You are on page 1of 17

Oxford Review of Education,

Vol. 29, No. 3, September 2003

Differential Teacher Effectiveness: towards a model


for research and teacher appraisal

R. J. CAMPBELL, L. KYRIAKIDES, R. D. MUIJS & W. ROBINSON

ABSTRACT The article reviews the research on teacher effectiveness and develops the case
for a model of teacher effectiveness in which differential effectiveness is incorporated. Five
problems with current concepts of teacher effectiveness are identified: undue influence of
available techniques upon the concept; emphasis on school, to the detriment of teacher,
effectiveness; tenuous relationship to teacher improvement; narrowness of operational
definitions in research; and the development of generic, rather than differentiated, models.
In addition the failure of existing models to explain variance in pupil outcome at the
classroom level, the neglect of teacher self-evaluation, and the restricted measures of pupil
outcomes are noted. A differential model is proposed incorporating five dimensions of
difference. These refer to teacher activity, outside as well as inside the classroom;
curriculum subject; pupil background factors; pupil personal characteristics; cultural and
organisational contexts of teaching. The developmental functions of such a model for
research and for teacher appraisal are explored. Four problems for implementing a
differentiated model are raised: complexity, stakeholder expectations, values, and policy
acceptability. These are considered in the light of the controversial Hay McBer model in
England and of models developed in Europe and the USA in the early decades of the last
century.

BACKGROUND
The search for the characteristics of the effective teacher has had a long history.
Robinson (2004, forthcoming) shows that models of effective teaching were being
developed in England in the early decades of the last century, as educationists
attempted to move instructional skills from a largely intuitive craft practice on to a more
scientific basis, in tune with the emergent scientific rationalism of the time. Professional
debate and interest in defining the principles of effective teaching were particularly
prominent, for example, in the work of the Training College Association in England
and its journal The Journal of Experimental Pedagogy during the first two decades of the
twentieth century. There were also developments in experimental pedagogy in the new
University Departments of Education, notably through the pioneering work of Pro-
fessor Joseph Findlay at Manchester and Professor James Welton at Leeds. Drawing
upon embryonic forms of research based upon work with trainee and expert teachers in
demonstration and model schools, these men sought to identify a set of rational
principles which could form a professional consensus on effective teaching and which
were located at the nexus between theory and practice.

ISSN 0305-4985 print; ISSN 1465-3915 online/03/030347-16  2003 Taylor & Francis Ltd
DOI: 10.1080/0305498032000120292
348 Oxford Review of Education

Drawing on hitherto unpublished data, Robinson shows that five broad propositions
for effective practice on the part of the ordinary classroom teacher were suggested,
comprising what would now be thought of as two clusters of knowledge, two clusters
of skills, and an integrative capacity. They were:
(a) meticulous planning and preparation based on strong subject knowledge;
(b) an understanding of the different modes of interaction between teacher
and taught;
(c) the logical and systematic construction of a single lesson;
(d) core teaching skills such as questioning, exposition, narration and illus-
tration; and,
(e) the personal power and presence of the teacher.

Robinson also shows that this growing interest in models of effective teaching was not
a UK monopoly, but drew upon an international base. She cites Claparede (1911), in
an early meta-analysis of new research in pedagogy and child psychology, who demon-
strated that similar experiments were occurring in the USA and the rest of Europe, and
that these were being shared through networking and conferences.
These conceptions of effective teaching were innovative for their time in that they
were based on experimentation, demonstration and observation, and were thus
research-based. They were also broad by comparison with what came later, as we argue
below. They focused on classroom performance, but performance that drew upon
strong subject knowledge for planning outside the classroom, and was realised through
the individual’s ‘power’, by which was meant not merely charismatic classroom pres-
ence, but the capacity and commitment to integrate the knowledge and skills into a
classroom practice that was dynamic and responsive to individuals and contexts, rather
than instrumental rule-following.
In this sense they resonate with very modern models, such as that underlying the
controversial Hay McBer model adopted by the Department for Education and Skills
(DfES) in England in 2000 (DfES, 2000), which is discussed below. In another sense,
the relative lack of consideration given to learning and its assessment, the models look
defective compared to modern models, but this apart, their main preoccupations,
subject knowledge, planning, classroom interaction, lesson structure, questioning and
explaining, and teacher behaviour have a very modern feel. They were fundamentally
concerned with identifying a set of rational principles upon which teaching could be
based, and in which teachers could be trained. The work was characterised by a real
optimism that the teaching profession itself could take control of the development of a
practical pedagogy, rooted in teacher-based research, observation and debate around
the nature of teaching.
This early reliance on empiricism flowered; over the next 70 years, researchers,
especially in the USA, the Netherlands and the UK, systematically investigated teacher
effectiveness using pupil learning outcomes as the significant indicator of effectiveness.
It is possible to construct a four-phase, largely chronological, loose classification of
these studies using the principal criteria being investigated in them, as below.

Presage-product model (1930s–1940s): in which the psychological characteristics of teach-


ers were identified and investigated for their effect on learning. The characteristics
included personality (e.g., authoritarianism) attitudes and experience (Rosenshine &
Furst, 1973; Borich, 1996).
Differential Teacher Effectiveness 349

Experimental studies (1940s–1960s): in which the effects of different teaching styles upon
learning were investigated (Medley, 1979). The styles included rather vaguely formu-
lated polarities such as formal and informal, progressive and traditional, open and
closed.

Process-product model (1960s–1980s): in which the behaviour of teachers in classrooms,


such as the quantity of instruction-focused interactions, and the pacing of instruction,
was the focus of investigations into factors influencing pupil attainment and progress
(Everston et al., 1980; Brophy & Good, 1986; Doyle, 1986; Galton, 1987).

