Part I
Performance-related pay
1
Performance-related pay
Introduction
The concept of âperformanceâ does not always commend itself to teachers when applied to the job they do. It is a term more often associated with sport, industrial production, stage and arena appearances by actors or musicians, circus animals even. The word suggests that anyone giving a performance is probably in the limelight, showing off perhaps, not a notion with which quieter teachers readily identify. It also carries negative connotations, as in âWhat a performanceâŚâ, meaning a long-winded rigmarole, an unnecessary waste of time, a poor show. In an industrial context there is an assumption that performance can be measured fairly accurately, in terms of output, profits, the production of inert commodities. These factors can make the word seem too cold and dispassionate a description of the essentially human activity that teaching is thought to be.
The same reservations attach to associated terminology, such as âperformance indicatorsâ, favoured by economists and accountants in their quest for confirmation that something is working well, or that it can be regarded as a wise and effective investment of money. Teachers who take a holistic view of their work feel uneasy when it is atomised into separate elements to be measured, weighed and then ticked off, one at a time. The feeling is not unlike that of the German author Goethe, who described in a poem how he tried to discover why a dragonfly was so beautiful, only to find that it became a crumpled, dismembered heap in his hand. The secret of its beauty had been life itself.
In these circumstances the phrase âperformance-related payâ can arouse a similar negative reaction. Those who dislike their work being seen as a âperformanceâ are hardly likely to embrace enthusiastically the practice of offering teachers cash rewards according to their perceived quality. On the other hand, the idea of paying people more money, if they are thought to be doing their job especially well, has a commonsense appeal to the general public. Some heads and teachers are in favour of both principle and practice, as this research will show, arguing that it is just and proper to pay more to those who are working harder. Indeed, making additional payments to certain teachers, but not others, has been a feature of the salary and reward structure for the profession in many countries, but the word âperformanceâ has not always figured in the transactions, as they may have been paid extra for seniority, or for particular duties and responsibilities.
This book reports the findings of a three year research project, the Teachersâ Incentive Pay Project (TIPP), funded by the Leverhulme Trust at the University of Exeter. The principal aim of the study was to investigate, at school, regional and national level, what happened when the United Kingdom introduced one of the largest and most widely spread performance-related pay schemes ever devised. All primary and secondary schools in England and Wales had to select those teachers who were regarded as the most competent and reward them with additional salary payments. Such a mass introduction of cash rewards, based on performance-related criteria, was novel in the UK, so we were in a unique position to study its genesis and effects, as the project began before the scheme was launched.
In order to âimproveâ what they do, people must change. Nobody improves by staying exactly the same. Yet in the Leverhulme Appraisal Project (Wragg et al. 1996), we found that fewer than half the teachers studied said they had changed what they did as a result of appraisal. Classroom observations endorsed this finding. It was going to be interesting to discover whether the awarding of cash appeared to produce different results.
In the past there had been different types of incentive pay for the teaching profession in England and Wales, usually in the form of extra payment for additional responsibilities or duties. The introduction, in 1998, of the post of Advanced Skills Teacher, signalled a new style of government incentive to teachers to stay in the classroom, rather than seek extra pay outside it. The nationwide scheme launched in 2000 was a huge step beyond that, for it involved the vast majority of teachers, not just a select few.
The research programme studied a large national stratified random sample of primary and secondary schools, using a mixture of quantitative and qualitative methods, and undertook a smaller set of intensive case studies. The project fitted in well with several of our previous large-scale studies of a similar kind in recent years, looking at classroom skills (Wragg 1993), teacher appraisal (Wragg et al. 1996), factors influencing the effective teaching of literacy (Wragg et al. 1998) and teachers alleged to be incompetent (Wragg et al. 2000).
Background
How to improve the pay of classroom teachers in primary and secondary schools was part of a problem that faced the government in the late 1990s, when recruitment and retention had become an important matter. It was already known that almost half of the 400,000 teachers working in 24,000 primary and secondary schools would be aged over fifty by the middle of the first decade of the twenty-first century. Large numbers of teachers were leaving the profession before the age of retirement, making retention at least as serious an issue as recruitment. Part of the challenge was to find ways of preventing teachers having to become headteachers, or take other posts outside the classroom, in order to earn a higher salary.
