Researching Effective Pedagogy in the Early Years

Figure 34: Summary: amount of pedagogical interactions made by practitioners with 96 different levels of training Figure 35: Cognitive pedagogical interactions of staff with different qualifications 96 Figure 36: Social Pedagogical interactions of staff with different qualifications 97 List of tables Table 1: Questioning 55 Table 2: Children’s most common learning activities while experiencing different curricular areas 66 Table 3: ICT provision 86 Table 4: Children’s most common learning activity when in the presence of staff 97 with qualifications List of appendices Appendix 1: Manager interview schedule 129 Appendix 2: Staff interview schedule 133 Appendix 3: Childminder interview schedule 139 Appendix 4: Parent interview 141 Appendix 5: Target Child Observation 142 Appendix 6: Episode Analysis of the Target Child Observations 146 Appendix 7: Cognitive challenge 148 Appendix 8: Examples from the Target Child Observations (TCO) of children learning through child initiated play 149 Appendix 9: ICT subscale based on the ECERS-E 155 Acknowledgements: Thanks are due to the EPPE Team researchers, especially Brenda Taggart and Bet McCallum, and the Principal Investigators for sharing data from the project. We also appreciated and valued the support provided by the Steering Committee members of the Effective Pedagogy in the Early Years project, and in particular their comments and feedback on presentations and reports. We are grateful and indebted to the staff, parents and children from our 14 research settings for their co-operation and support. Thanks are also due to Angela Hobsbaum, Viv Moriarty and John Siraj-Blatchford for comments on parts of this report, and to Rebecca Goldman for her critical reading of earlier versions. Note: The work was commissioned by the then DfEE, which became DfES in the course of the research. Both DfEE and DfES are used in the report, since some early research 'instruments' of necessity incorporated the DfEE logo. 5 Glossary Adult but child ex: (Adult initiates but child extends) is an observation category where the adult has initiated an activity but a child has extended it. Adult throughout: is an observation category where the adult has initiated an activity and led it throughout. As expected: (learning as expected) refers to a description of the target child’s ability by their practitioner; i.e. the target child was deemed to be learning as expected. CC: Combined centre; where care and education are integrated with some teacher input. CGFS: Curriculum guidance for the foundation stage (QCA, 2000) for children from 3 years old to the end of reception year. Child but adult ex: (Child initiates but adult extends) is an observation category where the child has initiated an activity but an adult has extended it. Child throughout: is an observation category where the child has initiated the activity and led it throughout. Co-construction: In the traditional constructivist account of learning new understandings are considered to be founded upon the child's prior understandings. In our analysis we emphasise the importance of recognising that learning encounters involve processes of co-construction where each party engages with the understanding of the other. A necessary condition for this to occur is that both parties are involved and that the content should be instructive (See below). Conceptual knowledge: Refers to the ideas and understanding of principles and of their relationship to each other. Creative: Refers to the Creative development area of the CGFS. Critical moments: An event occurring during an observation which influenced and developed an idea or skill substantially; this may have come from a child, a practitioner, or a resource. Curriculum differentiation: The provision of differentiated curriculum experiences as a response to the identification of the specific needs of individual or small groups of children (See ‘formative assessment’). DC: Local Authority Day Care centre, normally with a Social Services history but most are now managed by local education authorities. DfES: Department for Education and Skills, the UK central government department who commissioned the EPPE and EPEY research projects. Direct teaching: These are pedagogical interactions referred to by the Target Child Observations which include simple questioning, description of the activity, didactic instruction, task management, reading to the target child, and organising and allocating tasks. ECERS-E: (Early Childhood Environment Rating Scale: Extension) Sylva, Siraj-Blatchford and Taggart (1998) University of London, Institute of Education: A rating scale of 4 subscales which assesses pedagogy and the curriculum within the setting, including the areas of mathematics, science, literacy and diversity (whether staff plan to meet particular / individual needs). 6 ECERS-R (Early Childhood Environment Rating Scale: Revised) Harms et al (1998): A rating scale consisting of seven sub-scales which provide an overview of the pre-school environment, covering aspects of the setting from furnishing to individuality of care and the quality of social interactions. EEC: Early Excellence Centre (as designated by the DfES); a centre which offers fully integrated care and education for pre-school children with other services for older children, families and practitioners. EPPE: The Effective Provision for Pre-school Education project, a ‘value-added’ longitudinal research study of the effects of early childhood provision on the developmental progress of 3000+ children. EPEY: The Effective Pedagogy in the Early Years project is the formal name of the research reported here. Excellent settings: The group of 3 case-study settings (identified by EPPE) whose social or/and cognitive outcomes were excellent; i.e. where the children were found to make substantially more developmental progress than could be predicted by their family background and their pre-test assessments (see ‘good settings’). Formative assessment: The process whereby information is collected to identify the extent of children’s learning, and subsequently applied in the provision of feedback and in adapting the curriculum and pedagogy to provide for their particular needs. Foundation Stage: a Stage of education for children from 3 years old to the end of reception class (see CGFS). Good settings: The group of 9 (7 settings for the systematic observations) case-study settings whose outcomes were found to be good in some areas of social or cognitive development, but where the children largely made predictable progress according to their family background and their pre-test assessments (i.e. not ‘excellent’ see above). High cognitive challenge: An overall qualitative judgement of how much the episode is ‘stretching’ the child’s mind (see Appendix 7). Intervals: The 30 second intervals which make up each 20 minute Target Child Observation. Instruction/Instructive: The term ‘instruction’ has been applied to include both direct (demonstrating, questioning etc) and indirect instructional behaviours and intentions such as the encouragement of parental involvement and the provision of ‘pedagogical framing’. Involvement: The concept of involvement that we have applied has been adapted from the work of Ferre Laevers (1994) and Pascal and Bertram (1997). The term is applied to adults and children who have focused their attention upon a shared activity and are persistent. They are ‘intrinsically motivated, rarely distracted, and appear fascinated and absorbed’ (op cit) by their shared activity. Iterative triangulation: A mixed methodology design was employed in the study and a process of iterative triangulation is described. This was employed to contribute to the internal validity of the study and to encourage a process of working; ‘back and forth between inductive and deductive models of thinking’. Knowledge: Refers to the Knowledge and Understanding of the World area of the CGFS. Learning episodes: Coherent activities initiated either by children or adults. 7 Level 2-4: Practitioner qualification levels including NVQ level 2-4 and those which are equivalent to it. Level 5: Practitioner with NVQ level 5 qualification or qualified teaching status. Literacy: Refers to the Communication, language and literacy area of the CGFS. Modelling: The process where early years educators provide a ‘model’ in terms of their language, behaviours, skills and/or attitudes for young children to imitate. Maths: Refers to the Mathematical development area of the CGFS. Monitoring: These are pedagogical interactions referred to by the Target Child Observations which include the practitioner observing the target child, and when the practitioner was available to the target child in their social context but not interacting with them. Nodes: act as ‘containers’ in NVivo for categories and coding of the qualitative data e.g. in the coding of a transcript of an interview. NC: Nursery Class, normally attached to a primary school and where children aged 3 and 4 attend half-day nursery education. NS: Nursery School, whole schools catering for children 3-5 years old. Outcomes: In this report they are referred to in respect to how much developmental progress the children in the EPPE case-study settings made which cannot be explained by background characteristics. PDN: Private Day Nursery. Pair: refers to a child pair, rather than an adult and child which is 1:1 (see the last entry in the glossary). Pedagogy: Pedagogy: The practice (or the art, the science or the craft) of teaching. See page 27 for a more detailed discussion. Pedagogical content knowledge: Different pedagogic techniques are often required to make different forms of knowledge, skill and understanding accessible to young children. In teacher education the identification of appropriate strategies is often referred to as ‘pedagogical content knowledge’. Requires knowledge of the ‘subject’ being taught and the child’s level of learning. Pedagogical interactions: Face to face interactions practitioners engage in with children; they may take the form of cognitive or social interactions (see figure 5). Pedagogical framing: Is the ‘behind-the-scenes’ work that practitioners do with regards to provision of materials, arrangement of space, and the establishment of daily routines to support learning through (e.gs.) exploration, cooperation and the equitable use of resources. Pedagogical strategies: practices which support learning, for instance, social interactions, assessment, the organisation of resources or management. PG: Playgroup; a form of pre-school provision for children (2 ½ years and above), historically associated with the voluntary sector. Usually run by parents. Physical Development: Refers to the Physical development area of the CGFS. Procedural knowledge: Knowing about the things in the world, and how to act upon them. 8 PSE: Refers to the Personal, social and emotional area of the CGFS. QSR N-Vivo: computer software programme used to code and analyse qualitative data using a node (see glossary above) structure to support the organisation, categorisation, shaping, and linking of the data. Reception/RC: refers to the 2 Reception classes which were part of the case-study sample. In the UK, more than 90% of children enter the Reception class of a Primary (Elementary) School aged 4. Setting: the term refers to local authority day nurseries, playgroups, nursery classes, nursery schools, combined centres, reception classes, early excellence centres and private day nurseries. Small group: Code from Target Child Observation system referring to a group of between 3-8 children. Struggling: refers to a description of the target child’s ability by their practitioner; where the target child was deemed to be a struggling learner. Sustained shared thinking: An episode in which two or more individuals “work together” in an intellectual way to solve a problem, clarify a concept, evaluate activities, extend a narrative etc. Both parties must contribute to the thinking and it must develop and extend. Systematic observations: Analysis from the timed Target Child Observations. V capable: refers to a description of the target child’s ability by their practitioner; where the target child was deemed to be a very capable learner. Target Child: focal child being observed during the timed Target Child Observations. TCO: Target Child Observation instrument (Sylva, Roy and Painter,1980). A timed systematic observation instrument used to look at children’s experiences of the classroom. Whole class: Code from Target Child Observation system referring to a group of 9 or more children. ZAD (zone of actual development): Refers to the cognitive levels at which a child is currently operating. ZPD (zone of proximal development): Refers to the higher cognitive levels which a child can achieve when supported by a practitioner or more knowledgeable other (the cognitive potential). 