were best placed to identify the implications for policy making. Since there are few policy-makers on the Review Group, we have used the points made at the seminar and, as far as possible, the voice of other policy-makers consulted, to report on the implications for this group.

The organisations represented in this process were as follows:

Department for Education and Skills (DfES), Teacher Training Agency (TTA), Qualifications and Curriculum Authority (QCA), Centre for British Teachers (CfBT), General Teaching Council (GTC), Specialist Schools Trust National Educational Research Forum (NERF).

Individually oriented CPD

Policy-makers were struck by the following:

• the paucity of the evidence about the impact of individually oriented CPD
• the weak evidence of impact uncovered by studies that did address this type of CPD
• the comparison between this evidence and the strength of evidence about collaborative CPD that has been uncovered

They suggested that policy-makers involved with learning and teaching and/or planning CPD opportunities should encourage and/or require providers and facilitators to consider:

• whether collaboration or structured peer support can be built into development strategies; or
• how to encourage and enable schools and/or teachers to develop collaborative opportunities/structured peer support to complement and help embed the contribution from specialist expertise.

The focus of professional learning

Policy-makers noted the importance of identifying a focus for professional learning that addresses teachers' concerns about their pupils' learning, and their current interests. They also noted that this was refined and interpreted through peer support, resulting in teacher ownership, whether or not participation was voluntary. The implications of this for policy-makers include a need to:

• recognise that debates about whether CPD should be voluntary are over-polarised and that ownership can emerge from collaborative interpretation of externally framed needs over time; and
• ensure that the diagnostic contribution of performance management to identifying learning needs is introduced in ways that enable teachers to work together to refine and select potential development strategies and contexts. CPD participants also need the capacity to develop further  the learning focus in the light of:
  • pupils' responses; and
  • sustained professional dialogue about both the strategies and pupils' responses.

Specialist expertise

The consistency was noted between the findings in the first and second reviews about the importance of specialist, external input in relation to the following:

• an aspect of pedagogy
• supporting adult learning
• working flexibly in response to the imperatives of school life.

Policy-makers were concerned that, where schools have experienced poor specialist support in one or more of these respects, there is a risk that all such support will be dismissed as unreliable or too costly. There are a number of current policy initiatives that incorporate specialist expertise closely related to this model, including the following:

• the consultant leaders programme (Primary National Strategy (PNS) and the National College for School Leadership (NCSL))
• the development of the role of Advanced Skills Teachers (ASTs)
• the work of the consultants for the PNS and the Key Stage 3 strategy.

The need for specialist input and its relation to peer support is relevant to these and other policy programmes. The review findings could be used to reinforce and/or refine the forms of support used in such programmes.

There was concern that schools could find it difficult to identify appropriate external expertise. Specialists in turn may find it difficult to identify cost-effective ways of working flexibly with individual schools. The changes in local education authority (LEA) and higher education institution (HEI) funding and roles are also thought to be making it more difficult to access and organise such resources and to manage succession planning.

Policy-makers are urged to consider the nature of specialist input needed, the potential sources of such expertise and the ways in which access can be facilitated and sustained.

Alternative forms of accountability

Looking across the two reviews, policy-makers were impressed by the extent to which participating teachers expressed a desire not to let each other or their students down. While recognising that the participating teachers were motivated more by improving their practice than by accreditation, policy-makers were keen to draw attention to the possibility of taking such approaches into account within existing and developing accreditation schemes for CPD.

Policy-makers are urged to consider building teachers' reciprocal accountability for professional learning and the links between professional learning and concerns about specific students or groups of students into evaluations of the effectiveness of CPD and systems for recognising and/or accrediting such work.

Time

Policy-makers noted the findings in relation to time. Collaborative CPD with a positive impact lasted at least one term but further extensions of the work did not necessarily result in benefits. The Review Group is therefore urged to explore, in future reviews, how this relates to the scale of the learning goal and the stage of development of participants.

