Chapter 1
School Self-Evaluation for the Improvement of Educational Practice

Some theoretical issues

Introduction

Self-evaluation of the kind that is initiated and conducted within schools by teachers seems to have emerged from two distinct historical contexts: the curriculum development movement in the early and mid-1970s, and increased pressure for public accountability subsequent to the Tyndale affair. In relation to the first, the apparent lack of success of many curriculum projects to ‘take’ in the schools generated such axioms as that of the Humanities Curriculum Project which declared: “no curriculum development within teacher development”. Thereafter some of the ideas and methods associated with the experimental and evaluation stages in national curriculum development projects were recognised as offering possibilities to individual schools and teachers seeking professional development and curriculum change in their particular situation. It is significant that many of those who have promoted the notion of school self-evaluation have also been prominently associated with national projects: Marten Shipman, John Elliott, Barry MacDonald and Helen Simons, for example. Although Lawrence Stenhouse rarely refers to self-evaluation, preferring the notion of the ‘teacher researcher’, many of the principles underlying his conception are similar to ‘self-evaluation’ insofar as they share a purpose directed towards teacher and curriculum development.

Self-evaluation has, however, acquired a second purpose in response to increasing political pressure to make education more accountable. It is argued that the process of self-evaluation can become the basis for schools and teachers to render public accounts of their work. The fact that writers who take this stance are often those who also advocate self-evaluation as a means to improvement in practice (e.g. MacDonald, 1978, Elliott, 1979a, 1979b, Simons 1979, 1981) tends to obscure the fact that these two purposes are in many ways distinct and often make different demands on teachers and schools. Whilst acknowledging that these two strands are very much interwoven in the literature, and conceptually related if one starts to analyse ‘professional’ concepts and models of accountability (see Sockett, 1980) my chief concern is to draw out some of the issues which need to be resolved if self-evaluation is to contribute to professional development and educational improvement. I can offer no ‘answers’ and I suspect that what is appropriate in one context might not be so in another. However I believe the enterprise is worthwhile because, what people generally want to know about any self-evaluation is what happened as a result i.e. what difference it made. What I am suggesting is that change is more likely to occur if certain questions concerning the relation of ‘means’ to ‘ends’ are considered beforehand.

Needs, purposes and strategies

Helen Simons (1981) advances three arguments in support of school self-evaluation. Significantly all are closely related to conceptions of professionality and accountability. Her main point is that the rationale for school self-evaluation lies in its potential to improve professional practice. As such it needs to become a routine procedure, initiated internally in response to the perceived needs of the institution. She believes that this approach is more likely to be sustained, to reflect the way schools really operate, and to preserve the autonomy of teachers. She does not deny the possibility of public accounts as ‘by-products’, but she believes that a self-evaluation procedure that responds merely to political pressures is likely to be half-hearted, ‘one-off’, distorting of reality, engendering defensiveness and hostility in teachers, and incapable of bringing about any genuine change.

Some support for Simons’ views can be found in the work of social scientists who have studied change in a number of organisational settings. For instance Benne, Bennis and Chin (1969) identified three broad categories of strategies for change which they termed power-coercive, normative-re-educative and empiricalrational. Power-coercive strategies are based on the intervention of those with legal authority to alter conditions. Since the commitment of participants will often be lacking, compliance or ‘surface change’ may be all that is achieved. Normative-reeducative strategies, on the other hand, involve either problem-solving concerned with ‘activating forces within the system to alter the system’, or the process of attitude change based on the study of one’s own behaviour. The third group of strategies called empirical-rational, has a largely intellectual appeal. Their influence depends on the effectiveness of an idea or practice being clearly demonstrated by research. Of these three groups of strategies the normative-re-educative category appears to have the greatest potential to bring about genuine improvement, because it is based on fundamental changes in attitudes, values and roles. Helen Simons’ emphasis on the need for self-evaluation to be initiated in response to internally perceived needs, and the importance she attaches to the disclosure of assumptions and values underlying current practices, suggest that she supports a normative-reeducative approach to effective change in schools.

Simons’ second argument is really a development of her first. If public accounts are to be required of schools then time is needed for teachers to acquire the skills necessary to produce such accounts (a point made also by MacDonald, 1978). It is a challenging but, initially at least, also a threatening exercise and pressure to do too much too soon may encourage ‘gilding the lily’. Unsure how the public will react, the temptation might be to present the best possible face, whether or not it accords with reality. Moreover, aspects of schools that seem to require remedial attention might well be ignored unless there exists some added incentive such as a definite promise of help in the form of extra resources. In other words schools are likely to be highly selective in what they choose to investigate for accounting purposes if they are uncertain how information will be used and if no obvious benefit is likely to accrue to them as a result. In the short term, therefore, Simons’ believes that self-evaluation should be encouraged without any reference to accountability demands, although, in the long term, the ‘process’ model she describes might provide the most valid alternative to many current models of accounting procedures. In her words: ‘Evaluation on process lines allows schools to demonstrate and account for what they can reasonably be held to be accountable for, i.e. creating the opportunities for children to learn and for the quality of provision’. This is her third argument in support of school self-evaluation. Apparently, therefore, Helen Simons’ justification for self-evaluation is predominantly educational and professional.