1:1: One to one referring to a one adult and one child. 9 Introduction This report presents the analysis of the Effective Pedagogy in the Early Years (EPEY) Project. A condensed account of our conclusions is included in the Executive Summary below. In the interests of clarity and ease of presentation the main body of the report begins with detail of the methodology that was employed in the study and by a critical review of the literature with an account of the questions that were addressed. This is followed by an extended discussion and analysis of our findings. It should be understood that a good deal of the literature review, and the elaboration of our research questions and arguments were constructed in the process of our data collection and analysis. The linear structure of the presentation should therefore not be taken to represent the chronological sequence that the study followed. In particular it should not be assumed that all of our questions arose directly from our review of the literature. The process was rather an iterative one where our field observations stimulated an engagement with the literature just as often as the reverse. The project is based on research into centres for which we have robust child developmental outcome data provided by the 5 year longitudinal Effective Provision for Pre-school Education (EPPE) Project (DfES funded). All of the centres that we base our evidence on in this report can be described as providers of good practice to children and families. Of the 14 settings some are excellent and highly effective, most have good to excellent child outcomes across the board, none have an ineffective profile on the social and behavioural outcomes in EPPE. A minority have 1 or 2 low outcomes on the four reported cognitive assessments in EPPE. Two of these 14 case study settings are reception classes, selected for the EPEY study and chosen on the basis of professional judgement as providers of ‘good’ practice. The study engages with staff pedagogy in all types of Foundation Stage settings and the impact on children’s learning as a consequence of this pedagogy. Childminding is an important provision, and so we attempted to explore the practice of childminders in this study. However we had no child outcome data for these providers; they were selected, like the two reception classes, on the basis of local professional judgement as effective providers. Childminders’ perceptions of their practice and the impact of the Curriculum guidance for the foundation stage (QCA, 2000) are thus reported separately at the end of the study on the basis of telephone interviews with 46 childminders from 4 regions in England. 10 Part 1 Executive Summary The Effective Pedagogy in the Early Years (EPEY) study was developed to identify the most effective pedagogical strategies that are applied in the Foundation Stage to support the development of young children’s skills, knowledge and attitudes, and ensure they make a good start at school. The term Pedagogy is applied here to refer to the instructional techniques and strategies which enable learning to take place. It refers to the interactive process between teacher and learner, and it is also applied to include the provision of some aspects of the learning environment (including the concrete learning environment, and the actions of the family and community - see 3.4 below). The EPEY study was largely based on intensive case studies conducted in ‘effective’ Foundation Stage settings reflecting all types of provision. Twelve of the settings were chosen on the basis of child social/behavioural and cognitive outcomes from the Effective Provision of Pre-school Education (EPPE) project as 'good' practice settings. Two reception classes, and 46 ‘effective’ childminders were identified according to the professional judgement of local authority experts from 4 of the 5 EPPE project regions. Careful, detailed case studies were conducted in each setting which involved detailed documentation of naturalistic observations of staff pedagogy, and systematic structured observations of children’s learning. Information was also gathered and analysed using interviews with parents, staff and managers and through intensive and wide ranging documentary analysis and a literature review of pedagogy in the early years. Further information was collected from the 46 childminders using telephone interviews to identify the impact of the Curriculum guidance for the foundation stage (CGFS) on their practice. For most practitioners the declared priorities in the early years are on the development of positive dispositions to learning, self-confidence and independence. Staff and parents normally give priority to social development, but our evidence suggests that those settings which see cognitive and social development as complementary achieve the best profile in terms of child outcomes. Our analysis has also gone a long way to provide explanations for the statistical relationships which were found in the EPPE (Effective Provision of Pre-school Education) project data analysis. Four areas of impact were identified for special attention. Our close analysis of the relevant EPPE research instruments (e.g. the Early Childhood Environment Rating Scale) suggested that we needed to investigate each of the following practices further, to identify how some practitioners supported the children in making greater developmental progress than others in less effective centres: • adult-child verbal interactions; • differentiation and formative assessment; • parental partnership and the home education environment; • discipline and adult support in talking through conflicts. Adult-child verbal interactions If learning comes from a process of cognitive construction that is only achieved when the child is motivated and involved, we have argued that it is entirely consistent to treat the part played by the effective educator in the same way. The cognitive construction in this case is mutual, where each party engages with the understanding of the other and learning is achieved through a process of reflexive 'co-construction'. A necessary condition is that both parties are involved, and, for the resultant learning to be worthwhile, that the content should be in some way instructive. Our analyses of the qualitative and quantitative data have substantiated this model and our research has also shown that adult-child interactions that involve some element of 'sustained shared thinking' or what Bruner has termed ‘joint involvement episodes’ may be especially valuable in terms of children’s learning. We found that the most effective settings encourage ‘sustained shared thinking’ but we also found that this does not happen very frequently. As a group, in excellent settings there were significantly 11 more ‘sustained shared thinking’ interactions occurring between staff and children, than in the ‘good’ settings. When it does occur, it has been shown to extend children’s thinking. Our investigations of adult-child interaction have led us to view that periods of ‘sustained shared thinking’ are a necessary pre-requisite for the most effective early years settings, especially where this is also encouraged in the home through parent support. Parent interview data suggest that in some of our very middle class case study settings (notably the private day nurseries), it is less the staff’s interventions and more the parents’ pro-active behaviour towards their children’s learning in the embedded, cultural context of the home, that has provided a good basis for sustained shared thinking. The analysis has shown that practitioners’ knowledge and understanding of the particular curriculum area that is being addressed are vital as well. A good grasp of the appropriate ‘pedagogical content knowledge’ is a vital component of pedagogy and it is shown to be just as important in the early years as at any later stage of education. The research found that, even in these effective settings, there were examples of inadequate knowledge and understanding of curriculum areas, especially in the teaching of phonics. In the most effective (excellent) settings the importance of staff members extending child-initiated interactions is also clearly identified. In fact, almost half of all of the child-initiated episodes which contained intellectual challenge, included interventions from a staff member to extend the child’s thinking. The evidence also suggests that adult ‘modelling’ is often combined with sustained periods of shared thinking, and that open-ended questioning is also associated with better cognitive achievement. However, open-ended questions made up only 5.1% of the questioning used in even these ‘effective’ settings. In the excellent and good settings the balance of who initiated the activities, staff or child, were very equal, revealing that the pedagogy of these effective settings encourages children to initiate activities as often as the staff. The children in reception classes experienced a different balance of initiated activities, with a much greater emphasis upon staff initiated episodes. We have found that qualified staff in the most effective settings provide children with more experience of academic activities (especially literacy and mathematics) and they encourage children to engage in activities with higher cognitive challenge. While we found that the most highly qualified staff also provided the most direct teaching, we found that they were the most effective in their interactions with the children, using the most sustained shared thinking interactions. Further, we found that less qualified staff are significantly better pedagogues when they are supervised by qualified teachers. Differentiation and formative assessment Our teacher observations suggest an association between curriculum differentiation, formative assessment, and curriculum matching in terms of cognitive challenge, and ‘sustained shared thinking'. The interviews, teacher observations and documentary evidence suggest that the better the setting does on each of these dimensions of good pedagogic practice, the more effective it will be in supporting children’s cognitive progress. The evidence thus confirms the importance of formative assessment to meet