In the United States Furtwengler (1994) identified three major promotional strategies or âladdersâ for classroom teachers. These were:
1. Performance-based ladders: Progression up a scale based on evidence of professional competence.
2. Job enlargement ladders: Progression based on extra responsibilities.
3. Professional development ladders: Progression based on completion of staff development activities, coursework, advanced degrees.
Traditionally schools in the United Kingdom had made use of Strategy 2. This was usually carried out through the payment of specified sums of money to teachers given additional responsibilities, like acting as head of department or subject co-ordinator, taking charge of special educational needs, or running the school library. Gradually, however, the emphasis had moved much more towards Strategy 1, with certain teachers regarded as especially competent being awarded a higher salary. The initiative begun in 1998, introducing the post of Advanced Skills Teacher, allowed schools to pay certain teachers a higher salary if they were deemed to be particularly good at their job and were willing to advise and support other teachers in the same or a different school.
The major Strategy 1 type of initiative to be sponsored by the government, however, was the introduction of a full-scale system of performance-related pay which was to be widely applied. Initially it was thought that perhaps half of classroom teachers would be paid higher salaries. In a 2000 document entitled Performance Management Framework (DfEE 2000b), the Department for Education and Employment proposed that there should be annual appraisals for teachers covering both pupil progress and personal development, with three to six targets being identified and regular monitoring of teachersâ progress during each year. These annual reviews were then, in the words of the document, to âinformâ the awarding of extra salary by school governors.
From September 2000 schools were required to draw up a âfair and openâ policy involving all their teachers. Priorities and objectives had to be agreed, and between three and six âchallengingâ but âflexibleâ targets identified for each teacher, covering pupil progress and personal development. At the end of the year, teachersâ performance was to be assessed and a decision would then be taken about whether they should progress further up the salary scale.
The research programme
The main general aim of the research was to focus on schools in England, during the critical three year period when the new systems were being introduced and implemented, and to find out, by observation, questionnaires and interviews, what different primary and secondary schools actually did to implement performance-related pay: how they drew up procedures and set targets; how they monitored them; what happened in lessons and what teachers appeared to do to meet their targets, or whether there seemed to be little change; which factors were thought to facilitate and which to hinder implementation; how decisions about pay were made in practice, including a specific analysis of the part played by pupil performance data; and finally what heads and teachers thought about the process and the outcomes in retrospect, in the light of their own experiences.
Within the main broad aim of the research there were several specific objectives. These included eliciting the processes and outcomes in schools as experienced directly by the people involved, like heads, teachers, union officials and local authority officers. The design, over the three-year period, employed a mixture of methods. Two large-scale questionnaire surveys of over 1,000 headteachers would explore their views and experiences at early and later stages of the implementation process, while intensive case studies over a two or three year period would assemble the picture in individual schools and classrooms. Thus the approach overall was to use both a âlong shotâ and âzoomâ technique. A large national sample of schools responding to questionnaires gives the broader picture, while an in-depth study of schools and individual heads and teachers offers a set of intimate local insights and thus provides the local/national comparison framework.
There were many research questions. How would decisions be made in different schools? What sort of training would assessors receive? Would there be any significant differences in practice between the primary and secondary sectors? Would teachers change their styles and strategies? How would pupil âprogressâ be assessed? What would be the impact on teachers who were successful, as well as on those who were unsuccessful? Would personal relationships within schools be affected? What would be the role of the external assessor, whose responsibility it was to check that headteachers were acting according to the principles and practices laid down in the procedures?
The measurement of pupil learning was one of the most controversial parts of the governmentâs policy. A number of heads, teachers and their union representatives voiced opposition to this kind of evidence even being part of the process. The 2000 performance management framework document sent to schools, however, was not quite the crude mechanical âpayment by resultsâ model that had been anticipated by some critics. It stated that it was important for teachersâ targets to be âbased on pupilsâ prior attainmentâ. The document also said that team leaders, when making their assessment, should address âfactors outside a teacherâs control [that] may affect the achievement of objectivesâ. This modified âvalue addedâ approach is capable of many different interpretations, so that is why one objective of the research was to explore how pupil data were used.
