Teacher Evaluation and School Improvement
by Milbrey Wallin McLaughlin - 1984
Teacher evaluation can be used as a tool in school improvement. Principal inconstancy, insufficient time and resources, and lack of evaluating skills are some difficulties in teacher evalution. The use of evaluation as a means of professional development is discussed. (Source: ERIC)
Policymakers at all levels of government are pressured to respond to critical conclusions about the status of American education1 and to escalating public demand to do something about the schools. Questions of education quality, consequently, are on the agenda of most state legislatures and local school boards2 and the debate about solutions is heated. Teacher evaluation, this article argues, can be a powerful strategy for achieving these school-improvement goals.
Practitioners may find this position surprising if not wrong-headed. Many doubt that teacher evaluation can serve both accountability objectives and improvement concerns. Indeed, many practitioners have divorced improvement and assessment purposes in teacher evaluation so that staff-development activities will not be seen as punitive. But it is also true that few districts have actively pursued links between teacher evaluation and improvement. Most educators see current teacher evaluation practice as a waste of time and resources. In a majority of school districts, teacher evaluation constitutes an uneven, desultory ritual that contributes little to school improvement but much to teacher anxiety and administrator burden.3
As teacher evaluation typically is conceived and practiced, it could be little more. Most teacher evaluations comprise standard checklists completed by the principal after a brief classroom observation. Principals usually base ratings on their own sense of good practice; not surprisingly, assessments based in the I know what I like school of evaluation can vary among schools and classrooms. Evaluation in this instance reflects a principals individual preferences rather than a consistent set of criteria to inform either accountability or improvement.
However, principal inconstancy is less problematic than it might be because most teachers receive satisfactory or outstanding ratings; needs improvement or unsatisfactory findings are rare. Administrators explain the preponderance of these salutory assessments in terms of the political and bureaucratic problems associated with teacher evaluation. Low ratings risk conflict with the teachers organization; evaluators do not have the skills to confidently do more; support from downtown is often not forthcoming in the event of a negative or controversial appraisal; insufficient time and resources are available to respond to less than satisfactory ratings anyway.
Teachers are no happier with the current state of assessment practice. Strong teachers complain that this system does not acknowledge excellence, provides feedback too general to be useful, and fails to document incompetence. Weaker teachers are also dissatisfied. They express resentment that hoped-for diagnosis and assistance is not a result of their evaluation and that their satisfactory marks are meaningless.
Teachers assert that a checklist approach to evaluation, especially one grounded in a process-product model that assumes specific teacher behaviors lead to particular learner outcomes, is an irrelevant and inappropriate evaluation tool. A number of serious concerns are raised about this deterministic approach to teacher evaluation. Among them:
1. Learner outcomes are cumulative; it is difficult to isolate the effect of any one teacher on student performance.
2. Teacher behaviors and activities interact with numerous factors to affect student performance. Student socioeconomic status, school climate, pupil abilities, previous instructional treatment, are but a few of the many factors that determine teacher effectiveness for any given student.4 Teacher effectiveness, however defined, is highly contextual and conditional.
3. Teachers vary enormously in the practices that work for them and the problems they confront in their particular classrooms.5 As Good, a longtime student of teacher effectiveness, put it: One myth that has been discredited by classroom observation is that schooling is a constant experience with teachers behaving in similar ways and pursuing similar goals with a common curriculum.6 No single instructional program works for all teachers or all students; effectiveness depends on the classroom context. Thus there can be no single, simple method of evaluating teacher effectiveness because there is no single concept of what the teacher should be undertaking in the classroom.7
4. Teachers effectiveness varies depending on the goals defined for the student or the class. Not only are the objectives described for students multiple and substantively diverse (e.g., academic, emotional, or social outcomes) but the strategies successful in achieving one goal (memorization of facts, for example) are often counterproductive for other instructional objectives (e.g., higher-order problem-solving skills).8 Further, the effectiveness of particular teacher practices may be curvilinear: Too much of a good thing can depress outcomes.9
Yet most teacher evaluation activities, with their closed-ended checklists, prescriptive categories, and ambiguous standards, disregard this complexity. The incompetence of principals as teacher evaluators compounds instrumentation problems. Teachers seldom respect principals as experts on classroom practice or as skilled classroom observers,10 and in the absence of principal credibility, teachers consider the evaluation illegitimate comment on their performance and ignore its findings. Given the state of teacher evaluation practices in most districts, then, misgivings about the ability of teacher evaluation to contribute to school improvement are unsurprising. As an essentially bureaucratic mechanism, present teacher evaluation practices can inform neither practice nor policy in a meaningful way.