children’s particular needs, particularly formative feedback during activities. In the most effective settings, child-related information, especially about curriculum and learning aims, was most frequently shared between parents and staff, and the parents were often involved in decision making about their child’s learning programme. Parental Partnership While the data that we collected have shown a clear association between children’s cognitive and social outcomes and the pedagogic principles that are identified above, the evidence has also shown us that some private settings were effective even where these pedagogic principles were not always applied. Our findings suggest that where a special relationship in terms of shared educational aims has been developed with parents, and pedagogic efforts are made at home to support the children, good developmental outcomes may be achieved even in the absence of 12 consistently good pedagogic practice. However, in more disadvantaged areas, staff in effective settings had to be proactive in influencing and supporting parents’ role in developing the home education environment to support children’s learning. Only supporting the parents’ needs or involving them as helpers did not appear to influence children’s developmental progress. Discipline and adult support in talking through conflicts We believe that the research evidence linking early childhood social and behavioural outcomes with certain pedagogic practices is compelling. We therefore argue that direct teaching should be complemented by dialogues between children and adults which centre on sustained shared thinking. There should be a more equal emphasis on free play and direct learning. We have found that the most effective settings also adopted discipline/behaviour policies in which staff supported children in being assertive, at the same time as rationalising and talking through their conflicts. Staff who were consistent and pro-active in supporting the children in developing their social skills through e.g. using ‘story books and group discussions to work through common conflicts’ had better social/behavioural developmental outcomes. The most effective settings in this respect also had clear discipline and behaviour policies which prioritised talking through conflicts. In settings which are less effective in this respect our observations have shown that there was often no follow up on children's misbehaviour. On many occasions, children were 'distracted' from interfering with other children, or simply instructed to stop. Organisation Three major approaches to early education were identified in a review of the literature: • The teacher-directed, programmed learning approach. • An open framework approach where children are provided with ‘free’ access to a range of instructive learning environments in which adults support children’s learning. • A child-centred approach where the adults aim is to provide a stimulating yet open-ended environment for children to play within. We argue that effective pedagogy in the early years involves a balance of the first two approaches, both the kind of interaction traditionally associated with the term ‘teaching’, and also the provision of instructive learning environments and routines. We argue that where young children have freely chosen to play within an instructive learning environment, adult interventions may be especially effective. However, we have also noted that these interactions are not as frequent as they should be – even in settings we have classified as “effective” on the basis of child outcomes. In all of the case study settings we found that the children spent most of their time in small groups. But our observations show that ‘sustained shared thinking’ was most likely to occur when children were interacting 1:1 with an adult or with a single peer partner. Freely chosen play activities often provide the best opportunities for adults to extend children’s thinking. It may be that extending child-initiated play, coupled with the provision of teacher initiated group work, are the main vehicles for learning. The case study analysis has shown that excellent, and good settings, tend to achieve an equal balance between adult-led and child-initiated interactions. However, in the excellent settings adults extended about half the child-initiated activities, while in the good settings they did this less often. Thus we have found that the most effective settings therefore combine the provision of open- framework, free play opportunities with more focused group work involving some direct instruction. The evidence also suggests that the achievements of settings against the cognitive outcomes appear to be directly related to the quantity and quality of the teacher/adult planned and initiated focused group work that is provided. Our findings suggest that the most effective settings provide both, and achieve a balance between the opportunities provided for children to benefit from teacher initiated group work and the provision of freely chosen yet potentially instructive play activities. 13 The reception classes differed from the pre-school settings in a number of important ways. They had a different curriculum balance, the children spent much more time in larger groups and the learning episodes that we observed were most often teacher initiated. Reception classes had much more literacy, mathematics and physical development and the least personal and social education (PSE) compared to the pre-schools. This may be especially significant as we found that more of a child’s time was devoted to literacy, mathematics and physical development activities in the most effective pre-school centres. Children in the good, but not excellent, case study settings, spent more time in creative and knowledge and understanding of the world activities. Physical development and PSE, however, were given approximately equal attention in all of the settings. The majority of episodes occurring in reception classes were initiated by adults, and the episodes of high intellectual challenge to the child occurred in teacher initiated episodes, a very different approach to that of the pre-school settings. The Curriculum guidance for the foundation stage (CGFS) and transition into school Our research findings support the general approach taken in the Curriculum guidance for the foundation stage (QCA, 2000) where the emphasis is upon an ‘emergent’, ‘cognitively orientated’ approach to learning. Effective practitioners assess the children’s performance to ensure the provision of challenging yet achievable experiences, they model appropriate language, values and practices, they encourage socio-dramatic play, they also praise and encourage, ask questions and interact verbally with the children to encourage sustained shared thinking (SST). Most of our respondents reported that they endorsed the content of the CGFS, but some settings had clearly found the new initiative stressful. In several of the settings where respondents reported little change in their practices, it was felt that the standard of provision that was suggested by the CGFS fell below their prior or current practice. However, it is clear that in many other cases the CGFS is only partially being applied and in most cases this may be simply because the settings current practices are considered more appropriate. Despite these responses it may be that many of the changes brought about by the CGFS are under-reported. In one setting we were informed of significant changes in the staffing responsibilities and in another of a raised awareness of special educational needs. Even where the respondents felt there was little impact upon their curriculum or pedagogy, many reported changes in their assessments. The training programmes providing curriculum knowledge based on the Curriculum guidance for the foundation stage (CGFS) had not yet made a major impact on the practice of the effective childminders in this study, and many of these childminders regard the Early Learning Goals (ELGs) as too formal. The case studies have shown how the curriculum is currently being differentiated according to age. Children who are younger, and those children considered to have learning delay, experienced more PSE and also more creative development; while children who are older (and in Reception) experienced more literacy and mathematics and less PSE. The curriculum was also differentiated between excellent and good settings, with children in the most effective settings experiencing a greater proportion of literacy, maths and physical development activities. This makes the most effective settings more like good Reception Classes in their core curricular balance. The only difference being that excellent settings had the highest proportion of PSE. However, the emphasis in the most effective centres on literacy, mathematics, physical development and PSE left less time for creative activities and knowledge and understanding of the world when compared to other provisions. Our survey shows that the single element of the CGFS considered most popular by practitioners in this study has been its emphasis on continuity into the primary school. A significant degree of pleasure was expressed by some pre-school respondents regarding the positive impact that the CGFS was having on reception classes. The Foundation Stage has clearly improved the relationship between pre-schools and school; but our study shows that there are still major 14 concerns. The issues relate to transition from the nursery to the school reception class and also at the end of reception, from the foundation stage to year 1 of the National Curriculum. For the majority of case study settings, the practice of introducing the children to more formal organisational structures such as carpet and circle time activities (large group activities) was considered of more importance than any specific curriculum work. Most made provision for transition by attempting to make children more familiar with school (through visits etc) and/or by making the nursery practice a little more like school in the latter months (whole class activities, etc). Many of our respondents were critical of the common practice of schools taking one intake in the year rather than taking children into reception classes throughout the year. The extent to which chronological age cannot be taken to indicate young children’s development was also stressed by one of the reception class teachers. Information and Communications Technology (ICT) Generally speaking, pedagogical practice in terms of the ICT curriculum was found to be relatively limited at this time. All but one of the case study settings was found to be equipped with at least one computer that was available for the children’s use every day. Unfortunately the computers were used in a limited way, predominantly in supporting creative development, and by programmes that were chosen to encourage the acquisition of literacy skills. While many respondents recognised the part to be played by cassette recorders, telephones, scanners, and digital cameras we found little evidence of settings using programmable toys, and the integration of work on the computer screen with activities off the screen. The most innovative work that was recorded involved the use of digital cameras and of scanners, but many early years practitioners in this study lacked confidence in the area. While some references were made to the value of the computer in supporting number and language and literacy, and frequent references were made to the children’s on-screen painting, few