Behavioural issues within a primary school

Introduction.

This proposal sets out the rationale, aims and methodology, for a research project into behavioural issues within a primary school setting, and their impact on teaching and learning. The specific enquiry being pursued is whether or not current mainstream thinking and official policy on teaching and learning represents an effective means of influencing pupil behaviour. The weight of expectation regarding the moderation and re-shaping of behavioural problem now appears to rest significantly upon the teacher, their, planning, assessment, and delivery of learning: if poor behaviour is exhibited, it is often attributed, to a significant degree, to failures in these areas.

It is argued here that important questions are raised by this position. For example, what is the provenance of this logic in evidential, i.e. research terms? How widely should the teaching/learning and behaviour relationship be interpreted, i.e. are all behavioural difficulties supposedly mediated through such means? These are immediate issues of concern to all teachers, with extensive implications for their professional classroom practice, and school praxis as a whole. From my own position as new practitioner, this raises both specific and wider issues for personal practice, as well as the way in which I apprehend the overall educational system: what skills-sets are required in order to implement this approach successfully, and how might they be developed? What are the implications for planning and assessment in terms of different kinds of behavioural difficulty? Going beyond this level of interaction, how should I, as a practitioner, assess such policies and frameworks as they emerge and are disseminated across the profession? Is it appropriate that teachers, schools and parents simply accept the logic each successive policy change as given, or should there be a more fundamental dialogue between practice, research, and theory?

Literature Mapping,

As might be expected regarding such an important issue, there is a large canon of literature around the subject of behavioural issues: the same is true of teaching and learning. It is far beyond the scope of this limited review to comprehensively survey all of the related sources: instead, they are examined here through some representative texts, concepts and controversies. It is also the case that these literatures overlap to a significant degree. However, it is not particularly helpful to envisage such genres as cumulative and undifferentiated, or informed by the same logic or perspective: quite the reverse is true. In fact it is probably more helpful, in reality, to regard them as being composed of distinct sub-genres, whose character and focus is determined largely by their intended audience and substantive function. For the sake of this discussion, these sub-genres may be envisaged as the official, i.e. government publications, the semi-official, i.e. teacher’s network and other local authority texts, the academic or analytical, in the form of research publications by higher education specialists, and the self-help or didactic sub-genre. The latter is probably the most expansive, and incorporates texts developed specifically for teachers, parents, and other agents with an interest in the classroom praxis created around behavioural issues and learning. It must be acknowledged that, although it is being represented collectively here, this group incorporates multiple strands of thought and opinion: one noticeable contrast occurs between those who echo or endorse successive official interpretations, and those who critique them.

One theme which features pervasively across all of these sub-groups, is the subjective and qualitative nature of the subject analysis itself. After all, in discussing behavioural problems, we are dealing with a phenomenological social subject, which can never be known or defined in absolute and specific terms. Consequently, one issue which is raised from a holistic overview of this literature, is the nature of the relevant research, and its translation into both official policy, and classroom practice. In an environment where – for better or worse – these two are intrinsically bound together, is it not important to ask how such transitions are made, and how the objective integrity of such processes is assured? In other words, in the construction of an increasingly evidence-based practice, who assesses the relevant architecture, and ensures that behavioural issues are being weighed effectively in the educational mix?

Since there is a clear tautology running between the research, policy, and practitioner discussion of this praxis, it is not surprising that there areas of agreement, and also, significant areas of contention. It is around these areas that the rationale for this research has been developed. As Croll has observed, the claim of any educational research to have established best practice in absolute terms is at best questionable, and yet this is essentially what happens in practice. (Croll 1986: p.182) As Taber warns, despite the tautology which runs from the political centre and through the relevant intermediaries, it is ultimately the teacher’s responsibility understand and assess what truly represents ‘best practice.’ Teachers are essentially apprised of the research findings deemed most relevant when initially training, and subsequently updated through continuing professional development, but this is a discourse rather than a dialogue. The government initiates research, construes its findings, formulates policy, and issues guidelines in sufficient quantities, some judge, ‘…to allow teachers never to think for themselves again.’ (2007: p.4). This is obviously an extreme view, and part of a separate debate about the wider nature of training and practice as a whole: it does nevertheless, raise questions about how teaching staff should collectively assimilate and interpret the continually evolving policy framework.

Official Sources.