In the meantime, policy-makers and CPD providers working in areas where extended programmes are the norm are encouraged to review progress at the end of a first term to ensure that goals are refined so that they remain sufficiently challenging to justify the cost and opportunity costs of a continued process.

The nature of collaboration

Policy-makers were also interested in the additional practical detail about the nature of collaboration and the importance of issues featured in the first and second review in relation to the roles of questions, structured dialogue, surfacing beliefs, experimentation and building shared interpretations.

Policy-makers working across the range of national education strategies are encouraged to consider how far programmes plan for, provoke and support such dialogue between professional learners on a sustained basis. They should consider using this framework to provide a more detailed scaffolding to the well-established 'plan, work with evidence and review' cycles. The importance of ensuring that all sustained CPD involves an element of planned experimentation and planned collaboration connected directly to the teachers' own classroom should also be taken into account in designing initiatives targeted at developing learning and teaching.

Implications for practice

For practitioners we used a different approach to identifying implications. This was partly on account of the practicalities involved and partly because recently retired practitioners had been active in the review process and were involved in identifying implications. In addition to working with these colleagues, CUREE was also involved in a range of consultation exercises and seminars in England relating to CPD during September and October 2004, which enabled the development of a detailed and up-to-date picture of current CPD practice in schools and LEAs. This discussion was used to identify potential hot spots where the review evidence was likely to connect with, or inform, practitioner concerns. Part of the consultations and seminars involved discussion with the participants about the implications from both this review and the first review of CPD. Participants noted that the picture is a complex one and is often dependent on the organisational and working contexts of the CPD interventions and programmes.

Impact

There is evidence that collaborative CPD of the kinds identified in these research reports is effective in bringing about development in teaching and learning.

CPD co-ordinators: You should consider whether CPD programmes involve regular, structured opportunities for collaboration.

Teachers: You should consider seeking more opportunities to collaborate.

Combined expertise

Combining external expertise with peer support appears to be a consistent feature in delivering the desired outcomes of collaborative CPD.

Both teachers and CPD co-ordinators: You should consider how you can integrate learning from external specialist expertise with in-school learning.

Collaboration

Peer support is a key feature of effective collaborative CPD and peer collaboration often acts as the principal vehicle for professional development. It is possible that lack of collaboration might be a significant factor in CPD programmes that do not have long-term impact.

Teachers: If CPD is oriented towards participants as individuals, you may want to maximise your opportunities for peer support by developing partnerships with other teachers and setting time aside for shared planning or talking together about shared experiences. You could also consider how you can follow up individually oriented CPD by acting as a coach for other teachers.

CPD co-ordinators: You may want to consider how best to develop a critical mass of coaching skills among school practitioners.

Securing commitment

Collaboration may be an effective vehicle for securing teacher commitment and ownership of CPD where the agenda has been set by others.

Teachers: On those occasions when you are participating in CPD where the agenda is imposed, you could consider taking time with a colleague to interpret the CPD framework and themes explored in the CPD in the context of your own pupils, knowledge and skills. Think too about how you could integrate generic themes with your own concerns. For example, you could explore an emphasis on assessment in the context of the needs of a specific group of pupils or a specific subject; alternatively, you could take forward CPD activities with a very specific focus, for example on improving mathematics, through a more generic teaching and learning focus, such as thinking skills.

Locating CPD in classrooms

CPD based in the learning teachers' classrooms may be linked to positive pupil and teacher outcomes.

 CPD co-ordinators: You might consider how to use teachers' classrooms as a base for CPD activities and this might be timetabled so that a range of classes and settings are used. You need to consider building time and other resources into the CPD programme, rather than adding to teachers' existing workload.

Experimentation

Collaboration between teachers, based on active experimenting, may be more effective in changing practice than reflection and discussion about existing practice.