practitioners were using the equipment to achieve specific CGFS Early Learning Goals. The children tended to work at the computer in groups of 2-4 and the adults tended to help them access software and supported them when they get into difficulties. However, the children spent most of their time at the computer without an adult present. When an adult was present the adult was usually a qualified teacher but in very few cases did we observe extended support. The children were often left to develop computing skills independently, with the adult rarely questioning, instructing, engaging or guiding. Staff rarely engaged in scaffolding or interactions which encouraged sustained shared thinking during children’s computing activities, suggesting a very skills-based and technical approach. Training In the most effective settings the qualified teachers and NVQ level 2-4 staff members spent the greatest proportion of their time working with children on literacy activities. Teachers also devoted a high proportion of their time with children to mathematics activities. Untrained adults spent the majority of their time working with children in the area of creative development. We found in both interviews and observations that trained teachers used the most sophisticated pedagogy, including shared sustained thinking. When less qualified staff were working with qualified teachers we found significantly more sustained shared thinking interactions than when they worked alone or with other less qualified staff. Our study shows that early years staff may need support in developing their pedagogical content knowledge in the domains of the Early Learning Goals. ‘Effective’ Childminders Our findings suggest that the present training provision for childminders may be having a positive effect. It would seem to have provided an increased awareness and understanding of a range of 15 relevant issues and has resulted in increased status, higher levels of security and self-esteem. As a result childminders feel better equipped to develop more professional relationships with parents. Despite these improvements, this study suggests the training programmes geared towards implementing the Curriculum guidance for the foundation stage (CGFS) have yet to make a major impact on the general practice of even these ‘effective’ childminders. Over a quarter (26%) of those surveyed said they had no knowledge of the initiative. This is surprising as the sample was especially chosen to represent those identified as offering higher quality provision. Those who responded positively to the CGFS indicate a higher level of awareness of children’s learning with some evidence of providing higher quality educational provision. However, some childminders regarded the Early Learning Goals (ELGs) as too formal and although they may still be influenced by them, they felt that full implementation was not suitable for their settings. Whatever the level of training received, the childminders in this study felt confident that play (however they defined it) was central to learning for children in the Foundation Stage and this featured as a constant in their daily and weekly routines. Many also felt that children learn best from being supported while pursuing their own interests. Formal and informal networks which offer both personal and professional support were important in the lives of the interviewed childminders irrespective of where they lived. The relationships between the childminders and parents in our survey appear to have been exceptionally good. Although many childminders valued the personal relationships that may develop between childminders and parents there was a clear sense of professionalism in the way that they deal and make contracts with parents. It may be that the formation and development of the National Childminding Association (NCMA) should take some credit in this positive outcome. It would seem that further examination of the interactions between childminders and their children, particularly the extent to which modelling and scaffolding occurs, would be useful in assessing aspects of pedagogical understanding amongst childminders, and this would have implications for future childminder training programmes. Any review that is planned for childminding training programmes might consider input that examines the ways in which the children who are ‘minded’ react and adapt to the childminder setting and the relationships that underpin the childminder/parent contract, especially over the sensitive issues of values, attitudes, expectations and behaviour. There is also a need for more research to study more closely the practices of the full range of childminders. Parents and the home learning environment A total of 107 parents were interviewed from the 14 case study settings and each of the settings encouraged the parents to read with their children. While previous studies have shown that involvement in learning activities at home and parent involvement in decision making, leadership, and governance are often associated with better academic achievement (Epstein 1991), our findings show that it is the involvement in learning activities at home that is more closely associated with better cognitive attainment in the early years. The settings that encouraged continuity of learning between the setting and the home achieved consistently better cognitive outcomes. Whether parents and/or children are tuned into what the setting is trying to achieve seems to be linked, at least partially, with what happens at home. It appears to be associated with what the parents do with their children when they are not at the setting or what the children initiate with their parents. 16 Part 2 Context and methodology The Effective Pedagogy study draws upon the findings of Project EPPE (Effective Provision of Pre- school Education) to identify the pedagogical strategies which in the Foundation Stage support the development of the skills, knowledge and attitudes that enable children to make a good start at school. Twelve settings, which use effective pedagogy, had been identified by the EPPE team through their measured impact on academic, social and dispositional learning. Case studies, systematic observations, interviews and focus group discussions have been used in this report to document effective pedagogical practices and to illustrate their use across a variety of Early Childhood settings, in addition two reception classes were identified for similar case study. The Effective Pedagogy in the Early Years project therefore aimed to: • Draw upon and extend the analysis of case studies conducted in the twelve EPPE pre- school settings within a range of effective outcomes and pedagogical strategies. • Extend the analysis further with the inclusion of 2 good reception classes (identified on the basis of professional judgement). The total sample therefore consisted of 14 Foundation Stage settings: 1 Playgroup 1 Local Authority day nursery/combined centre 3 Private day nurseries 2 Nursery schools run by the LEA 3 Nursery classes in primary schools 2 Early Excellence Centres/Combined Centres 2 Reception classes • Document the characteristics and practices within the pedagogical strategies. • Collaborate with practitioners from a variety of settings through focus groups to adapt and implement the strategies in local settings and with different providers. The study provides new research evidence while drawing upon existing EPPE research. The research methods that were applied in the EPEY study included: • Review of the international literature on pedagogy • Survey of 46 childminders • Longitudinal study of child outcomes from EPPE - drawn from EPPE (see EPPE Technical Reports) • Case studies of 14 good/effective settings including reception classes • Systematic observations of children within 12 of the 14 good/effective settings • Focus Group Discussions with user groups 2.1 The EPPE design Project EPPE is a longitudinal national research study which follows the developmental progress of more than 3,000 children across England. Children and their families are recruited into the study when they are 3+ years, the age at which many enter their first Foundation Stage group setting. (For those who are already enrolled in a group setting, they enter the study at age 3+ while continuing in their attendance in a setting.) Developmental progress is assessed regularly, beginning when children enter the study and continuing through school entry, the end of Year 1 and also at the end of Year 2 in national assessments. After controlling for family background and child factors such as gender and health, the educational and care factors associated with children’s progress is revealed. The full design of EPPE is explained in EPPE Technical Paper 1 (Sylva et al 1999). Both qualitative and quantitative methods (including multilevel modelling) are used to explore the relationship between the quality of individual pre-school settings and children’s intellectual attainment and social/behavioural development at entry to school. EPPE is a study that identifies the ‘value added’ to children's developmental progress by the form of Early Childhood provision that they have experienced. In addition to controlling for the child’s developmental level at entry to pre-school, EPPE controls for the influence of the family and child 17 characteristics when establishing the ‘effectiveness’ of each setting in its sample. Thus, one of the main questions in EPPE is: Which are the centres where children make more developmental progress than would be predicted by their assessments at entry to pre-school? A total of 141 centres were included in the main EPPE study which selected a random sample of the six main forms of provision in five regions in England to include provision in urban, suburban and rural areas with a range of ethnic diversity and social disadvantage. We aimed for 20-25 children to be randomly sampled in each pre-school centre when they entered and their progress was followed to the end of Key Stage 1 in order to demonstrate ‘analysis centre’, effects in the multi-level analysis. The school entry data (when the children entered reception class) was analysed to identify the ‘pre-school centre effects’ through multi-level analysis, and a sample of twelve settings ranging from average, with some good practice, to very effective practice were selected for case study analysis. The centres were selected on the basis of their scores in children’s developmental progress in: Cognitive Outcomes Social/behavioural outcomes Pre-reading Independence and concentration Non-verbal Cooperation and conformity Language Less anti-social, worried and upset Number concepts Peer sociability Figure 1: The EPPE design. A ‘centre profile’ was created for each centre through systematic observation and questions to staff. The Early Childhood Environment Rating Scale: Revised (ECERS-R) (Harms et al 1998) was used in drawing up each centre’s environmental quality profile along with an extension to this scale based on the Early Learning Goals and what we know about how children learn (ECERS- English Extension or ECERS-E). The ECERS-R rating scale consists of seven sub-scales which provide an overview of the pre-school environment, covering aspects of the setting from furnishing 25 playgroups 500 children 25 private day nurseries 500 children 25 local authority/voluntary nurseries 500 children 7 combined (care and education) 200 children home 380 children 25 LEA nursery classes 500 children 25 LEA nursery schools 500 children Pre-school Provision (3+yrs) Reception Year 1 Year2 (3+yrs) (6 yrs) (7 yrs) School entry assessment N= 3,000+ Age 6 assessments N= 3,000+ Age 7 assessments N= 3,000+ 18 to individuality of care and the quality of social interactions. The ECERS-E describes the curriculum within the pre-school, and includes the areas of mathematics, science, literacy and diversity (whether staff plan to meet individual/particular needs). Each sub-scale is comprised of a range of items describing the ‘quality’ of the specific type of provision. Each item was rated 1 (inadequate) to 7 (excellent). The ECERS-R and ECERS-E are one approach to describing the ‘processes’ through which children are cared for and educated (see EPPE Technical paper 6, Sylva et al 1999, for full details.) An ICT subscale was developed at a later date for another project (see appendix 9 Siraj-Blatchford J and I. 2000) and used in our case studies, along with the child observations, to ascertain the quality of ICT provision in the settings. 