In its latest exposition to parents, the DCFS concedes that ‘Teachers cannot teach effectively and pupils cannot learn effectively in classes disrupted by poor behaviour.’ (2009: p.4) This and similar documents outline the various rights and responsibilities contingent upon this requirement. A re-occurring point, and one which seems to have achieved hegemony in OFSTED and other official circles, is that behavioural standards and the quality of teaching and learning are closely linked. Sir Alan Steer’s review of behaviour in schools concluded that ‘…all schools should be required to produce a written policy on teaching and learning..’, with specific regard for its behaviour implications. (Steer 2009: p.5) The notion that ‘…bad behaviour in schools is a complex problem which does not lend itself to simple solutions…’ is specifically balanced by the assertion that ‘…Consistent experience of good teaching promotes good behaviour.’ (Steer 2009: p.72) This theme is echoed in many OFSTED inspections, reports and publications, including Improving behaviour and attendance in primary schools, (2005), which advises that, where the prescribed Behaviour Improvement Plan has been implemented, ‘…small but significant changes have…improved attitudes, motivation and pupil behaviour…An ethos that values pupils’ academic and social achievements has been created…’ (2005: p.2). This is obviously a desirable state of affairs, but the way in which it is achieved will probably vary between each context.

The important point here is that, in contemporary official parlance, good practice is that which takes account of the Assessment for Learning framework, using it to mutually reinforce achievement and behaviour. For example, as the QCA points out with regard to primary mathematics, there are many classrooms where ‘…pupils do not perceive the structure of the learning aims that give meaning to their work. Therefore they are unable to assess their own progress.’ (QCA 2003: p.3) As children’s progress is measured incrementally through objectives and outcomes, and constantly updated through day-to-day and periodic assessment, the next target should be both challenging and supportive, and sensitized to each learner’s capabilities. By ‘capabilities’, we are here referring to their operational and higher order skills, as well as their emotional capacity to push themselves through the next learning barrier. Therefore, day-to-day objectives should be mediated partly through previous feedback, self and peer assessment, facilitating the engagement of the learner: periodic assessment should bring the relevant national standards into the classroom, but with the same loop operating between the positioning of the learner’s progress, and promulgation of the next medium-term planning task. In this model, both practitioner and learner can regularly ‘step back’, visualising progress in value-added terms, thereby merging day-to-day and periodic assessment. In terms of behaviour, this mechanism supposedly helps to avoid the suspension of learning which occurs when a learner encounters objectives which are poorly matched to their ability level, insufficiently attentive of previous progress, and poorly adapted to their learning style. As Meadows points out, when certain emotional states are experienced frequently and saliently, they effectively become ‘fixtures’ within feelings about the self, subsequently influencing a wide range of behaviours, such as ‘….perception, emotional expression, cognitive processing and social relations.’ (Meadows, 2006: p.438). It is as this point that poor, learning-contingent behaviour might occur. The official literature therefore reflects the fact that a Goleman-esque focus on emotional intelligence has become deeply embedded in the teaching, learning, and behaviour management approach: the most literal expression of this is to be found in the primary Social and Emotional Aspects of Learning initiative. (DFES 2005).

Unofficial Sources.

The self-help or practitioner genre is effectively split between those who examine behavioural issues in isolation, and those who discuss them in more holistic discussion of teaching and learning. The work of Rogers, Muijs, and Reynolds falls into the former category, whilst that of Cowley and Robertson features in the latter. As Lawrence et al. point out, the real extent, nature and effect of behavioural problems in schools, and their effect upon teaching and learning, have always been, for a variety reasons, opaque issues. (Lwrence et al. 1984: p.6) In his study of behaviour ‘recovery’, Rogers notes that, whilst teachers provide significant emotional and social support for students, they do so ‘…at a cost.’ (Rogers 2005: p.5) He also notes that teachers themselves are part of a stakeholder chain in addressing such problems: often however, they operate in a kind of ‘structural isolation’: part of this problem is their ‘…natural reluctance to share concerns…for fear they may be seen as failing, or…incompetent.’ (Rogers 2005: p.6). The important point here is that behavioural difficulties, like special educational needs, are the responsibility of the whole school. The rise of the Inclusions role and multi-agency working remains implicitly focussed on achievement in school, placing teaching and learning at the centre of the multi-disciplinary support effort. (DCSF, 2007: p.1)

Research Question and Aims and Objectives.

The thematic objective of this research is to examine the relationship between contemporary teaching, learning, and behaviour in a primary context. It does not, however, envisage this issue in simple pejorative or attributive terms: it would, after all, be simple to engineer which illustrated the fact that there are behavioural problems in primary classrooms, and yes, they do affect both teaching and learning in different ways. Arguably, this would be something or a cul-de-sac, as well as merely repetitive of much earlier published material. Instead, it attempts to explore how the relationship between these two elements is being – or could be – mediated by the contemporary policy frameworks. Specifically,

How effectively are pupils’ behavioural problems addressed through the current teaching and learning approach, including day-to-day and periodic assessment, objectives, planning, and teacher, peer and self assessment?