But theory and experience suggest that teacher evaluation of another stripe can support teacher growth and development, strengthen the role of the principal, and contribute significantly to the vitality and coherence of the school.11 A number of school districts have adopted teacher evaluation practices based in principals strengthened supervisory, diagnostic, and prescriptive skills. In districts that are moving away from the deterministic, process-product model of teacher evaluation, principals are trained to observe classroom practices, assess teacher solutions to classroom problems, gauge the quality of teacher-student interactions, and analyze the structure of instructional processes. Principal training framed in this model acknowledges the conditional nature of teacher effectiveness and focuses on individual teacher judgments and choices within broad and widely held categories for effective teaching.12
TEACHER EVALUATION AND IMPROVEMENT
Experience shows that teacher evaluation based in this process perspective supports the formal authority of the principal as evaluator with functional authority based in technical knowledge, evaluation skills, and shared language. Teacher evaluation grounded in this design is a potent tool for school improvement because it can affect factors that are fundamental to how teachers and principals go about their jobs and how well they carry out their responsibilities for instruction and management.13 Most important are:
teacher motivation and sense of efficacy
effective communication and shared goals
principals instructional leadership
teacher learning and development
TEACHER MOTIVATION AND SENSE OF EFFICACY
It is axiomatic that teachers motivation and their sense of professional effectiveness are central to school-improvement efforts and to maintaining high-quality classroom practices. Teachers sense of efficacy is tied to an educators primary source of satisfaction, the intrinsic rewards associated with the teaching roleservice to youngsters or transmitting knowledge associated with a particular discipline.14 The extrinsic rewards attached to a teaching career are low; the ancillary benefits (with the exception of a long summer vacation) are effectively nonexistent.
Yet it is difficult for teachers to collect the intrinsic rewards that motivate them and provide satisfaction. The greatest obstacle to teacher sense of efficacy, ironically, is lack of feedback about their performance-credible information about how well they are carrying out their responsibilities. To this point, significant and recurrent doubt about the worth of their work with students is a consistent teacher characteristic.15 Efficacy, as this suggests, is not entirely an internal construct; it relies on environmental response that acknowledges good performance.16 A number of factors frustrate teacher participation in what Lortie calls craft pride or efficacy.17 The structure of the profession itself makes it difficult for teachers to experience professional accomplishment. Most professions ideally are characterized by explicit career progress, not by a static position of competent practice.18 Teaching, however, is a relatively flat occupation, with few of the stages or plateaus that mark accomplishment and success in other professionsmedicine or law, for example. Thus there is little in the structure of the profession to tell teachers they are doing a good job.
Another impediment to teacher sense of efficacy is inherent in the teaching task. Unlike other areas of professional or semiprofessional activity, there is no agreed-upon technical core of knowledge or unambiguous set of guidelines for successful practice.19 Furthermore, outcomes for teachers are relatively complex, ambiguous, indeterminate, and long-run. A lawyer can judge success by case outcome; an agronomist can measure achievement by the number and type of new agricultural techniques in place. Teachers, however, have no such unequivocal or unitary measure. Student achievement scores, the outcome measure favored by school boards, citizens, and policymakers, are not seen as adequate measures of effectiveness by teachers.20 Classroom effectiveness, in the teachers view, rests in the successful diagnosis of classroom problems and the selection of strategies to meet them and in producing long-term changes in youngsters attitudes and capacity. As one particularly acid commentator on the process-product school of teacher effectiveness research put it: Teachers are not hired to cram information into students heads to be retained just long enough to enable them to pass objective tests. Teachers are hired to educate children, to produce important, lasting changes in their behavior, not short-term changes in test scores.21 Successful teaching outcomes, in this view, are as indeterminate as the practice itself.
Ironically, then, while self-reflection lies at the heart of professionalism,22 self-monitoring and assessment are difficult for teachers to carry out. There is no template for success that teachers can lay beside their performance and assess the extent to which they have achieved their personal and professional goals. And long-term outcomes may never be evident to teachers. Consequently, teachers must rely on the reflection and feedback of others to gauge their effectiveness and support professional pride. For this feedback to be credible, it must come from individuals who teachers believe can make authoritative judgments about their performance.