2.2 Intensive case studies: sample and data collection All 12 EPPE case study settings are providers of good to excellent practice which boosts children’s development much more than could be expected by their scores at age 3 (and entry to study). Figure 2 reflects all the case study centres, we have not provided our precise details to ensure anonymity of the settings. Researchers on the EPPE project who had been visiting the centres for 1-3 years to conduct child assessments and ECERS ratings for EPPE were trained in qualitative data collection. To some extent they were introduced to certain aspects of qualitative data analysis. They received intensive training to conduct naturalistic observations, to engage in semi-structured interviewing and to collect and analyse documents. Each researcher was familiar with the 14 centres under study (2 reception classes were added), some had made up to 40 visits to their case study centre during the EPPE assessment period, and each researcher spent two whole weeks in each of the following settings collecting qualitative data. Figure 2 - Profile of centres selected for in-depth case studies Centre Type of No. on roll Some Under 3 No. Provision 017 NC approx. 50 - 106 NC approx. 40 - 214 DC (CC) approx. 60 yes 219 PDN approx. 50 yes 225 CC (EEC) approx. 120 yes 306 PDN approx. 90 yes 324 NS approx. 120 - 401 PG approx. 40 - 413 PDN approx. 120 yes 417 NC approx. 50 - 421 NS approx. 100 - 426 CC (EEC) approx. 200 yes Code No of centres in study CC Combined Centres also } 2 EEC Early Excellence Centre } 19 DC Local Authority Day Care – also CC 1 NC Nursery Class 3 NS Nursery School 2 PG Playgroup 1 PDN Private Day Nursery 3 RC Reception classes (501 and 502) 2 Total 14 Since the EPPE children had left the centres by the time of the case studies, we needed to adopt criteria for the selection of Centres to ensure that they had not changed drastically. All centres were selected for case study after we had reassured ourselves that: a) No significant change in management had taken place (same manager/deputy and senior management team) since EPPE. b) A similar or better score in ECERS-E (which had shown a significant correlation with child outcomes) as in EPPE. c) No difficult circumstances (e.g. pending/recent OfSTED). d) Selection of good (in some aspect/s) to excellent centres related to children’s outcomes (‘blind’ – selected by the member of the central team not involved in the fieldwork or analysis). e) As balanced a range of providers as possible. The last criterion was difficult, although we found effective and less effective settings in all types of pre-school provision, in some types there were very few centres in the top range. The day care centre and the playgroup reflect the top range of their provision, hence we had very few to choose from and there is only one of each in our sample of the 12 EPPE case study centres (see Sammons et al Technical Paper 8, forthcoming). The qualitative data was transcribed, ‘cleaned’ i.e. anonymised, and then entered into QSR N-Vivo (explained below in this report), to aid the coding and data reduction. The N-Vivo database includes (see Figure 3): 14 files of documentary analysis, one on each case study centre (a detailed narrative description); 42 transcripts of staff and manager interviews; 204 (approx 400 hrs. of) naturalistic observations of 2 practitioners for two days and selected vignettes and ‘critical incidents’ observed over a 2 week period - for each centre; 107 parent interviews and 14 centre plans of the inside and outdoor learning and play environment. The analysis focuses on patterns of good practice across the centres but particularly at what makes some centres more successful at achieving good outcomes, for example, for numeracy, social development or language development. The intensive case study analyses attempt to tease out the specific pedagogical and other practices that are associated with achieving very effective outcomes as compared to those centres with good or more average practice. In particular the case studies aim to investigate the processes which might explain the patterns of child outcomes in the EPPE study (see EPPE Technical Papers 2, 4, 6 1999; 7, 2000 and 8 forthcoming). 20 Figure 3 - Data entered into QSR NUD*IST - vivo Centre Documentary Staff Teacher Other Parent Centre Number analysis interviews observations observations interviews plan 017 1 3 8 (22:30) 11 08 1 106 1 3 8 (18:55) 6 08 1 214 1 4 9 (30:45) 6 09 1 219 1 2 10 (27:35) 6 10 1 225 1 3 8 (22:20) 7 07 1 306 1 2 4 (10:20) 7 05 1 324 1 4 8 (17:06) 4 08 1 401 1 2 10 (28:21) 3 06 1 413 1 3 9 (11:44) 0 09 1 417 1 2 8 (25:05) 6 08 1 421 1 4 8 (30:40) 12 09 1 426 1 4 10 (23:28) 8 08 1 501 1 4 8 (21:25) 4 06 1 502 1 2 8 (20:20) 8 06 1 Total 14 42 116 (300+hrs) 88 (100+hrs) 107 14 14 files of documentary analysis, 42 staff and manager interviews, 204 transcribed naturalistic observations of staff (400+hrs), 107 parent interviews and 14 centre plans Total = 381 files 1 million word + data set 21 2.3 Managing data reduction using QSR NUD*IST – vivo (NVivo) Qualitative data like interview transcripts and observation notes are notoriously difficult to manage especially in a project of this size. Data reduction can become a ’nightmare’ unless it is well organized and structured to allow for ease of coding and theory building/testing. NUDIST (non-numerical unstructured data indexing, searching and theory building) has been widely used by qualitative researchers to manage small and large data sets. NVivo is a more recently developed sister product which, while based on earlier versions of NUDIST, has proven itself simpler to use and more efficient for the purposes of analysis (Bazeley and Richards, 2000). Software such as NVivo and NUDIST allow the researcher to develop an advanced organizing system that becomes the main tool for organizing and analyzing the data (Tesch, 1990). Exploring, viewing and coding documents are made much easier. NVivo allows the researcher to model her preliminary ideas. In the case of this study, these were taken from a broad reading of the literature on effective practice in the early years (curriculum and pedagogy in particular), and models were created of ‘components’ of particular, key indexing categories. For instance, in our analysis of pedagogy we used the NVivo modeler to create a rudimentary cognitive involvement model using ‘nodes’. From our knowledge of other aspects of pedagogy e.g. instruction, co-construction etc. we modeled these nodes and, from an early stage, we were able to suggest how they might relate to each other. In these initial analyses, through coding transcripts, we gradually amend our models and nodes using anticipatory and procedural data reduction techniques (Miles and Huberman 1994), and theory building. This is undertaken through the iterative process of coding the documents and refining the node definitions and structures (‘trees’) as we sift through the data (back and forth), adding new nodes as well as taking away those which are not evident in the data. The NVivo software package allows us to create attributes for files e.g. for gender, qualification or type of provider. These attributes together with the nodes support the process of analysis further. For example, in our analysis of the 107 parent interviews from the 14 settings we created several nodes to code the text e.g. of how parents choose a centre, what they think makes a centre effective and what educational activities are undertaken in the home and who initiated these. Drawing out all our responses from the 107 parents on home education, there appeared to be some fundamental differences between parents of children at private day nurseries (PDN) and the ‘others’. It was simple to separate the PDN parents, and the other ‘types’, using our attributes and nodes to create new files and compare the responses. We report on this finding later. Once the ‘nodes’, ‘trees’ and ‘sets’ are established further searches can be made to explore the data. For instance, if we developed a particular interest in the relationship between a particular teaching technique (of instruction); (e.g. demonstration, questioning or modelling) within a particular area of the curriculum (e.g. creativity, language, maths etc.) then we could at any time separate and explore the relevant documents to create new data sets. These new sets then provide a means of exploring any links that existed between these categories further. The permutations of analysis and for the modification and creation of data sets, are potentially, vast. However, the main areas for analysis are ultimately determined by the researcher and not the software package! Qualitative analysis is an iterative and constructive process which involves developing ‘hunches’ or hypotheses and going back to interrogate the data to confirm or contradict them. From the quantitative findings of the EPPE study we wanted to 22 explore the associations between the outcome findings and the practices that we had documented (see findings later in the report). 2.4 Case-study Interviews and observations During the EPPE research we had interviewed the 12 managers of the settings. These were informal semi-structured interviews aimed at mopping up any anomalies found between our observations of staff pedagogy and our documentary analysis. As part of the Effective Pedagogy project we devised several new semi-structured interview schedules in an attempt to explore how the CGFS had influenced the practice of the centres and staff we had studied under EPPE. We also asked specifically about ICT and parent involvement. A staff and manager interview were constructed and piloted (see appendices 1 and 2) and a face-to-face interview lasting up to one and a half hours was conducted with the manager and staff (normally those staff we had observed in the EPPE study) of each setting, a total of 30. In addition to this we analysed the 107 parent interviews (see appendix 4). All these interviews were transcribed and added to our NVivo database, coded and analysed. Some aspects of the interviews were analysed using regular data reduction techniques e.g. the examples of routines and how the day was structured, the data for these two aspects was very diverse. Our 400 hours plus naturalistic observations were analysed for pedagogical content. An instrument developed in research conducted for the IBM KidSmart initiative and the Developmentally Appropriate Technology in Early Childhood (DATEC) project was used to evaluate the overall standard of ICT provision in the case study settings (Siraj-Blatchford, J. & I. 2001a/b). This ICT subscale (see appendix 9) was developed to be used in conjunction with the Early Childhood Environment Rating Scale: Extension (ECERS-E Sylva et al). The instrument provides a rating of the quality of the early years ICT environment as well as the resources and the provision offered. The ICT subscale covers three items that reflect the provisions of the Foundation Stage curriculum applied in England and Wales. These are: • Information handling and communication skills • Access and control of ICT tools • Learning about the uses of ICT Each one of the three items includes examples that describe the ICT provision, and others related to equipment, facilities, pedagogy and social interactions. In common with the other ECERS a seven-point scale is applied to indicate a range of practices and provisions that are considered inadequate (Level 1) to excellent (Level 7). A level 5 rating represents the kind of 'good' practice that may be expected to fully satisfy the requirements of the Foundation Stage curriculum and Early Learning Goals; it was developed with this material as reference. 