Hoyle (1980) also offers a professional rationale for self-evaluation in a reformulation of a distinction between ‘restricted’ and ‘extended’ professionality, which he first made in the early seventies. In the following extract he suggests that systematic self-evaluation may be a means to ‘extend’ professionality.

Stenhouse (1975), however, is critical of the restricted/extended dichotomy. In commenting on a previous formulation he points out that Hoyle’s ideal of the extended professional implies an unquestioning commitment to theory. He argues that such a position undermines the autonomy of the teacher because it invests theory with the status of ‘received wisdom’ to be accepted, rather than ‘provisional’ and open to experimental testing by the teacher in his or her own classroom. It is not altogether clear whether Hoyle is attempting to meet Stenhouse’s criticism in his more recent writing. In any case Stenhouse offers an alternative formulation of extended professionality – revealing that he does not entirely reject the notion:

Both Hoyle and Stenhouse tend to emphasise the development of individual teachers, while Simons focuses more particularly on the development of schools. However, nowhere in the current literature of self-evaluation, is the precise relationship between the professional development of individuals and the improvement of practice in institutions fully worked out. This leaves a number of questions unanswered. For instance, are we to assume that schools will automatically improve the quality of their educational provision, if all the individuals within them become fully developed professionals; or is an institution more than the sum of its parts? Are there special conditions to be satisfied and courses of action to be undertaken, for the purposes of institutional improvement, which are quite distinct from the processes associated with professional development? Indeed, are there occasions when individual and institutional development are in conflict? I am inclined to think that some conflict is quite likely, particularly when ‘development’ is associated with a concept of ‘career’. Individuals who are energetically engaged in their own professional development frequently envisage the pattern of their careers extending beyond the single institution; experience in a variety of settings may also be a requirement of full professional development. But if a whole staff is active in this way, and if one consequence is an increase in staff turnover, then the stability of the institution and hence its power to develop as a unit, may be seriously impaired. On the other hand if improvement of practice in the institution, as a whole, becomes the overriding goal then many of the professional needs of individuals may be ignored. However, to formulate the question as either institutional development or personal development would over-simplify a complex relationship, and it should be possible to establish conditions in which the professional development of individuals assists the development of the institution as a whole and vice versa.

While I cannot ignore the implications of the process for individuals, my concern, in this paper, is more with the institution. In this context, therefore, the prefix ‘self’ refers more particularly to the whole school or a significant part of it, such as a department. For this reason, and those outlined in the previous paragraph, it may be helpful to look at another notion developed by Hoyle: that concerning the ‘creativity of the school’. In an article discussing innovation in British schools (Hoyle, 1975) he argued that “institutionalisation has been a problem, since there has been a lag between innovations in curriculum, method and organisation of teaching/learning and necessary changes in what might be termed the ‘deep structure’ of the school.” (Hoyle, 1975, p. 343). Thus, if innovations are to become part of normal institutional practice much depends on the ‘creativity of the school’, that is, its capacity, as a social system, to adapt to and sustain change. According to Hoyle this may necessitate changes in a school’s ‘organisational character’, in particular, its authority structures, its decision-making procedures, its professional relationships, and even its pedagogical ‘code’ (e.g. from traditional/closed to progressive/open or flexible). Here we are faced with a dilemma because such changes in internal organisation are themselves major innovations. Where then is the ‘prime mover’? At the time of writing, the process of self-evaluation is, in most instances, in the category of innovation. Therefore, if Hoyle is right, whether or not it will be effectively institutionalised will significantly depend on the degree of creativity exhibited by the institution and, in particular, its willingness to change its internal organisation.