Teacher knowledge and beliefs model (1990s–present): in which teachers’ subject knowl-
edge and pedagogical knowledge (Fennema & Loef-Franke, 1992), and their beliefs
such as self-efficacy or expectations (Dempo & Gibson, 1985; Anderson et al., 1988;
Askew et al., 1997) were investigated to explore the relationship between these factors
and pupil attainment and progress.
There are a number of qualifications to the above classification. The most obvious is
that the chronology is spuriously precise, since studies in one category often overlap in
period with another. For example, Shulman’s highly influential study of the role of
subject and pedagogical knowledge was published in 1987 (Shulman, 1987). However,
the chronological sequence in terms of the focus of the research, is broadly accurate.
Secondly, the categories themselves are impure. The influential ORACLE study in the
UK, for example, (Galton and Simon, 1980, re-run in 1999 (Galton et al., 1999)) could
be thought of as concerned both with teaching style and with teacher behaviour. The
important point however is that the category system identifies the principal factors, and
theoretical assumptions, under investigation in the studies. Thirdly, any classificatory
framework is to some extent subjective; Watkins and Mortimore (1999) for example, in
their review of pedagogy, also construct a four-stage analysis, namely, teaching styles,
classroom contexts, teaching and learning models, current views of pedagogy. Never-
theless, supporting our argument for a model reflecting differential effectiveness, their
framework, despite differences in content, drew attention in its fourth category to the
emerging interest in models of teaching and learning stressing differential effectiveness.
Such a model draws attention to the creation of classrooms as learning
communities in which knowledge is actively co-constructed… .This model of
pedagogy would … be increasingly differentiated by details of context, con-
tent, age and stage of learner, purposes and so on. (Watkins & Mortimore,
1999, p. 8)

PROBLEMS WITH THE CURRENT RESEARCH


There are five principal sources of difficulty with the current state of research in teacher
effectiveness, which some researchers themselves have been quick to recognise. The
first is that the conceptualisation of teacher effectiveness appears to have been, to some
extent understandably, largely driven by the available technicalities of measurement.
When robust personality measures became available teacher effectiveness researchers
attempted to investigate the effect of teacher personality (for example, whether authori-
tarian or not) on learning even though there was no a priori basis for believing that
personality was influential. When (admittedly less robust) measures of teaching style
350 Oxford Review of Education

(whether progressive or not) became available, teacher effectiveness researchers


attempted to investigate the effect of teacher style on learning, though in this case it is
possible to detect ideological fashion being the driver as much as the technology of
measurement. Both attempts were unable to show strong positive relationships to pupil
cognitive outcomes, though it is possible to interpret the recent revived interest in
‘preferred learning styles’, a typology originally developed by Kolb (1973), as having its
provenance in these failures, although the empirical evidence for these is again limited
so far. Similarly, now that sophisticated and mathematically complex techniques, such
as multi-level modelling and structural equation modelling are available, it is becoming
possible to analyse teacher effectiveness in relation to cross-level interactions between
pupil, classroom and school levels. This is not to deny the real advances in theory and
research methodology that arose from the development, at the end of the 1980s, of
good techniques for analysing hierarchical data about effectiveness, as Goldstein (1995)
argues. Nevertheless, reading some of the literature, it is difficult to avoid a sense of the
horse and the cart being in the wrong places; the technology of measurement has been
creating the concept of effectiveness rather than the concept requiring an appropriate
technology. It follows that current concepts of teacher effectiveness may be open to
question.
The second problem flows from the vagaries of education policy in which school,
rather than teacher, effectiveness captured the interest of policy makers across the world
in the 1980s and 1990s. Although there were a small number of highly influential
studies (eg Mortimore et al., 1986), the seduction of policy makers by school effective-
ness probably arose out of an attractive but romantic belief in the importance of school
leadership for raising standards in the system—a belief that has led to knighthoods and
damehoods for head teachers of schools in England judged to be particularly effective.
Yet school effectiveness theorists such as Creemers (1994) and organisational theorists
such as Caldwell and Spinks (1992) had agreed that, a priori, pupil performance was
more heavily dependent on classroom factors than school factors. The latter were
conceived of as probable conditions for teacher effectiveness. Teacher effectiveness
research has been both the casualty and the beneficiary of the belief in school
effectiveness; initially passed over for funding in favour of school effectiveness research,
it became high profile as the school effectiveness researchers, perhaps belatedly,
identified classroom factors as contributing the greatest amount to variance in pupil
outcome (see Scheerens & Bosker, 1997; Wright et al., 1997; Yair, 1997; Hextall &
Mahoney, 1998; Kyriakides et al., 2000; Teddlie & Reynolds, 2000). The logic of this,
that there could be effective teachers in ineffective schools and ineffective teachers in
effective schools, and that the relationship between school and teacher effectiveness is
problematic, has become more obvious in England, as appraisal of individual teacher
performance has been implemented. So, to knighthoods and damehoods have been
added the ‘Oscars’ for individual teachers.
It would be inappropriate to polarise. Much of the school effectiveness research can
be used to show how school-wide policies, consistently applied, support teachers in
areas shown to be connected to effective learning, such as maintaining high expecta-
tions, common rules of classroom and extra-classroom behaviour (see Reynolds et al.,
1996). What may be needed is a model which, while focused on the individual teacher,
reflects the context within which she is working.
Thirdly, there is the problem of teacher effectiveness research and teacher improve-
ment. In contrast to some of the research in the USA (for example Brophy & Good,
1986), research in the UK has been relatively indifferent to the implications for
Differential Teacher Effectiveness 351