CPD co-ordinators: You could encourage groups of teachers to choose a shared focus for experimentation in their classrooms. In this way, they could offer each other support and reflect together on their experiences, and include colleagues who may not have attended conferences where ideas were first presented.

Peer support/peer coaching

Peer support/peer coaching may be a cost-effective way of extending the reach of external specialists into day-to-day school life. Coaching is emphasised and supported in many national programmes, such as the primary and Key Stage 3 strategies.

Teachers: You should consider seeking opportunities to participate in peer-coaching programmes to acquire generic coaching skills while at the same time pursuing personal CPD priorities as agreed, for example, through performance management or individual CPD planning.

Pairs and groups

Paired or small-group collaboration may have a greater impact on CPD outcomes than larger groups. Small groups may be able to meet more regularly to reflect on their CPD than a larger group would be able to.

Teachers: You could discuss with CPD co-ordinators and/or course providers the possibility of working in smaller groups or pairs when finding yourselves in large groups. You should also consider asking whether you could work with a colleague whenever they are offered CPD opportunities.

CPD co-ordinators: You could consider initially setting up small groups or pairs to undertake the CPD together.

Implications for research

Our priority has been to work on implications for practitioners and policy-makers. The 'implications for research' below were developed following presentation at the British Education Research Association Conference (September 2004) and in consultation with our academic colleagues on the Review Group.

• Researchers need to report, at least in brief, information about the context and process of the CPD intervention, including the sample characteristics, recruitment strategies and details of the methodology.
• Research on different forms of CPD is a fertile area of study, more so given the current policy direction and the work within the Strategies in England and the broad international consensus of the importance of collaboration and networking.
• Research is needed that looks at individually oriented forms of CPD.
• People engaging in research about CPD who wish their work to be considered in systematic reviews, need to consider ways in which their reporting facilitates or inhibits inclusion in systematic reviews, within their own research models and frameworks.
• There is a need for much greater clarity in providing clear titles and abstracts for studies that accurately reflect the content of the papers in order to enable search enquiries to identify relevant materials.
• Researchers need to explore the organisational context(s), including for example the contribution of school and CPD leaders, when reporting studies of the impact of CPD, in order for others to make connections.
• When reporting research, researchers should consider both the CPD processes and outcomes to ensure that practitioners know both whether and how an intervention is effective.
• Researchers need to explore the literature about both the pedagogic interventions they are targeting and the literature about CPD in securing an evidence base for their research. We found that most studies tended to include a literature review on only one area rather than both.
• Researchers are professional learners too and need to consider working collaboratively with other researchers and with practitioners in schools in terms of designing and implementing the research, and developing a sense of ownership in the research by practitioners.
• Journal editors need to consider all these issues when selecting articles for inclusion, while continuing to work within their own frameworks and models.

References

Cordingley P, Bell M (2002) Literature and evidence search: teachers’ use of research and evidence as they learn to teach and improve their teaching. London: TTA. (Unpublished)

Day C (1999) Developing Teachers: The challenges of lifelong learning. London: Falmer Press.

Harland J, Kinder K (1997) Teachers’ continuing professional development: framing a model of outcomes. British Journal of In-Service Education 23: 71-84.

Hargreaves DH (1993) A common-sense model of the professional development of teachers. In: Elliott J (ed) Reconstructing Teacher Education: Teacher development. London: Falmer.

Joyce B, Showers B (1988) Student Achievement through Staff Development. London: Longman.

Joyce B, Showers B (2002) Student Achievement through Staff Development. 3rd ed. London: Longman.

Rich I (1993) Stability and change in teacher expertise. Teachers and Learning 9: 137-141.

This report should be cited as: Cordingley P, Bell M, Thomason S, Firth A (2005) The impact of collaborative continuing professional development (CPD) on classroom teaching and learning. Review: How do collaborative and sustained CPD and sustained but not collaborative CPD affect teaching and learning? In: Research Evidence in Education Library. London: EPPI Centre, Social Science Research Unit, Institute of Education, University of London.