Levels of evaluation

In her article, Process Evaluation in Schools (1981), Simons emphasises the need to evaluate the school as a whole, or a policy issue which concerns the whole school. The case she makes for whole school evaluation rests principally on her assumption that school policy is an important area for inquiry. She also believes that self-study involving the whole, or a major part, of the institution will contribute to extending the professionality of the teacher group. (In this respect she seems to lend tacit support to Hoyle’s notion of the ‘creativity of the school’.) There is an ambiguity in Simons’ account however. When she writes of studying the school as a whole, she doesn’t make clear whether she means the ‘whole school’, in a literal sense, or some theme, aspect or issue which has ‘school- wide’ implications. The first would seem to me to be totally unmanageable in a single exercise although something approaching it might be achieved in a series of exercises over a number of years: a gradualist, incremental approach. (It is worth considering how ‘practical’, in these terms, are some of the LEA school evaluation schemes which are emerging at this time.) A related issue is whether school self-evaluation can be interpreted as more to do with efficiency (a managerial perspective) than effectiveness of teaching and the quality of learning (an educational perspective). Basically the question is whether self-evaluation at the level of the institution can constitute curriculum evaluation, that is (in the broad sense of ‘curriculum’), the evaluation of the total experience provided for the pupil or student by the school; or whether such exercises will inevitably become the evaluation of teachers (rather than teaching) and management procedures. Thus, when we refer to self-evaluation for institutional improvement we need to consider carefully the sense in which we understand the term ‘improvement’. So far I have assumed, like Helen Simons, that the purpose of institutional self-evaluation can be improvement of the quality of education in the institution as a whole. Other writers deny that self-evaluation operating at this level can have this effect.

Lawrence Stenhouse, for instance, insists that all well-founded curriculum development, and therefore the betterment of educational practice whether at the level of the individual teacher or the school, must be based on the study of classrooms (Stenhouse, 1975). This position is one that has been strongly supported by John Elliott who has argued that the focus of evaluation should be classrooms because this is where the most important educational transactions take place. Moreover, he has suggested that whole school evaluation can be developed on the basis of evaluation across individual classrooms. In a paper outlining strategies for process evaluation in classrooms (Elliott, 1979b) he describes a way in which teachers might ‘share’ their personal classroom evaluations with their departmental colleagues, so contributing to an evaluation of their subject in the school. By extending this process of ‘sharing’ still further, an evaluation of a whole school curriculum could then be produced. Underlying Elliott’s description is a conviction that “self-evaluation in the classroom should be the foundation stone of the total evaluation process in the school”, because “the main purpose of curriculum evaluation is the improvement of teaching in classrooms”, and “whether such improvements occur ultimately depends on the attitudes of classroom teachers”. (Elliott, 1979b, p.6).

Having insisted, however, that improvement of practice in schools depends most significantly on the attitudes of teachers, we discover that, in an article written only a little later, Elliott (1980) acknowledges powerful constraints on a teacher’s capacity to effect change. The following passage illustrates this:

While, at the time of writing, Stenhouse maintains his position that improvement in practice depends on the attitudes and actions of individual teachers, I suspect that Elliott is increasingly prepared to acknowledge the extent of structural influences. (A result perhaps of his move from a project based in classrooms i.e. The Ford Teaching Project, to a project focussing on whole schools i.e. the Cambridge Accountability Project.) The tension that is highlighted is a real one and has implications for the level at which school self- evaluation needs to operate.

The arguments of Stenhouse and the ‘earlier’ Elliott are persuasive and evaluation across classrooms may be able to provide a composite evaluation (i.e. an aggregate of individual self-evaluation). However, it may fall short of holism because it cannot provide an evaluation of the complex relationships between individuals and the total context in which they work, nor the interrelationships among the various features of that context. Either through dilution, or because personal issues are more pressing, an ‘aggregate’ of classroom evaluations may pay insufficient regard to those structural features which have an important bearing on educational practice. Therefore, while the importance of classroom and inter-classroom evaluation cannot be denied, there may still be a strong argument for a significantly different exercise at the level of the school as a whole.

Initiatives and control

According to Elliott (1979b) it is appropriate and desirable that evaluation at classroom level is carried out by the classroom teacher; on his own initiative. He may be greatly helped by the support of a sympathetic outsider but ideally it should be he who solicits this help. “Evaluation should be viewed as a participatory process rather than something that is done to teachers” (Elliott 1979b, p. 6). However, if institutional issues are the focus of self-study it is not always feasible to involve, in a direct way, all those whose practice would be implicated. For instance, if the use of resources, the organisation of the whole curriculum, or general assessment policy are to be evaluated, information may be required from large numbers of teachers. But it is unlikely that they would either initiate or conduct such an evaluation themselves, en masse. How then can their confidence in the evaluation be maintained, so that they willingly heed any diagnosis and suggestions for improvement in areas over which they do have control? In other words, how can evaluation be accomplished at institutional level without undermining the autonomy of individuals and thus inhibiting, rather than promoting, their professional development?