improving teaching. The widely cited meta-analysis by Gipps (1996) What we Know
about Effective Primary Teaching is succinct and analytical, but does not show how a
teacher could move from ineffective to effective practice. From this and most of the
other research the best that could be inferred was a set of characteristics about the
effective teacher, but a teacher thought to be ineffective on the criteria used to define
effectiveness would not be helped to identify ways of improving. The explanation for
this does not lie in researchers’ lack of interest in improvement, so much as in the
research methodologies. These involve correlational studies based on variation in
existing practices. Thus even where there were attempts to transfer findings to class-
room practice (Brophy & Good, 1986) or to professional development programmes
such as the Active Mathematics Teaching (Good et al., 1983) or The Teacher
Effectiveness Enhancement Project (Muijs & Reynolds, 2000), prescriptions for appli-
cations usually remain within the ranges of teacher behaviour observed. Therefore,
until very recently, training programmes based on teacher effectiveness research were
not available. Some moves to connect training with research findings are now beginning
to emerge. For example, the text by Muijs and Reynolds (2001) for student teachers
uses teacher effectiveness research to draw implications for practice, and Gipps,
McCallum and Hargreaves (2000) have produced a text drawing implications from the
research for primary teaching.
An important point arises from the above approaches to researching teacher effective-
ness. They have been mainly good at distinguishing ineffective from averagely effective
teachers (with regard to pupil achievement in standardised tests) but not good at
distinguishing highly effective from averagely effective teachers. In this sense, one could
say that teacher effectiveness research has looked at ‘basic teaching skills’ rather than
‘higher order’ teaching skills.
Moreover they have not investigated leading edge or innovative teaching. It could be
argued that by concentrating on the basic, they have rendered mere competence
acceptable, and may even have had the unintended effect of discouraging outstanding
and inspirational teaching. Bennett’s (1976) study depicted ‘formal’ teachers as more
effective than ‘informal’ teachers. But one ‘informal’ teacher was later shown to be an
inspirational superteacher, whose pupils made more progress than any other class in the
study. Despite the methodological problems with the study itself, this remains a
cautionary tale for current effectiveness research.
Fourth, there is the narrowness of the operational definition of effectiveness, usually
restricted to the teacher’s classroom instructional behaviour and its association with
pupil cognitive outcomes. Clearly effective instruction is a major dimension of teaching,
but the work of teachers is substantially broader than classroom performance. Teaching
in modern societies has been analysed by a variety of supra-governmental agencies
(e.g. OECD, 1990; ILO, 1991). These show that under modernising tendencies
and as societies become more secular, schools become the main site of moral and
social value formation. Under these trends the role of teachers is expected to be
broad; affective, moral and welfare in orientation as well as cognitive. A model of
teacher effectiveness might therefore be required to incorporate measures of effective-
ness across these different roles rather than as now be limited to aspects of the
cognitive. Furthermore, empirical studies of teachers’ work (e.g. Campbell & Neill,
1994; PricewaterhouseCoopers, 2001) show that teachers typically spend less than half
their working time on classroom instruction, with a range of extra-classroom activities,
such as administrative and clerical tasks, lesson preparation, marking, report writing,
meetings, curriculum development, social and welfare tasks with pupils and parents,
352 Oxford Review of Education

school management and leadership roles, and professional development taking up the
remainder. Under reform initiatives, the significance of work outside classrooms in-
creases, as studies across the world have shown (e.g. Campbell & Neill, 1994; Carlgren,
1994; Tedesco, 1997; Day, 2000; Klette, 2000; Bajunid, 2000).
Even within the classroom, the measurement of teacher effectiveness ought in
principle to include how effectively the teacher manages and develops other adults,
including teaching assistants, technicians and other para-professionals, volunteers, and
pupil teachers, and how well she works in relation to pastoral matters. Cognitive gain
by pupils cannot stand as proxy for all these activities. In England, the 1998 Green
Paper (DfEE, 1998), proposing the ‘modernisation’ of the teaching force, envisaged a
significant shift in the teacher’s role from direct teaching to the management of other
teaching personnel and of technology. Despite reservations about the underlying
consequences of such a shift in terms of control and workforce flexibility (see Merson,
2000), an evaluation of teacher effectiveness concentrating exclusively on the teacher’s
ability to instruct classes directly, looks inappropriately narrow, if not anachronistic.
Fifth, there is the problem of generic effectiveness and differential effectiveness.
Much of the earlier research, and the policies for teacher appraisal that claim to be
connected to it, have tended to identify a general set of characteristics that define the
effective teacher; they provide a profile of teacher behaviour, knowledge and beliefs
assumed to be generic, irrespective of age of pupil, ability of pupil groups, social and
organisational context, or even the subject being taught. Effective teaching was thereby
constructed as a Platonic ideal, free of contextual realities. A move from a generic
model has begun for three reasons. In the first place, the generic model is obviously
counter-intuitive; an effective teacher teaching her specialist subject to a fast stream of
17-year-olds in a selective or prestigious private school, with ten pupils in the class,
would not be expected to exhibit the same classroom behaviour (except at such a level
of generality—show enthusiasm, say—as to be banal) as an effective pre-school teacher
with a class of 25 three-year-olds. Second, researchers in the UK and the USA have
drawn attention (e.g. Teddlie & Stringfield, 1993; Borich, 1996; Watkins & Mortimore,
1999; Harris 2001; Hopkins & Reynolds, 2001; Muijs & Reynolds, 2001) to the issue
of differential effectiveness. Third, the commissioning of separate research projects on
the effective teaching of literacy and numeracy respectively, by the Teacher Training
Agency, a UK government agency, appeared to recognise that the two subjects might
require different models of effectiveness. The outcomes of these two projects suggested
that effectiveness in numeracy might depend less strongly on subject knowledge than in
literacy (Askew et al., 1997; Medwell et al., 1998). In a similar way, but with a focus
on social class, American research has drawn attention to the different behaviours or
characteristics needed for effectiveness in different contexts. Teddlie and Stringfield
(1993) suggested that effective schools in different social class contexts displayed
different characteristics depending on the socio-economic context in which they oper-
ated. Likewise, Borich (1996) concluded that different classroom behaviour in low and
high social class contexts were needed for teachers to be effective. Thus any model of
teacher effectiveness needs to incorporate what Hopkins and Reynolds (2001) call
‘context specificity’.
Two of these difficulties seem to be a particularly English problem. Lack of interest
in improving teaching has not generally been associated with the USA teacher effective-
ness research, where some of the most significant studies (Rosenshine, 1971; Good et
al.,1983; Brophy & Good, 1986; Griffin & Barnes, 1986; Borich, 1996) led directly into
training programmes or incorporated applied programmes in their research design.
Differential Teacher Effectiveness 353