The methodology followed closely that employed in our previous project, described below. Questionnaires were piloted extensively on samples similar to the constituency being polled. Interviews were recorded and transcribed whenever necessary. All four researchers have observed hundreds of lessons and trained together to high levels of agreement on quantified methodology, while on qualitative data a ârate until agreeâ approach was used to resolve different interpretations.
Anonymity was guaranteed to everyone. Questionnaires were filled in anonymously, which may increase the likelihood of frank responses, but removes the opportunity to carry out a follow-up with the same sample. Headteachers were not told which members of staff were being studied and teachers did not know what their heads had said in interview. All names of individuals and schools in this book are invented, to protect the people concerned.
Part of the research strategy was to consider practice in other countries and indeed in other professions that make use of performance-related pay, as some have a long tradition of using it. The General Electric Company in the United States conducted a seminal study of its effects as early as the 1960s (Meyer et al. 1965), concluding that, among other findings, it was important to separate in time the act of appraisal and the awarding of a cash bonus or reward, otherwise the one interfered with the other. We looked at certain schemes in industry and the professions that appeared relevant to teaching, and reviewed practice internationally. In the United States, for example, several states have tried it, though some have discontinued it. Furtwengler (1994) documented the response of state governors after 1984, when Lamarr Alexander, governor of Alabama, declared that no teacher in the United States had ever been paid a cent more for teaching well. Despite many promises to the contrary, most governors did little beyond setting up a committee or a pilot study. By 1995 only five states were still running largescale schemes, though others had changed their system to one where schools rather than teachers received additional cash.
In addition to the direct study of an incentive and reward scheme there were several important underlying issues, many of which were to do with the dynamics within a school, or the way that power was exercised, either by those who wielded it officially, or by those who wrested it. In our study of the competence of several hundred teachers (Wragg et al. 2000) we found that power and control over decisions could fluctuate from one party to another as time progressed and circumstances changed. Matters like decisionmaking about a group of teachers by their managers lie at the very heart of what happens in a professional community. The United Kingdom was introducing one of the most ambitious and far-reaching schemes of performancerelated pay ever conceived in a climate of some reservation among teachersâ unions and individual teachers and heads. It was an innovation that seemed likely to be of worldwide interest.
Teaching and teacher âeffectivenessâ
The debate about teacher âeffectivenessâ is central to the performance pay issue. Teaching can be an extraordinarily busy job, with thousands of interpersonal transactions taking place inside a week and as little as one second between them in which the teacher must make a decision (Wragg 1999). As a result teachers lay down deep structures, favoured styles and strategies which they can produce rapidly and modify according to the context in which they find themselves. Teachers with twenty yearsâ experience have probably asked over a million questions, given several million pieces of information and explanation, and offered praise or reprimand on thousands of occasions. If one purpose of performance-related pay is to change behaviour, then there are many repeats and rehearsals to unscramble, the equivalent of trying to unpick the grooved swing of a seasoned professional golfer.
The notion of âeffectivenessâ is by no means a clear and uncontested matter. Different criteria or measures may produce different candidates. In the nineteenth century, teacher training institutions were known as ânormal schoolsâ, on the grounds that there was a single agreed ânormâ of teaching (Rich 1933). The pluralism of the twentieth century led to a diversity of teaching styles and there was no clear research evidence that one approach was universally more effective than another, since context, subject matter, pupil ability and prior experience will often influence both process and outcome. Indeed large-scale summaries of the relationship between process and product in classrooms over many decades have often stated precisely that.
Barr (1961), summarising several hundred American studies, concluded that teachers who were preferred by administrators or by pupils were not necessarily those whose pupils did well on tests. Doyle (1978) observed that reviewers of research into teacher effectiveness had concluded, with remarkable regularity, that few consistent relationships between teacher variables and effectiveness criteria could be established. Even reviewers of the same studies have sometimes reached different conclusions about them, as Giaconia and Hedges (1985) pointed out in their synthesis of research findings on effectiveness.
A number of usually smaller scale analyses have sometimes found advantages for one style of teaching over another. Sometimes such findings can contradict each other. For example, Gage (1978 and 1985) cited studies that tended to show a superiority of âtraditionalâ over âprogressiveâ methods so far as basic skills in the early years of schooling were concerned, while Kulik et al. (1979) reported a meta-analysis of American studies of the Keller Plan, a form ...