2.5 Interviews with childminders Forty-six from an original sample of 50 childminders from four regions involved in the EPPE project were interviewed. As the project team was concerned with identifying effective pedagogical practices, only those who had had training in supporting children’s learning were selected for inclusion in the sample. In order to guarantee this criterion our initial intention was to sample from child minders who were accredited to the National Childminding Association (NCMA) and eligible for the DfES Nursery Education Grant. Unfortunately this proved to be impossible as only one of the areas at this time had a fully functioning accreditation system in place. The other 23 areas were at different stages of developing their accreditation systems. It was therefore decided, in order to preserve some level of consistency, that our sample should come from recommendations made by the local authority (LA) experts responsible for the registration of childminders and advisers involved in providing childminder training and professional support. The criterion we gave them was to identify childminders who had had some training, preferably to NVQ2 or above, who according to their professional judgement, offered good practice. Thus the names of ‘good’ childminders were identified using professional judgement at LA level (of these, 10 were randomly selected from three areas). All 10 were interviewed from two of the areas, and 8 were interviewed in the third. The other 20 were taken from the area with the fully accredited NCMA network. Of these 17 were interviewed. All childminders had to be caring for one child or more aged between 3 to 5 years at the time of the interview. Information was collected using semi-structured telephone interviews lasting from 45 minutes to 1 hour over a period of two months. One in three interviews was taped and all interviews were conducted by the same, trained researcher (see appendix 3 for childminder interview). Each childminder was asked at the end of the interview to supply a diary record of two sessions (session = half a day) during one week. Of the respondents only one was male. Twenty-one were aged between 28 and 40 years, the remaining 25 were between 40 and 54 years old. Although the range of childminding experience extended from between 2 and 21 years, 61% of the sample had provided services for families for between 10 and 20 years. 2.6 The systematic observations The systematic observations explored the everyday pedagogical practices in 12 of the 14 case study settings. Not all of the 14 settings could be visited due to the time- constraints of the project. Each visit lasted for one week and systematic 20 minute timed observations (see Appendix 5) were carried out with a stratified random sample of children who were observed throughout the sessions – indoors, outdoors, during snack time, and on departure. The aim was to capture typical learning and teaching episodes so they could be analysed for pedagogy, challenge, and for social context. The systematic observations allowed us to focus on individual children within a setting to record for every 30 second interval, for a total of 20 minutes. For each 30 second interval we coded the child’s learning activities, the social groupings in which they spent their time, and the kinds of pedagogical interactions they had with staff. We were eager to learn how children experienced the six areas of learning in the Foundation Stage curriculum (QCA, 2000), and the ways in which adults supported the target child. In analysing the pedagogy of a setting we made a firm distinction between ‘pedagogical interactions’ which were face to face encounters and ‘pedagogical framing’ which was the behind-the-scenes’ work including provision of materials, arrangement of space, and the establishment of daily routines to support cooperation and equitable use of resources. The systematic Target Child Observations focused on the inner and middle layers of the pedagogical model in Figure 4 which illustrates how the aspects of pedagogy all fit together. 24 Figure 4: Pedagogical model The deliberate sampling of children at different ages and abilities allowed us to investigate different pedagogical strategies for younger children, very capable learners and children who were struggling, as well as the different experience of boys and girls. Careful recording of the qualifications of staff also allowed us to explore possible differences in pedagogy related to different training. The Target Child Observation (TCO) instrument [refer to Appendix] was applied to 254 children (about 20 children in each setting) to investigate broad trends across all case study settings. After considering the twelve case study settings collectively, the most effective of all (3 settings) were compared to the rest. This comparison provides information about practice which distinguishes “excellent” from “good” settings. The excellent settings had outstanding developmental outcomes while those in the good settings made strong – but not outstanding progress. Training and piloting the Target Child Observation system took place in a range of Foundation stage settings which were unrelated to the EPEY case study settings. This phase lasted for 2 months which began in January 2002. The target child observations were collected from visits to case study settings between March and July 2002. Inter-rater reliability of the Target Child Observation system was calculated using Kappa, and the results for all codes ranged between 0.572 and 0.841. Details are to be reported in further technical papers. Analysing longer episodes in the systematic observations Of the 254 Target Child Observations, 141 were randomly selected for this more fine- grained qualitative analysis. The 20 minute Target Child Observation was first divided into learning episodes. The learning episodes recorded in this detailed way were coherent activities which were initiated either by children or adults, and lasted for at least one minute. An example would be the target child painting at an easel, Pedagogical Interventions Pedagogical Framing Planning Resources Face to face interactions Parents Arrangement of space Management Structure Assessment Development of behaviour routines for collaborative play Community relations Staff Development Liaison with outside professionals 25 during which she used her fingers and brushes to put the paint onto the paper, and then started to mix different colours together on the paper. The child then stopped, took the paper from the easel, hung it to dry, took off her apron, washed her hands and then wandered around the class coming to the socio-dramatic play area where she started to do the ironing. The observation of the child then continued in the socio- dramatic area. As the child started to wander towards a new activity, this would be marked as the end of the episode. The learning episodes were examined to provide an analysis of who initiated and choose each activity (staff member or child), the cognitive challenge (i.e. did the activity really ‘push’ the child intellectually), the adult’s most important contribution, the resources and materials which supported learning, and the routines (daily/weekly timetables and ‘rules’ such as those for sharing) which the staff established so that children could learn on their own (see Appendix 7). This closer analysis of the timed observations (30-second intervals across the day) and the longer episodes (randomly selected activity) allowed us to answer questions about child choice, curricular balance, pedagogical interactions and the kinds of social groupings which nurture learning. 26 Part 3 Literature on effective pedagogy 3.1 The role of the teacher/practitioner in early childhood Teaching involves appropriate interaction with learners, using a number of strategies (central to these are play, hands-on experience and language development) to enable learning to take place. These interactions will be effected by practices implicit in the social context in which the learning is taking place (Wood, 1986). A typology of commonly applied 'early childhood education curriculum models' (sic) is provided by Weikart (2000) and shown in the model below. The categories applied are broadly consistent with others developed by Weikart (1972), Kohlberg & Mayer, (1972) and Baumrind, (1971) and the EPPE research suggests that as 'ideal types' they apply just as well to the UK early childhood context as to the US context where they were first developed (Siraj-Blatchford et al, 1999). The only analytical difficulty may be in terms of the definition of curriculum applied by Weikart, and the way the term 'initiative' is used. According to Weikart the major organising principle to be considered is the role of either high or low curriculum 'initiation' on the part of the teacher/adult and the child (Weikart, 2000, p58). But in his subsequent elaboration of the various categories of 'educational approach', high teacher initiative is described predominantly in terms of the highly structured pedagogy, and high child initiative in terms of the learner's control over the curriculum. The major organising principle might therefore be better conceived in the binary terms of pedagogy and curriculum. This would also be consistent with Bernstein's (1981) elaborate analysis of pedagogic codes and their modalities of practice. While a comprehensive structural analysis of the various coding principles employed in early education lies beyond the scope of this review, we can employ Bernstein's (1981) formulation of classification and frame to distinguish between the different forms of early childhood practice. As Bernstein himself has noted, while this more limited use of the terms may not have been intended when he first coined them, it does demonstrate their analytical value (Bernstein, 1996, p3). Classification refers to the degree of boundary maintenance between curriculum subject contents. Where the curriculum content is clearly defined in terms of subjects we can therefore refer to that as strong classification. Framing is about who controls what; who selects, sequences, paces etc. the learning. When framing is weak the child (or parent) has more apparent control, when strong it is the educator/ professional who is most clearly in control. So for example, a collaborative, Role of Teacher High Initiative Programmed Open Framework Role of Low Approach Approach High Child Initiative Initiative Custodial Care