Equally important, the development of a cumulative set of findings about effective


teaching over 30 years in the USA has not been undermined by interest in school
effectiveness. According to Gage (1996), one of the leading researchers across the
period,
meta-analysis has yielded knowledge concerning the impressive magnitude,
consistency, and validity … for quantifying and analysing the generalizability
of research results … continuing the effort to build sciences of behav-
iour… (p. 5)
It is right to be sceptical about the extent of achievement of this effort. Labaree (1998)
in a detailed analysis of the status and nature of educational knowledge production in
American schools of education, noted that educational research ‘is well short of the
condition Gage calls a science of educational behavior’. Nevertheless, the state of play
in the US in 1986, when the ground breaking Handbook of Research on Teaching
(Wittrock, 1986) was published for the American Education Research Association, was
much closer to producing research-based principles of teaching effectiveness than was
the case until very recently in the UK. Most of the teaching methods and classroom
organisational arrangements required for England’s national literacy and numeracy
initiatives in 1999—direct interactive instruction, establishing classroom rules, clarify-
ing objectives for the session with pupils, reviewing work done against objectives—are
traceable to, and could be justified by, the studies by Brophy and Good, Rosenshine
and Stephens, and Doyle in that handbook. (See the literature review prepared for the
UK Government’s Numeracy Task Force, Reynolds & Muijs, 1999.)
There are three further issues, which arise principally from technical matters rather
than from conceptual problems. They have been raised in school effectiveness research,
but could have a bearing on how any new conceptualisation of teacher effectiveness
might be developed. They are, first, that the current models, using generic measures of
school effectiveness, have identified a considerable contribution to variance in pupil
achievement at the classroom level which is not able to be explained. A differential
effectiveness model might be able to explain much of this variance if it can be shown
to arise from the spread of achievement in a class. Second, the school effectiveness
movement has accepted the value, or potential value, of school self-evaluation
(MacBeath, 1999) in the process of measuring school effectiveness, but teacher effec-
tiveness research has so far largely ignored the role of self-evaluation. Kyriakides et al.
(2003) show that self-evaluation has the potential to contribute to improvement of both
the individual teacher and the school as a collectivity. Third, pupil cognitive outcomes
have been limited to performance mainly in standardised basic tests and therefore
researchers have eschewed opportunities for investigating the effect that teachers have
on the process of learning. This might be an important task if research were to focus
on innovative and higher order teaching, as suggested above.

TOWARDS A MODEL OF DIFFERENTIATED TEACHER EFFECTIVENESS:


FIVE POSSIBLE DIMENSIONS OF DIFFERENCE
A new model of teacher effectiveness might reasonably be expected to start from a
definition of effectiveness. Such a definition has been notably missing from the
literature. This may be explained by the fact that a definition is likely to tend to the
tautological, or to be self-evident to the point of banality. The Shorter Oxford English
Dictionary (SOED) defines effectiveness as ‘the quality of being effective’ which is not
354 Oxford Review of Education

terribly helpful, especially as effective is defined as ‘concerned with, or having the


function of, effecting’. The use of the term as an idea in research is much closer to
the SOED’s definition of efficacious which is ‘that produces, or is certain to produce,
the intended effect’. Similarly effectiveness as understood in the research is close to
efficacy, defined as ‘the capacity to produce effects; power to effect the object in-
tended’. Thus we would propose a definition of teacher effectiveness as follows: the
power to realise socially valued objectives agreed for teachers’ work, especially, but not
exclusively, the work concerned with enabling pupils to learn. This is not perfect, but
the adoption of a working definition should reduce the problems that arise from not
having one made explicit. An assumption we attach to the definition is that there must
be some means of measuring the extent to which the objectives have been realised,
though these need not be restricted to pupil test outcomes. In particular, adopting a
definition along these lines should help to question the tendency of some school and
teacher effectiveness research to treat pupil outcomes, usually restricted to a narrow
selection of test items in mathematics or the native language, as unproblematic.
Similarly, the incorporation of ‘socially valued’ reflects the fact that teachers are
employed for publicly approved purposes, whilst simultaneously raising questions
about the values underlying the objectives towards which teachers work. This may be
particularly important given the tendency of teacher effectiveness researchers to side-
step questions of value assumption. (See Richards, 2001, for a discussion of values
underlying the UK’s Office for Standards in Education (OfSTED) inspection methods
for evaluating effective teaching.)
Using that definition, we see five possible dimensions of differential effectiveness.
These refer to: differences in activity; differences in subjects and/or components of
subjects; differences in pupils’ background factors; differences in pupils’ personal
characteristics; differences in cultural and organisational context.
As has been argued above, the range of activities expected of teachers in modernising
education systems is extremely broad. These include social, pastoral, welfare dimen-
sions and management of other adults, in addition to the formal instructional dimen-
sion. Accepting that the sampling of these kinds of teacher function is desirable for
an appropriate approach to defining teacher effectiveness, what is required is a multi-
dimensional concept of teacher effectiveness. This would include effectiveness in
achieving, for example, pastoral goals, in the management of other adults to improve
learning, and so on.
Second, there is the issue of differential effectiveness across different subjects in the
curriculum, or across different components. Evidence from inspections conducted by
the Office for Standards in Education (OfSTED) in England at primary school level is
beginning to demonstrate that the same teacher can be judged to be outstanding at, say,
teaching mathematics but only adequate at, say, teaching history or physical education.
This is because the inspections now provide a profile of teacher performance in relation
to each lesson observed (typically up to four lessons in an inspection). The lessons are
commonly of different subjects. The data are logged onto a national database and may
be made available to researchers, whilst data at the level of the individual school are
made available to governing bodies of the school. Although there is as yet little evidence
about different performance in different subject components (algebra as against num-
ber in mathematics for example), because the inspection data are not presented at that
level of detail, there is no logical reason not to believe that a teacher can perform
differently in different subject components. This analysis is not restricted to primary
schools; secondary teachers often teach subjects other than their specialism, especially
Differential Teacher Effectiveness 355