By what criteria is this relationship assessed within school(s)?

How do the learners themselves consider that their behaviour is changed, modified, or affected by the above framework? Do they see improved and assessment-informed teaching and learning as an effective way to remove any residual barriers to learning which obtain through behavioural problems?

With specific regard to learners with recognised behavioural issues, how do they perceive that their learning is affected by their own behaviour?

To what extent is it possible to objectively assess this relationship, i.e. through the triangulation of planning, assessment, and behaviour records?

What are the attitudes, and what have been the experiences, of practitioners themselves, in attempting to mediate recognised behavioural problems through an improved teaching and learning practice?

The aim will be to elicit responses to questions based around these lines of enquiry, and pitched appropriately for each cohort of respondents.

Methodology.

The qualitative nature of the wider research base has already been discussed in the context of the relevant literature: it is, however, equally important to examine this issue with regard to the methodology for this particular work.

In paradigmatic terms, this is a predominantly phenomenological piece of work, however it is also planned to employ some element of positivist analysis: consequentially, it is important to assess the implications and limitations of both paradigms. As Collis and Hussey have indicated, in the context of practical research, and especially where it concerns a social context, positivist and phenomenological data should be visualised, not as discrete disciplines, but rather the two ends of a spectrum which the researcher will inevitably move along, stopping at the most appropriate point. (2003: p.48) They are in fact ‘ideal types’ of data, whose claims to that identity seldom stand up to rigorous scrutiny in the context of published research.

In other words, statistical data is only completely positivist where used to illustrate the exact values it expresses: qualitative data is only genuinely so when used to describe the specific behaviour or events from which it was recorded. The important point here is that neither type of data is usually afforded this degree of care in research interpretations: statistical data is frequently extrapolated to suggest tendencies amongst a much wider sample than that which produced it, whilst qualitative data is used to illustrate phenomena which did not generate it. In some respects, this is of course unavoidable: the total population(s) results, especially in terms of educational cohorts, are unobtainable, unless particular statutory output figures, i.e. examination results, are considered. The only research body with the resources to achieve a 100 per cent survey of any given population is the state, and it only does so in particular contexts. Considering these qualifications, this particular work will be predominantly qualitative, whilst some limited statistical data will be employed to suggest trend s along which a more expansive study might be pursued. It also has to be acknowledged that the reception of research findings may itself be mediated through the supposedly dominant paradigm. This may especially be the case However, as Muijs points out, to attach ‘radical subjectivist views’ to quantitative researchers is misleading, whilst to envisage all quantitative researchers as ‘positivists’ is ‘…equally inaccurate.’ (Muijis 2004: p.5)

The planning of the research also had to encompass decisions regarding other epistemological options, especially whether it would use probability or non-probability sampling, a cross-sectional or longitudinal approach, and an exploratory or descriptive questioning framework. In practical terms, the sampling options were prescribed by practical and ethical issues, as well as those of availability, and the necessary authority to proceed. However, it remains important to distinguish between the relative merits of the two models: probability/random sampling allows the researcher to claim that their cohort of respondents is representative of the wider population, due to their selection at random. This means that, in theory at least, each sub-group within the whole has a chance of inclusion. (McGivern 2006: p.277). A non-probability/purposive model is instead composed of specifically elicited respondents, eliminating the grounds for any claim to representation of the whole population, but assuring their relevance to the research. (McGivern 2006: p.277). A cross-sectional design focuses on one chronological context in depth, and can therefore be used to detect trends, patterns, and suggest possible causality: however, due to its discrete timeframe, it cannot usually prove causality. (McGivern 2006: pp.99-100). By contrast, a longitudinal design allows a cumulative series of contact with the subject(s), so, depending on its duration, it may allow firmer conclusions to be drawn from a more extended set of data. (Malhotra and Birks 2003: p.68). Deliberating between an exploratory or descriptive format of questioning is basically informed by the direction of the research and the amount and quality of the data already held. A descriptive approach relies on a firm hypotheses and a correspondingly defined set of parameters, (Malhotra and Birks 2003: p.65). whilst the exploratory alternative implies a lot of holistic or ‘open’ questioning, to identify trends and further refine the research.