Child-Centred Approach Approach Low Initiative 27 progressive and permissive classroom illustrates weak framing and a traditional didactic one strong framing. The Programmed approach is highly teacher directed providing for little initiative on the part of the child. The rationale for this method is drawn significantly from learning theory. This pedagogy is usually applied where curriculum objectives may be clearly (and objectively) classified and is likely to be most effective where learning involves the development of basic skills or memorisation. The curriculum content is often highly differentiated. The Open-framework approach provides the teacher with a strong pedagogic structure (or framework) that supports the child in their explorations and interactions with, and reflections upon, the learning environment. In this model, the curriculum classification is weaker as the child has a good deal of freedom to make choices between the various learning environments that are on offer. But the optional environments (e.g. sand, water, block play, puzzles etc.) are often provided to achieve particular (usually cognitive or conceptual) curriculum aims, although these may be more or less identified by the setting. In some settings children's choices are carefully monitored and a broad and balanced curriculum is encouraged over the medium or long term. In the most extreme applications of the Child-centred approach the teacher responds entirely to the individual child's interests and activities. More often, topic or project themes are adopted that have been chosen especially to appeal to the children's interests. The curriculum emphasis is on encouraging children's independence, their social and emotional growth, creativity and self-expression. The classroom/playroom environment is often rich in stimuli, permissive and provides for open-ended exploration and discovery. Of course each of these approaches remain 'ideal types' and the practices in many settings will involve a combination of all three. 3.2 So what is pedagogy and why does it matter? The term ‘pedagogy’ has often been defined quite broadly in continental Europe and the term is sometimes applied in a similar way in UK early childhood contexts. At times this results in accounts where the use of the terms pedagogy and curriculum appear indistinguishable. The recent approach in the House of Commons Select Committee on Education and Employment (2000) enquiry into early learning has been quite different; where ‘pedagogy’ (referred to as ‘teaching’) has been understood as analytically distinct and complementary to, the term; ‘curriculum’. Here curriculum may be understood as denoting all of the knowledge, skills and values that children are meant to learn in educational establishments. For many alternative purposes curriculum has also been defined to include all of the hidden and/or unintentional learning as well. Pedagogy is often referred to as the practice (or the art, the science or the craft) of teaching but in the early years any adequate conception of educative practice must be wide enough to include the provision of learning environments for play and exploration. The term ‘teaching’ may therefore be unhelpful in this context, but effective early childhood pedagogy must still be ‘instructive’. Instruction may therefore be defined to incorporate all of those processes that occur within the classroom that aim to initiate or maintain learning processes and/or to be effective means to achieve educational goals (Creemers, 1994). In these terms it is clear that every effective form of pedagogy must be instructive in some way. 28 Different early years practices are informed by different educational philosophies and values and by the different assumptions that are held about learning, child development, appropriate styles of instruction, and curricula. So it isn’t at all surprising that the pedagogic practices applied in different educational sectors and settings should be different. In fact there are wide differences of opinion and variation in understanding regarding the degree of influence that adults should have over the curriculum (Bennett et al.,1997). However, in both oral and written submissions of evidence to the House of Commons Select Committee on Education and Employment there was widespread agreement that ‘Early Years education should be play-based’. In accepting the terms set out above it can be seen that the provision of an instructive play environment in itself constitutes an instructive technique and to be an effective pedagogue requires the practitioner to be skilled in the selection of appropriate techniques to facilitate learning. As Rumbold (DES, 1990) observed, in the UK there has been a widespread consensus regarding the importance of play among early childhood educators for many years. However, this has not led to pedagogic uniformity (Bennett et al., 1997). In common with the views of many authorities and practitioners, particularly in Scandinavia, early years pedagogy has often been seen in direct opposition to practices involving ‘teaching’. It is therefore extremely significant that the role of the practitioner is defined in the new Curriculum guidance for the foundation stage (DfEE/QCA, 2000) as an explicit ‘teaching’ role that includes: “establishing relationships with children and their parents, planning the learning environment and curriculum, supporting and extending children’s play, learning and development, and assessing children’s achievements and planning their next steps”. (p1) While the introductory statements refer to the different ways that children learn, and the need for practitioners to draw upon; ‘a range of teaching and care strategies’, the desirability of applying ‘play’ based pedagogy is re-emphasised: “Practitioners need to plan learning experiences of the highest quality, considering both children’s needs and achievements and the range of learning experiences that will help them make progress. Well-planned play is a key way in which children learn with enjoyment and challenge during the foundation stage.”(p7) In arguing for a ‘scientific basis for the art of teaching’, Gage (1985) has referred to the distinction between knowledge that is general (nomothetic knowledge), and knowledge that applies to the understanding of a particular event or individual (ideographic knowledge). He argued that teachers creatively apply their nomothetic knowledge to the ideographic problems posed by the unique class of children that they are faced with; with all of their specific needs, socio-cultural status and cognitive and affective demands. Pedagogy refers to that set of instructional techniques and strategies which enable learning to take place and provide opportunities for the acquisition of knowledge, skills, attitudes and dispositions within a particular social and material context. It refers to the interactive process between teacher and learner and to the learning environment (which includes the concrete learning environment, the family and community). This provides a definition wide enough to take in such indirect instructional behaviours as the encouragement of parental involvement while distinguishing it from curriculum in the terms previously suggested. A wide range of authorities (incl. Mortimore, 1999, Leach & Moon, 1999, Bowman et al 2001) have applied essentially the same model where, in their day-to-day interactions with children and parents, 29 capable educators are seen to draw upon a repertoire of pedagogical techniques that have been in some way tested and/or stood the test of time and experience. From this perspective the task of this study is to identify more fully the range of techniques applied by the most effective pedagogues in early childhood education. 3.3 What has research already told us about effective practice? In recent years a number of writers have argued that any focus on ‘effectiveness’ in the early years is mistaken. As Pascal & Bertram (1999) have argued: "What constitutes an 'appropriate' curriculum [sic] for young children is still hotly disputed but it is increasingly apparent that simplistic outcome measures, which only look to knowledge and skill competence in the short term, are seriously underestimating the issue" (p93). However, some longitudinal studies from America have provided a powerful justification for pre-school education. The High/Scope Perry Pre-school evaluation (Schweinhart, Barnes & Weikart, 1993) showed the substantial benefits to be gained through pre-schooling for children brought up in poverty and at high risk of school failure. Most notably a benefit-cost analysis was conducted that has subsequently been widely cited; the evaluation found that: "Dividing the $88,433 in benefits per participant by the $12,3546 in cost per participant, results in a benefit-cost ratio of $7.16 returned to the public for every dollar invested in the High/Scope Perry Pre-school programme". (Weikart, 2000, p65) But while the benefits of a High/Scope pre-school programme have been widely accepted, this does not, in itself, show that pre-school education in general is beneficial or effective. To provide any evidence of this there has been a need for a much more comprehensive study. We have also needed to know more about the specific pedagogy that was applied by High/Scope as well as the variation of pedagogy applied across the early years sector. The research has suggested that social and motivational elements of early years programmes are as important as academic outcomes. In fact in some studies these have been demonstrated to be more important; programmes such as High/Scope have led to greater social outcomes than academic ones. This has been attributed to the process of teaching and learning in the High/Scope programme which emphasises interaction, reasoning, reflection and responsibility for self-learning (not uncommon in other frequently cited programmes e.g. Reggio Emilia). We need to be especially clear about the distinction between pedagogy and curriculum in this context. The High/Scope programme includes strong cognitive and socio-affective priorities, and the plan-do-review pedagogy may be especially effective in providing for both. Research findings from the longitudinal study in New Zealand, ‘Competent Children’ (Wylie, 1998) suggests that by six years of age children gained higher or lower educational outcomes depending on a few factors including; age at which children started early childhood education (before 3 had better impact); quality of staff interactions with children; and the extent to which children were allowed to complete activities. Some longitudinal studies have also shown us that children provided with direct or 'programmed' instruction sometimes do better in the short term than those provided with other forms of pedagogy (e.g. Millar & Bizzell, 1983, Karnes et al, 1983). But the studies also suggest that even when apparent, these gains are short lived, with all significant differences having 'washed out' within a year of the provision ending. Direct instruction has also been found to result in children showing significantly 30 increased stress/anxiety behaviours (Burts et al, 1990). A more recent and rigorous longitudinal study conducted by Schweinhart and Weikart (1997) showed little difference in the academic performance of children provided with direct instruction but significantly more emotional impairment and disturbance which later required more special educational provision. More importantly, the Schweinhart and Weikart study showed that the direct instruction group experienced more