to the younger classes (see OfSTED, 2001), and the argument about subject compo-
nents applies equally to them as to their primary colleagues. Thus a differential model
would need to include a consistency dimension across the instructional role.
Third, teachers may be differentially effective in promoting the cognitive progress of
different groups of pupils according to background variables. The principal ones are
ability, age (or developmental stage), sex, socio-economic status and ethnicity. For
example a teacher might be extremely effective in promoting the learning of pupils with
special educational needs but less so with very able pupils, or vice versa. Any model of
differential effectiveness would need to be able to identify such strengths and enable the
interactions among these variables to be examined.
Fourth, teachers may be differentially effective in promoting the learning of pupils
according to the pupils’ personal characteristics, such as their personality, cognitive
learning style, and extent of motivation, and self-esteem.
Fifth, teachers may be differentially effective in response to the various cultural and
organisational contexts in which they work. For example, effectiveness in a two teacher
rural primary school may be different from in a two thousand pupil urban school or
college; in different departments or faculties in the same school; with several para-
professionals to manage, compared with none; in homogeneous, compared with hetero-
geneous, classroom groups; in schools with strongly framed cultures, compared with
schools with weakly framed cultures. This would require a model in which the
inter-relationships between school context and teacher effectiveness were reflected. At
the moment the recognition of this issue has been reflected in the adoption of the term
‘educational effectiveness’ (Creemers, 1994) but its use may serve to fudge matters by
avoiding the necessity to tease out the interactions, which in many contexts are
extremely complex.

TWO FUNCTIONS FOR A DIFFERENTIAL EFFECTIVENESS MODEL


We see the development of a differential teacher effectiveness model along the lines
indicated above as serving two different functions. The first would be to guide research
design in the future, in order to have a more sophisticated and appropriate knowledge
base about teacher effectiveness. This would assume that a concept of teacher effective-
ness defined more broadly than at present could find acceptance among researchers and
their funding sources. If so, the available techniques, such as structural equation
modelling, are capable of examining not merely the five dimensions of difference
outlined above but also the interactions within and across them. It should be possible,
in principle at least, to investigate such relationships as those between pastoral and
academic effectiveness, between performance in particular aspects of the curriculum
and teachers’ subject knowledge, between teachers of pupils from different social,
cultural, linguistic and ethnic backgrounds, or of different gender. It should also
provide demonstrable distinctions between teachers who are averagely effective and
those who are outstanding at higher order teaching skills. Perhaps most important of
all, the patterns of relationship between teacher effectiveness and the school and
departmental contexts can be explored. Less ambitiously, research might be expected
to describe differences in the effectiveness behaviour traits of teachers of different age
groups, from pre-school through primary and secondary schools, to university and
beyond in life long learning and occupational training. None of the above should be
seen as ruling out the potential of a differentiated model for building more general
356 Oxford Review of Education

theoretical models identifying factors associated with effective teaching in any subject,
any context, etc.
The second function would be to provide a more equitable and appropriate model for
the evaluation of teachers’ performance; and this could be used to appraise teachers’
performances across a spectrum of work more broad than the current focus on
classroom instruction. If the five dimensions were accepted as a basis for the concept
of teacher effectiveness by which performance could be evaluated, a number of
advantages would in principle follow, and we identify four main ones below.
First a relatively full range of teacher performance would become available for
appraisal, and the fuller the range of sampled behaviour the more robust one could
assume the appraisal would be.
Second, the widespread objection to a narrowly instrumental set of criteria from the
profession and academics would be able to be partly addressed. The profession’s
objections in England (see ATL, 1999; NUT, 1999) tended to call attention to
the divisive nature of rewards based only on instructional performance, and to the
absence of values in the criteria, despite the moral nature of education. However,
part of the criticism, for example the analysis developed by Lawn (1995), may be
directed at any form of differentiation of rewards based on performance, as part of an
egalitarianism that is strong in the teaching profession, and especially the unions
(Katzenmeyer & Moller, 2001). This objection could not be dealt with by the model we
propose.
Academic objections have been strong in both the USA and the UK (see
Apple, 1986; Sultana, 1994; Ball, 1998; Osborne et al., 2000) and have criticised the
‘performativity’ ethic on the grounds that it renders instrumental efficiency as
more important than individual development, and undermines the idea of service. Since
these critics see the origins of performativity in the changing economic and political
values of globalising societies overall, it is difficult to argue that a different model of
performance measurement could address their objections, but what we propose would
at least allow for values, and performance less instrumentally defined, to be taken into
account.
Third, of particular interest in respect of the above objections, is the opportunity the
model provides for teacher self-evaluation as part of the process, as argued earlier. A
five-dimensional model enables a strong qualitative element to be incorporated into the
appraisal, either at the school level, where a school staff could identify and give different
weighting to particular dimensions, or to elements within each. Likewise, an individual
teacher would be able to self-evaluate, identifying those elements to which she would
wish particular emphasis to be given.
Finally, a profile drawn from the appraisal would identify development needs
beyond the classroom as well as within it. This would be particularly important as
the modernisation of the profession proceeds, since much of the work of teachers
is envisaged as managing others, developing leadership skills, and identifying the
learning needs of particular groups of pupils. Moreover, it would be possible to
take into account the effect of the school management and structures on individual
teachers’ performance, leading to an appraisal process in which accountability can
be facing two ways: from the individual teacher to perform duties effectively, and
from the school management to support the individual teacher. This would be a
significant change and could be used to develop a more equitable symmetry in
accountability.
Differential Teacher Effectiveness 357