This is essentially a political problem, i.e. concerning power and control, and one that MacDonald (1977) became aware of in his evaluations of national curriculum projects. In response he developed a political classification of evaluation studies and advocated democratic evaluation, as an alternative to more familiar bureaucratic evaluation (akin to management consultancy) or autocratic evaluation (akin to the ‘expert’ advice offered by the academic).

The principles outlined by MacDonald were developed in relation to evaluation by outsiders. However, his concepts of confidentiality and control, negotiation and accessibility can equally well be applied to institutional self-evaluation. In this context they can be employed to protect the interests of those whose work is evaluated and give them effective control over the evaluation process, whether or not they are the evaluators. In another article, Helen Simons (1979) works out the implications of applying democratic principles to school self-evaluation.

In this she recognises the unequal distribution of power in institutions and suggests that democratic procedures could protect the least powerful by giving them control of the information they possess. In other words, if a teacher is involved in an evaluation as an informant, in the first instance, then he or she should be able to decide what information those who act as evaluators will have access to, and how it may be used. This strategy is more likely to create confidence in the evaluative process thereby encouraging receptivity to anything diagnosed as in need of remedial action.

This does not altogether solve the problem of where initiatives should come from and who should be employed to carry them out. A deputy head, or equivalent, might be in the best position to stimulate and support self-evaluation because he has whole school responsibility and dual access to classroom teachers and the head. However, his managerial status may pose problems unless he enjoys the full confidence of all staff. Moreover, it would seem a pity to dismiss initiatives from elsewhere, particularly from those in a position to perceive a need not immediately visible to others, e.g. a remedial teacher’s concern with some aspect of special educational provision. Furthermore, in any institution certain individuals tend to come to the fore because of their ability to inspire the confidence of their colleagues, and it may be that these are in the best position to carry out, if not initiate, evaluations, whatever their roles or status positions. Realistically, however, it seems unlikely that someone with no position of special responsibility would have sufficient authority to take such an initiative alone. As a general rule, whoever takes on the role of evaluator needs the full support of those from whom information will be required. If this is lacking then he or she might expect some important information to be withheld. Self-evaluations which involve investigation of policy issues at institutional level are often vitally dependent on the willingness of the head and others at the top of the status hierarchy to open up their practices to evaluative critique. It is difficult to imagine their being very enthusiastic about this if the evaluator is a very junior member of their staff.

Summary

In this paper it has been my intention to consider those issues which I believe have a bearing on the effectiveness of self-evaluation in contributing to change in schools. Whilst I do not deny their existence, I have tended to ignore accountability issues except insofar as the concept of professional accountability, i.e. a teacher’s professional responsibility towards himself, his colleagues and his pupils, is implicit in procedures of self-evaluation which have a commitment to the improvement of practice. Although the discussion has been largely concerned with the literature, the importance I attach to certain ideas has come through an examination of a number of concrete examples: a departmental self-study which I conducted whilst I was still teaching in schools, a primary school self-evaluation supported by a teachers’ centre warden, and two exercises conducted as part of a secondary school’s staff development programme. These examples are reported in detail elsewhere (Open University, 1982), where I also consider the accountability implications of school initiated self-evaluation. By concentrating, in this paper, on the single purpose of professional development and change I hoped to identify, more starkly, those features of self-evaluation that are likely to influence its outcomes. As a final summary I would like to frame these issues as a set of questions which schools and teachers might consider in planning self-evaluation, and which researchers might consider worthy of empirical inquiry.

1. What strategy of change is likely to be most effective, given the particular school context (e.g. power-coercive, empirical rational, normative-re-educative)? What kind of programme for self-evaluation is congruent with that strategy?
2. If self-evaluation calls for changes in internal organisation, is the school prepared for this eventuality? Can structures be developed (e.g. decision-making, staff development) which will accommodate/facilitate/cope with such changes?
3. Is the proposed self-evaluation concerned with ‘curriculum’ (in its widest sense)? If not, how worthwhile is the exercise likely to be in terms of the school’s future educational provision?
4. Is it likely that more will be learned, and the possibilities of constructive action increased, if the classroom is made the basic unit for self-evaluation? Are there important educational issues which make ‘whole school’ evaluation desirable?
5. Who is in the best position to initiate and conduct self-evaluation if the intention is to encourage receptivity to diagnoses and suggestions for change?
6. What kinds of skills are required to carry out the proposed self-evaluation? What kind of in-service provision needs to be made for the acquisition of these skills, and how much time needs to be allowed for proficiency and confidence to build up?
7. How can problems associated with ‘threat’ be avoided? Does this require that those whose practice is to be evaluated should retain some control of the process? If so, what kinds of control and how much, are conducive to promoting change?
8. How can a suitable balance be achieved between the need for the personal and professional development of teachers, and the need for improvement in the institution as a whole?