suspensions from work and more than double the rate of arrests than either of the other two groups by the age of 25. In terms of serious crimes requiring a custodial sentence, 43% of the direct instruction group gained a felony record, compared with 17% of the child- centred group and only 10% of the open framework group by the age of 25. Both the Schweinhart and Weikart (1997) study and the High/Scope Perry Pre-school study showed a significant difference in the percentage of young adults married and living with their spouses: While 31% of the open framework group and 18 % of the child- centred group were married at age 23, none of those experiencing the direct instruction programme were (Schweinhart and Weikart, 1997). Other studies have shown that ‘formal’ approaches to teaching young children are counterproductive (Nabuco and Sylva, 1996) and can hinder young children’s learning, generating higher anxiety and lower self-esteem. These studies also compare different types of educational programmes comparing High/Scope with a formal programme and a free-play programme. Developmental outcomes have been found to be best in those settings which emphasise a balance between child-initiated and teacher-directed activities. Apart from the plan-do-review model, the other pedagogic strategy that might have contributed to the success of the Perry Pre-school program was the programme of outreach and parent involvement. This is clearly an area that would benefit from further research. A meta-analysis of the effects of a small group of pre-school programmes which were considered to be of excellent quality was carried out by Lazar and Darlington (1982b). The authors limited their selective meta-analysis to pre-school programmes planned from the start as evaluation studies. Each individual project had an adequate sample size, used norm-referenced assessment tests to establish outcomes, assessed outcomes for comparison/control groups, and included a follow up on children well beyond school entry. By these strict criteria the results of 11 carefully monitored programmes were subjected to meta-analysis, a statistical exercise which enables researchers to compare the size of effect across many different studies. These early intervention programmes represent a wide variety of interventions, almost all aimed at disadvantaged children and all of high quality but often on a small-scale. Lazar and his colleagues compiled information on the education and employment of more than 2000 individuals who had participated in early intervention programmes when they were young. In addition, the researchers carried out interviews with the young adults at age 19 and their families. Results from the meta-analysis showed that participation in excellent, cognitively oriented pre-school programmes was associated with later school competence and avoidance of assignment to ‘special’ education. Interviews revealed the individuals who had participated in the intervention programme talked to their parents more about ‘life in school’ and the parents themselves developed higher aspirations for employment of their children. This research suggested that the long-term effects of early childhood education lay not with IQ gains but with children remaining in mainstream education and developing positive views of themselves and their futures. We must note however that these high quality programmes were set up for 31 ‘demonstration’ and ‘research’ – making generalisation to all early childhood programmes impossible. Another important factor in relation to good, early education provision is how long a child receives it. The amount has a bearing on the extent to which children can show long term benefits; in other words, the more early education provision you get the more likely you are to retain longer term benefits. According to Zigler and Styfco (2001) there are two key benefits of early education; to support a good start to school through enabling a better transition, and most pre-school education is successful in this, and the longer term benefits which are based on more intensive pre-school intervention, particularly for children from poorer backgrounds. They argue that a short period of early education is not an inoculation against disadvantage, just as a short period of healthy nutrition is not a guarantee of health for life. Provision needs to be sustained. Recent Government policy in the UK to increase the amount of time children spend in early education is to be welcomed when it is coupled with practitioners who are skilled, knowledgeable and sensitive to an appropriate pedagogy and curriculum for young children. The successful Child-Parent Centres (CPCs) study that Zigler and Styfco refer to (Reynolds, A, Temple, J, Robertson, D, Mann, E, 2001), which showed substantial long term benefits, also showed that: “Quality in the CPC appeared to be relatively high with respect to staff training and curricula” (p. 2379). 3.4 What has the study of learning shown us? It has long been recognised that play has a major role in early childhood development (Franklin, 1999; US National Research Council, 2001). Both Piaget and Vygotsky have written about the strong link between symbolic play with the development of representation and transformation and language and literacy (Schwartzman, 1978; Pellegrini et al, 1991) There is also a good deal of evidence within the psychological literature that learning is an interactive event, where the child actively constructs his/her own understandings within a social and physical environment. Hohmann and Weikart (1995) call this “active learning” (p.17) and define this as “learning in which the child, by acting on objects and interacting with people, ideas and events, constructs new understandings” (p.17). This means that young children require direct and immediate experiences that will enable them to derive meaning from these experiences based on their previous ones. The learning environment must, therefore, provide children with opportunities to be active and take the initiative to learn. The role of the adult is to provide these opportunities and experiences through setting the physical and intellectual environment and through consistent planning and rigorous assessment so that appropriate opportunities may be given. Adult support is also important to encourage children to learn in an active and participatory way. Hohmann and Weikart (1995) drew upon the work of Piaget to inform the High/Scope curriculum programme that emphasises interaction, reasoning, reflection and responsibility for self-learning. In recent years High/Scope has drawn equally from the work of Lev Vygotsky, however, we are keen not to over-emphasise what we consider to be largely a false dichotomy created between these two theorists by many educationalists. High/Scope has taken a developmental perspective on learning in which they argue that human development takes place in predictable sequences, but that each individual is unique in his/her characteristics. Learning experiences should, therefore, be developmentally appropriate for each child. Appropriateness is defined as the extent to which a learning activity: • Exercises and challenges the capacities of the learner 32 • Encourages and helps the learner to develop unique patterns of interest, talents and reach goals • Presents experiences in which the learner is able to master, generalise and retain concepts, skills and knowledge and which relate to previous experiences, whilst linking to future learning expectations. (Adapted from Hohmann and Weikart, 1995, p.15). Piaget’s theories were based on the concept of cognitive structures and on an assumption that children are natural and active developers of their own understandings. Cognitive structures are defined as patterns of physical or mental action that underlie specific acts of intelligence and that are conditioned by stages of development. As the child matures, she/he is able to integrate more abstract structures into his/her understanding. Action and self-directed problem-solving, leading to control of the environment, are considered to be at the root of learning and development (Wood, 1988). Whilst Piagetian theory is a theory of child development and not of education, as Davis (1991) writes, it has had an important potential influence on education. Piagetian principles have been interpreted as implying a pedagogy based on the need for the educator to facilitate developmental change by enabling the child to discover contradictions or ‘disequalibria’ with her/his prior knowledge and understanding. In our analysis of the effective pedagogy project findings below we refer to this as a process of ‘cognitive construction’, where in this classic account the construction is founded upon the child's prior understanding. But before we look any further into this we must recognise that children do not learn in a social vacuum. In fact much of the child’s earliest learning is the product of interactions with adults and peers even where there has been no deliberate attempt to provide instruction. The child learns a great deal from observations of those more competent than themselves. We can draw upon the social cognitivist theory of Bandura (1986) to account for the way in which social (and media) experiences provide for this sort of observational learning. For Bandura the process of social learning begins with ‘imitative’ learning which is subsequently internalised through identification and thereby incorporated in the individual’s self-concept. The child’s construction in this case is founded upon the behaviour of the other person who is considered especially significant. The value of early years educators modeling appropriate language, behaviour, skills and attitudes should therefore be especially recognised; such modelling is likely to be consequential in terms of both cognitive, social and dispositional outcomes. ‘Significant others’ are also crucial in encouraging the child’s engagement with other stimuli. Even when an experience contrasts markedly with some aspect of a child’s prior learning, we can’t take it for granted that children will ‘construct’, we can’t assume that they will learn from that experience. For learning to take place they need to be motivated to engage with the cognitive challenge that is provided for them. In the case of imitative learning this is often provided by the child’s desire to ‘be like’ and/or ‘to be liked’ by the person imitated. In other cases we may need to make specific provision for the child’s affective involvement. Piaget described a learning mechanism which involved the child in the active elaboration of their own mental structures as they assimilated and accommodated new experiences. But crucially for Piaget, in his early accounts of this learning process, while he argued that it was triggered by the recognition of some ‘disequilibrium’ between the child’s new phenomenon or experience and their prior knowledge and skill, he said that it was ‘fuelled’ by the affect of their ‘interest’. The child’s interest in the new phenomenon or experience might be the spontaneous result of their natural curiosity but it could also be influenced by the adults and peers around them. Piaget therefore argued that the child’s intellectual adaptation was as much an adaptation to the social environment 33 as it was an adaptation to their physical and material environment. This provides a potentially strong foundation for early years educational practice as it accounts simultaneously for learning and for motivation. Unfortunately, this latter part of Piaget’s theory