DISCUSSION
There are four principal problems with what has been proposed. The first is technical,
and arises from the complexity of the model needed to reflect the nature of teachers’
work. For the purposes of research design, this is not an insoluble problem, although
it poses challenging questions about methods of data collection. For purposes of
performance appraisal, however, the form of implementation will be critical; the
complexity threatens to move the mechanisms of measurement and evidence collection
to the nightmarishly bureaucratic. It is not the aim of this article to develop workable
mechanisms for teacher appraisal, but the difficulty could be overcome by sampling
teacher performance within each dimension, especially if self-evaluation is part of the
initial process.
The second problem is conceptual and arises from the spread of activity included in
the idea of teacher effectiveness. There is some evidence from studies of pupil expecta-
tions of teachers, and from parents’ expectations also, that clear instruction, fairly
delivered, is the main expectation laid upon teachers (Brown & McIntyre, 1993;
Cooper & McIntyre, 1996). Any model which did not weight instructional performance
significantly more heavily than other aspects would run the risk of opposition from
these sources, even when mediated through governing bodies. Yet a significant weight-
ing, to the extent of overwhelming the importance of other dimensions, would render
the rationale for the new model redundant in practice. This would be less of a problem
if empirically it were to turn out to be the case that teachers effective in the instructional
dimension were also effective in others—but we do not know whether this is so. Again
this is a more significant problem for teacher appraisal than for research, where the
models could incorporate different weightings or none, and the main objective would
be to establish whether there was correlation between the various dimensions, and the
interactions between them.
Third, as we have indicated earlier, there is the problem of values. Teaching is a
normative, not a neutral, activity, and there is evidence that it is a dilemma-laden
occupation (Berlak & Berlak, 1981). There is no reason to suppose that effective, or
highly effective, teachers are somehow excused from facing difficult value choices. The
problem is more acute in teacher appraisal than in research. Some of the values
underlying effectiveness, for example, holding high expectations for pupils, having
strong commitment to their subject, commitment to treating pupils with fairness, are
unlikely to be contested. Others, however, are less clear cut. For example, there may
well be a tension between collaboration with colleagues and performing effectively
independently of colleagues, especially if teacher appraisal is focused exclusively on
individual performance. Even more problematically, at a superficial level, the value of
respecting others seems uncontroversial, but not all colleagues, pupils, or parents turn
out to deserve respect. Moreover, some important values, such as commitment to
enabling all pupils to realise their potential, either are not measurable or could only be
measured in time scales that are too long for purposes of appraisal. Added to these
difficulties, there would be understandable unease about measuring and rewarding the
extent of teachers’ adherence to a particular set of values, however much they are
socially approved.
The fourth problem is political, and also concerns the form of implementation of the
model; it is by no means sure that policy makers would welcome a more complex basis
for the appraisal of teachers’ performance along differentiated lines. For policy makers
there is virtue in simplicity. And a Platonic ideal is simple, even though it neglects the
358 Oxford Review of Education

complexity of modern teaching as an occupation. To adopt a concept of effectiveness


which implied that any generic measure of effectiveness would have to be adjusted in
the light of performance with different groups of pupils, say, or of organisational
context, would force policy makers to confront this complexity. Moreover, a simple
concept lends itself to public representation of teacher performance in apparently
simple accountability terms. A more complex concept would require a form of account-
ability less easily understood by lay consumers of education, and would make it less
easy for politicians to demonstrate claims for improvement in teachers’ performance
nationally. However, we would argue that complex realities have to be incorporated
into policy implementation, not wished away.
These difficulties are real, and should not be underestimated, but they are not
insurmountable. There is already in place a model of teacher effectiveness that contains
some of the features for which we are arguing. The Hay McBer model (DfES, 2000)
(www.dfes.gov.uk/teachingreforms/leadership/mcber/01.shtml), which has been
adopted by the English government as the basis for teacher appraisal, has three
dimensions in which are incorporated professional characteristics, teaching skills and
classroom climate. Under professional characteristics, there is a ‘professionalism clus-
ter’ in which values such as respect for others, challenge and support, confidence,
and creating trust feature. Characteristics such as leading others, analytical and concep-
tual thinking, initiative and a passion for learning are also incorporated. How far these
values have been reflected in the implementation of teacher appraisal in practice
is unclear as yet. Moreover, the origins of this model, and the empirical claims about
the extent to which it reflects real effectiveness have been questioned by members of
the British Educational Research Association, primarily because it has been difficult
to subject the original research data and findings to scrutiny, let alone replication
(see BERA, 2001). This is partly because, as a private company, Hay McBer
exercised its legitimate right to retain confidentiality on grounds of commercial interest,
reflecting one disadvantage in commissioning the private sector to conduct research
and development on public policy. Nonetheless, it illustrates that in principle, both
policy-makers and the profession can envisage and could adopt a complex, values-
based model.
We began this article by referring to Robinson’s work on models of teacher effective-
ness developed by pioneering innovators 80 years ago. A key feature was their commit-
ment to conducting, admittedly embryonic, research—experiments in pedagogy,
observation, demonstration—upon which the models could be based. Although
significant methodological advances have been made since then, the principle of
empirically developed models remains very important to defend in a period when quick
solutions for teacher appraisal are sought by politicians. However, the key problem
remains conceptual; to generate models that seek, as did the early pioneers, general,
possibly universal, principles for teacher effectiveness, whilst identifying and incorpo-
rating appropriate dimensions of differential effectiveness. The pioneers used the
concept of ‘power to teach’, to embody the ability of the teacher to respond to
individual and contextual differences. Given the development of modern education
systems in performance cultures, we need a more sophisticated form of differentiation
than theirs, and our argument is that differentiation in teacher effectiveness represents
the next stage for concept and methodological development by researchers. Otherwise,
education systems will be left with the kind of unaccountable obscurity represented by
some aspects of the Hay McBer model; a model developed in a short time frame and
in a regime of commercial confidentiality; adopted as policy with considerable concep-
Differential Teacher Effectiveness 359

tual confusion and contradiction unresolved; and focused on generic performance to


the neglect of differential performance.