The norms and the process of schooling preclude those individuals most able to provide that feedbackfellow teachersfrom doing so. The cellular structure of the school and the isolation of teachers in their classrooms is much remarked upon.23 Teachers have little opportunity to observe their peers, to compare classroom practices, or to comment on collegial practices. Instead, time spent with colleagues during the school day typically is perceived as stolen.24 But even if opportunities for peer observation were increased, the norms dominant in most school settings prohibit collegial assessment. Conventions of teacher autonomy join with norms of collegial support to make peer criticism unprofessional. So strong is this ideology of noninterference that even when teachers know about bad practices, they will make no move either to assist a colleague or inform responsible adminstrators.25
The principal, in his or her role as evaluator, thus has a crucial role to play in providing the credible feedback essential to a teachers sense of efficacy. Regular classroom observations, based in principal-evaluator classroom expertise and observational skills, can provide the review and diagnosis essential to teacher satisfaction, efficacy, and growth. Far from perceiving visits from a competent principal-evaluator as a threat or a waste of time, teachers view them as professionally and personally rewarding. To this point, districts studied as part of the preliminary research for Rands teacher evaluation study provide strong evidence of teacher support for this principal role. A teacher in one New Jersey district, inadvertently excluded from evaluation visits on two occasions, somewhat playfully filed a grievance for being overlooked. Teachers in a Minnesota district with a strong, diagnostically based teacher evaluation system voted to continue funds for teacher evaluation as a high priority when the districts budget was trimmed. In Washington state, teachers amended their collective bargaining agreement to include more and unannounced principal visits; they felt that principals were not seeing what was really going on and so the feedback to teacherspositive and negativewas less useful than it might be. In districts such as these, teachers have come to value their evaluations as an important source of information about their performance and primary support for their sense of efficacy and so for their professionalism.
COMMUNICATION AND SHARED GOALS
Open, frequent, and candid communication among teachers and school administrators is characteristic of effective schools26 and a factor in successful planned change activities.27 Effective communication is two-way and includes significant emphasis on instructional, not just administrative, matters.
Communication of this nature is not easy to achieve and is not part of the normal character of information transmission within school buildings. Bureaucratic pressures encourage one-way telegraphic communication rather than conversation between teachers and administrators. Even when the occasion for exchange presents itself, communication is less effective than it might be because teachers and administrators lack common language. An important result of principal training in clinical supervision is acquisition of this common language.28 Such training permits principal-evaluators to speak clearly, precisely, and very specifically to teachers about their performance, to interpret classroom events, and to analyze teaching practices. Principals are thus able to move beyond global statements about teacher performance (Keep up the good work! More discipline is needed during seat work.) to discuss particular concepts of classroom practice and provide teachers concrete examples gathered by observation (e.g., pointing out that a teacher spends most of her time teaching to the right side of the classroom). As one teacher, commenting on her principals supervisory and evaluation expertise, put it: It puts words on problems as well as strengths. I have a clear notion of what needs to be improved and [concerning her own documented improvement] I really dont think all of this would have been possible without evaluation.29
Providing principals with the skills to make classroom observations of this diagnostic nature and communicate findings in ways that teachers can relate directly to classroom practice supports communication in the other direction as well. Teaching staff in schools where principals possess this expertise report that they talk much more with principals about classroom issues because shared language makes such a conversation possible. Judith Warren Little noted this phenomenon in her study of school success and staff development: [Only administrator observation of classroom practices] and feedback can provide the shared referents for the shared language of teaching, and both demand and provide the precision and concreteness which makes talk about teaching useful.30 Teacher evaluation rooted in administrator observation and diagnostic skill, in short, provides both the language and the content of the communication associated with effective schools and improved practiceconcrete talk about instruction and strategies for improvement.