In consideration of these options, it is proposed that this research be conducted as follows.

Non-probability or purposive in terms of sampling, due to the practical considerations of suitability, the voluntary nature of the sample, the contingent availability and permission, and the ethical screening of some participants. A random or probability-based sample could be a viable alternative for a larger study, if greater resources and time were available to take account of the potential drop-out rate and self-exclusion.

A cross-sectional approach, due to the practical constraints of time upon the study. However, this does not preclude the development of a more extended, longitudinal variant, once further research options become are identified.

A balance of descriptive and exploratory questioning. There are two strands of reasoning supporting this approach. Firstly, the questioning must be supported by some prior knowledge of the issues involved, as well as knowledge of the respondents themselves: the former has been generated by a combination of academic study, and placement experience. The latter is unavoidable, because of the cooperation and engagement involved in identifying the suitable participants, and the ethical considerations contingent upon their particular needs or vulnerabilities. In essence the ‘random’ element has already been eliminated by the use of a non-probability approach: the logical corollary is that the questions, although standardised, should be well adapted to the subjects themselves.

Complimentary elements of ‘closed’ and ‘open’ questioning, to achieve a balance of positivist and phenomenological outcomes. Although substantially a qualitative piece of research, it is hoped that some binary ‘yes/no’ responses will generate some quantifiable data, suggesting patterns which can inform the direction of further research. Questions will be standardised, to avoid unnecessary variation in responses. (Bryman and Bell 2003: p.116)

Ethical Dimensions of the Study

As Padgett points out, there are two generic levels of ethical responsibility present in any research process: the macro-level, which demands the objective framing, conduct and interpretation of the research, and the micro-level, which requires ethical treatment of the respondents and their disclosures. (Padgett 1998: p.44) The first level of hierarchy arguably implies particular responsibilities in the context of qualitative research, due to the flexibility of the paradigm. As Padgett points out, the absence of a rigid epistemology can liberate the researcher, but carries with it additional responsibilities for the design, conduct and assessment of the process. (Padgett 1998: p.106) The second level concerns not only the respondents themselves, but everyone in the wider stakeholder group around them, who have a professional or vested interest in ensuring their care and development. As Black cautions, disclosures in research have the potential to damage feelings, reputations, careers, or even lives, and may affect institutions as well as individuals. (Black 2002: p.63) At the most basic level, individual participants effectively entrust their anonymity to the practitioner, and there are multiple ways of compromising it, even if their name does not appear anywhere in the published findings. (Padgett 1998: p.38).

The overriding point here is that this research is being conducted by a putative practitioner, and this in itself implies additional and particular responsibilities during the process. As the British Educational Research Association points out, this dual role always places additional responsibilities on the researcher. (British Educational Research Association, 2004: p.6) As Cohen et al. have argued, ‘procedural’ ethics are, in themselves, not enough to ensure a consistently ethical approach. One obviously has to ensure how the research ‘….purposes, contents, methods and reporting and outcomes abide by ethical principles and practices.’ (Cohen et al. 2007: p.79). However, there should also be a heightened level of sensitivity to the attitude and emotional state of the primary-age respondent during the research process itself, as well as their attitude to the findings. As the Social Research Council puts it, ‘…The boundary between tactical persuasion and duress is sometimes very fine and is probably easier to recognize in practice than to stipulate. In any event, the most specific generic statement that can be made about adequate consent is that it falls short both of implied coercion and of full-hearted participation.’ (Social Research Council, 2003: p.29)

One example of this would be the unintentional exacerbation of situational stressors amongst peer groups. It might, for example, be interesting or useful to elicit the views of other learners, whose experiences of class-based teaching and learning are mediated through the behavioural problems of other pupils. However, this might be deemed unethical in terms of the residual tensions raised, and the subjectivity of interpersonal or peer reflection.

Similarly, there may be defined parameters of the questioning which could be employed with regard to practitioners themselves. It would, for example, be unethical to ask teachers to comment pejoratively on official policy, given that this kind of dialogue does not have an obvious place or precedent in contemporary educational management discourse. As Mann has cautioned, any qualitative research process bears implications, both for those in an institution context, as well as for the researcher themselves. However, such implications may not be immediately obvious to either, until the ramifications of disclosure become clear. (Mann, 2003: p.67).

This does not mean however, that questions constructed with due consideration might not illustrate areas for development in current classroom practice. The important point here is that neither an ostensibly ethical approach, or the legitimate securing of consent, absolves the researcher from apprising respondents of the likely consequences of participating. (Social Research Council, 2003: p.35).