REFERENCES
ANDERSON, R., GREEN, M. & LOEWEN, P. (1988) Relationships among teachers and
pupils’ thinking skills, sense of efficacy and pupil achievement, Alberta Journal of
Educational Research, 17, pp. 86–95.
APPLE, M. (1986) Teachers and Texts: a political economy of class and gender (London,
Routledge).
ASKEW, M., RHODES, V., BROWN, M., WILLIAM, D. & JOHNSON, D. (1997) Effective
Teachers of Numeracy: report of a study carried out for the Teacher Training Agency
(London, Kings College London School of Education).
ATL (ASSOCIATION OF TEACHERS AND LECTURERS) (1999) Meeting the Challenge of
Change, Comment (London, ATL).
BAJUNID, I. (2000) Rethinking the work of teachers and school leaders in an age of
change, in: C. DAY, A. FERNANDEZ, T.E. HAUGUE & J. MOLLER (Eds) The Life and
Work of Teachers (London, Falmer).
BALL, S. (1998) Performativity and fragmentation in ‘postmodern schooling’, in: J.
CARTER (Ed.) Postmodernity and the Fragmentation of Welfare (London, Routledge).
BENNETT, S.N. (1976) Teaching Styles and Pupil Progress (London, Open Books).
BERA (BRITISH EDUCATIONAL RESEARCH ASSOCIATION) (2001) Report on the method-
ological seminar on Hay McBer enquiry into teacher effectiveness, Research Intelli-
gence, 76, pp. 5–9.
BERLAK, A. & BERLAK, H. (1981) Dilemmas of Schooling: teaching and social change
(London, Methuen).
BORICH, G.D. (1996) Effective Teaching Methods, 3rd edition (New York, Macmillan).
BROPHY, J. & GOOD, T.L. (1986) Teacher behaviour and pupil achievement, in: M.C.
WITTROCK (Ed.) Handbook of Research on Teaching (New York, Macmillan).
BROWN, S. & MCINTYRE, D. (1993) Making Sense of Teaching (Buckingham, Open
University Press).
CALDWELL, J.B. & SPINKS, M.J. (1992) The Self-managing School (London, Falmer
Press).
CAMPBELL R.J. & NEILL, S.R. STJ. (1994) Primary Teachers at Work (London, Rout-
ledge).
CARLGREN. I. (1994) Curriculum as social compromise or accident, in: D. KALLOS &
S. LINDBLAD (Eds) New Policy Contexts for Education, Educational Report 42
(Umea University, Sweden).
CLAPAREDE, E. (1911) Psychologie de l’enfant et pedagogie experimentale, Experimental
Pedagogy and the Psychology of the Child, trans. from 4th edition (London, Edward
Arnold).
COOPER, P. & MCINTYRE, D. (1996) Effective Teaching and Learning: teachers’ and
pupils’ perspectives (Buckingham, Open University Press).
CREEMERS, B.P.M. (1994) The Effective Classroom (London, Cassell).
DAY, C. (2000) Stories of change and professional development: the costs of commit-
ment, in: C. DAY, A. FERNANDEZ, T.E. HAUGUE & J. MOLLER (Eds) The Life and
Work of Teachers (London, Falmer).
DEMPO, M. and GIBSON, S. (1985), Teachers’ sense of efficacy: an important factor in
school achievement, The Elementary School Journal, 86, pp. 173–184.
360 Oxford Review of Education

DFEE (1998) Green Paper, Cm 4164, Teachers: facing the challenge of change (London,
Stationery Office).
DFES (2000) A Model of Teacher Effectiveness, Report by Hay McBer to the Department
for Education and Employment. www.dfes.gov.uk/teachingreforms/leadership/
mcber/01.shtml
DOYLE, W. (1986) Classroom organization and management, in: M.C. WITTROCK
(Ed.) Handbook of Research on Teaching (New York, Macmillan).
EVERSTON, C.M., ANDERSON, C., ANDERSON, L., & BROPHY, J. (1980) Relationships
between classroom behaviour and pupil outcomes in junior high math and English
classes, American Educational Research Journal, 17, pp. 43–60.
FENNEMA, E. & LOEF FRANKE, M. (1992) Teachers’ knowledge and its impact, in: D.A.
GROUWS (Ed.) Handbook of Research on Mathematics Teaching and Learning (New
York, Macmillan).
GAGE, N.L. (1996) Confronting counsels of despair for the behavioral sciences,
Educational Researcher, 25, 3, pp. 5–15.
GALTON, M. & SIMON B. (1980) Progress and Performance in the Primary Classroom
(London, Routledge).
GALTON, M. (1987) An ORACLE chronicle: a decade of classroom research, Teaching
and Teacher Education, 3, 4, pp. 299–313.
GALTON, M., HARGREAVES, L., COMBER, C., WALL, D. WITH PELL, A. (1999) Inside the
Primary Classroom, 20 Years on (London, Routledge).
GIPPS, C. (1996) What we Know about Effective Primary Teaching (London, Routledge).
GIPPS, C., MCCALLUM, B. & HARGREAVES, E. (2000) What Makes a Good Primary
Teacher? Expert Classroom Strategies (London, RoutledgeFalmer).
GOLDSTEIN, H. (1995) Multi-level Statistical Models (London, Arnold).
GOOD, T.L., GROUWS, D.A. & EBMEIER, D. (1983) Active Mathematics Teaching (New
York, Longman).
GRIFFIN, G.A. & BARNES, S. (1986) Using research findings to change school and
classroom practice: results of an experimental study, American Educational Research
Journal, 23, 4, pp. 572–586.
HARRIS, A. (2001) Department improvement and school improvement: a missing link?
British Educational Research Journal, 27, 4, pp. 477–486.
HEXTALL, I. & MAHONY, P. (1998) Effective Teachers, Effective Schools (London, Bid-
dles).
HOPKINS, D. & REYNOLDS, D. (2001) The past, present and future of school improve-
ment:towards the third age, British Educational Research Journal, 27, 4, pp. 459–
476.
ILO (INTERNATIONAL LABOUR ORGANISATION) (1991) Teachers: challenges of the 1990s.
Second joint meeting on Conditions of Work of Teachers (Geneva, ILO).
KATZENMEYER, M. & MOLLER, G. (2001) Awakening the Sleeping Giant: helping teachers
develop as leaders (Thousand Oaks, CA, Corwin).
KLETTE, K. (2000) Working time blues: how Norwegian teachers experience restruc-
turing in education, in: C. DAY, A. FERNANDEZ, T.E. HAUGUE & J. MOLLER (Eds)
The Life and Work of Teachers (London, Falmer).
KOLB, D. (1973) Toward a Typology of Learning Styles and Learning Environments: an
investigation of the impact of learning styles and discipline demands on the academic
performance, social adaptation and career choices of MIT seniors (Cambridge MA,
MIT Press).
Differential Teacher Effectiveness 361
KYRIAKIDES, L., CAMPBELL, R.J. & GAGATSIS, A. (2000) The significance of the class-
room effect in primary schools: an application of Creemers’ comprehensive model
of educational effectiveness, School Effectiveness and School Improvement, 11, 4,
pp. 501–529.
KYRIAKIDES, L., CAMPBELL, R.J. & CHRISTOFIDOU, E. (2003) Generating criteria for
measuring teacher effectiveness through a self-evaluation approach: a complemen-
tary way of measuring teacher effectiveness, School Effectiveness and School Improve-
ment, 13, 3, pp. 291–325.
LABAREE, D.F. (1998) Educational researchers: living with a lesser form of knowledge,
Educational Researcher, 27, 8, pp. 4–12.
LAWN, M. (1995) Restructuring teaching in the USA and England: moving towards the
differentiated, flexible teacher, Journal of Education Policy, 10, 4, pp. 347–360.
MACBEATH, J. (1999) Schools Must Speak for Themselves: the case for school self-evaluation
(London, Routledge).
MEDLEY, D. (1979) The effectiveness of teachers, in: P. PETERSON & H. WALBERG
(Eds) Research on Teaching: concepts, findings and implications (Berkeley, CA,
McCutchan).
MEDWELL, J., POULSON, L. & WRAY, D. (1998) Effective Teachers of Literacy: a report of
a research project commisssioned by the Teacher Training Agency (Exeter, University of
Exeter School of Education).
MERSON M.W. (2000) Teachers and the myth of modernisation, British Journal of
Educational Studies, 48, 2, pp. 155–169.
MORTIMORE, P., SAMMONS, P., STOLL, L., LEWIS, D. & ECOB, R. (1986) School Matters
(Wells, Open Books).
MUIJS, D. & REYNOLDS D. (2000) Effective Mathematics Teaching: Year 2 of a
Research Project. Paper given at the International Congress for School Effectiveness
and School Improvement, Hong Kong, January.
MUIJS, D. & REYNOLDS, D. (2001) Effective Teaching: evidence and practice (London,
Sage).
NUT (NATIONAL UNION OF TEACHERS) (1999) Teaching at the Threshold (London,
NUT).
OECD (1990) The Teacher Today: tasks, conditions, policies (Paris, OECD).
OFSTED (2001) Standards and Quality in Education, Annual Report of Her Majesty’s
Chief Inspector of Schools (London, OfSTED).
OSBORNE, M., MCNESS, E. & BROADFOOT, P., WITH POLLARD, A. & TRIGGS P. (2000)
What Teachers Do (London and New York, Continuum Books).
PRICEWATERHOUSECOOPERS (2001) Teacher Workload Study, www.teachernet.gov.uk/
remodelling
REYNOLDS, D., SAMMONS, P., STOLL, L., BARBER, M. & HILLMAN, J. (1996) School
effectiveness and school improvement in the United Kingdom, School Effectiveness
and School Improvement, 7, 2, pp. 133–158.
REYNOLDS, D. & MUIJS, R.D. (1999) The effective teaching of mathematics: a review
of research, School Leadership and Management, 19, 3, pp. 273–288.
RICHARDS, C. (2001) School Inspection: a critique (Northampton, Philosophy of Edu-
cation Society).
ROBINSON, W. (forthcoming 2004) Power to Teach (London, Woburn Press).
ROSENSHINE, B. (1971) Teaching Behaviours and Pupil Achievement (London,
NFER).
ROSENSHINE, B. (1979) Content, time and direct instruction, in: P.L. PETERSON & H.J.
WALBERG (Eds) Research on Teaching (Berkeley, CA, McCutchan).
362 Oxford Review of Education

ROSENSHINE, B. (1987) Direct Instruction, in: M. DUNKIN (Ed.) Teaching and Teacher
Education (Oxford, Pergamon).
ROSENSHINE, B. & FURST, N. (1973) The use of direct observation to study teaching,
in: R.M.W. TRAVERS (Ed.) Second Handbook of Research on Teaching (Chicago, IL,
Rand McNally).
ROSENSHINE, B. & STEVENS, R. (1986) Teaching functions, in M.C. WITTROCK (Ed.)
Handbook of Research on Teaching (New York, Macmillan).
SCHEERENS, J. & BOSKER, R. (1997) The Foundations of Educational Effectiveness (Ox-
ford, Pergamon).
SHULMAN, L. (1987) Knowledge and teaching: foundations of the new reform, Harvard
Educational Review, 57, 1, pp. 1–22.
SULTANA, R. (1994) Conceptualising teachers’ work in a uniting Europe, Compare, 24,
2, pp. 171–182.
TEDDLIE, C. & REYNOLDS, D. (2000) The International Handbook of School Effectiveness
Research (London, Falmer).
TEDDLIE, C. & STRINGFIELD, S. (1993) Schools Make a Difference: lessons learned from a
ten year study of school effects (New York, Teachers College Press).
TEDESCO, J.C. (1997) Enhancing the role of teachers, in: C. DAY, D. VAN VEEN, D. &
S. WONG-KOOI (Eds) Teachers and Teaching: international perspectives on school
reform and teacher education (Leuven and Apeldoorn, Garant).
WATKINS, C. & MORTIMORE, P. (1999) Pedagogy: what do we know? in: P. MORTI-
MORE (Ed) Understanding Pedagogy and its Impact on Learning (London, Paul
Chapman).
WITTROCK, M.C. (1986) (Ed.) Handbook of Research on Teaching (New York, Macmil-
lan).
WRIGHT, S.P., HORN, S.P. & SANDERS, W.L. (1997) Teacher and classroom context
effects on pupil achievement: implications for teacher evaluation, Journal of Person-
nel Evaluation in Education, 11, pp. 57–67.
YAIR, G. (1997) When classrooms matter: implications of between-classroom varia-
bility for educational policy in Israel, Assessment in Education, 4, 2, pp. 225–248.

Correspondence: Professor R. J. Campbell, University of Warwick Institute of Education,


Coventry CV4 7AL, UK. E-mail: r.j.campbell@warwick.ac.uk

You might also like