This kind of evaluation allows principals to inform teachers regularly about school-wide goals and to assess teacher performance in terms of these goals. Thus teacher evaluation can support the mutual understanding between teachers and administrators that is necessary to combat the segmented and sometimes incompatible practices seen in many schools. At the same time, it can provide the information to enable teachers and administrators to align instructional content, classroom activities, and instructional goals. This strategy thus moves both knowledge and practice of a schools professional staff toward the shared goals crucial to school improvement and effectiveness.31
The central and essential function of the principal in school improvement and school effectiveness as instructional leader or gatekeeper of change has become a truism. A substantial body of research focuses on the activities associated with this pivotal rolefor example, identifying and supporting good classroom practices, integrating school-wide instructional activities, and keeping activities of the professional staff apace of changing school needs, district priorities, and knowledge about more effective practice.32 Teacher evaluation can be the key to principal leadership because it provides the occasion, structure, and information that support these activities, namely, regular teacher observation, discussion about teacher work problems, and assistance to teachers through regular feedback and analysis.33
Teacher evaluation also allows principals to exercise effective control over the quality of classroom practice because it appeals to teacher incentives for improvement and growth and assists in the counseling out of less effective teachers. Schools are normative organizations and teaching is a craft in which excellence relies heavily on commitment, enthusiasm, and the desire to do ones best. Coercion and punitive oversight are not effective strategies for promoting excellence in teaching or school improvement broadly defined.34 Indeed, experience suggests that heavy-handed accountability measures can actually make things worse in the classroom because teachers do not see the outcome measures typically employednamely, student achievement scoresas legitimate or the process sufficiently sensitive to the complex process of teaching. The result, too often, is bitterness at the system, frustration, and decisions to give up goals of excellence and instead do just enough to get by. However, a teacher evaluation system that furnishes specific, detailed, and believable information about classroom performance can engage teacher commitment to growth and enthusiasm for learning new skills.
The same information that motivates teachers to grow professionally can also increase the quality of educational services in a school through the counseling out of teachers who appear ill-suited to teaching and unlikely to profit from in-service education opportunities. In the face of detailed, concrete information that points to performance problems, and given adequate remediation opportunities, most teachers who continue to have difficulty in the classroom are amenable to suggestions that they seek another vocation. The same norm of service that encourages teachers to gain new skills in the light of documented problems, it appears, supports decisions to resign when personal lack of fit with the profession can be demonstrated. To this point, one Lake Washington, Washington, principal who counseled out seven teachers in the past five years commented that with only one exception, they all left with a smile.35
TEACHER LEARNING AND DEVELOPMENT
Teacher evaluation that describes and diagnoses teacher practices in specific, concrete terms can provide the most effective and legitimate means of quality control because it appeals to internalized norms of professionalism and points the way to do better. Teacher evaluation of this stripe relies on the normative power of legitimate authority and informed feedback to stimulate teacher development and change.
Such teacher evaluation not only is compatible with notions of effective authority in the school organization, but is also compatible with what we know about how adults learn. Unlike children, adults seldom learn simply because someone tells them to. Indeed, demands to learn new skills, particularly where they involve replacement of existing routine, threaten an adults already well-organized self-concept and established level of accomplishment. Adult motivation to learn new things must come from within. Teacher evaluation has an important role to play in stimulating this internal motivation. To this point, Brundage concluded on the basis of a comprehensive review of adult learning that what seems most clear in discussions on motivation is that the tendencies which are labeled motives arise from within the learner. These are not something added on by an external agent. . . . The behavior of the external agent must be viewed as contributing either to feedback or to reinforcement and by this route indirectly to further motivation.36 Similarly, Knowles says that adults are motivated to learn as they experience needs and interests that learning will satisfy.37 Concrete information about areas in which teaching practice can be improved furnishes precisely the most powerful kind of motivation for teachers authoritative and legitimate feedback on ways to be a more effective teacher.
The salience of teacher evaluation in this role is amplified by what we know about how teachers learn to teach. Teachers learn to teach primarily in two ways, as students and on the job; preservice teacher education programs play a weak role in teacher development.38 Teacher learning requirements are developmental. As Nemser details, first-year teachers engage in formative skill development; it is only after teachers master fundamental teaching skills that they begin to concentrate on the relationship between what they do and student behavior.39 This means that on-the-job learning is most significant to teacher performance and that support of this learning, as well as assessment of performance, must be keyed to a teachers developmental stage. Experience has shown that unitary or uniform staff-development activities too often are too little too late. A strong teacher evaluation program is essential to the identification of differentiated strategies of diagnosis and assistance that can support teacher development. Given the centrality of on-the-job learning for teachers, teacher evaluation may be one of the most potent teacher development strategies available.
CONDITIONS NECESSARY TO EFFECTIVE TEACHER EVALUATION
Teacher evaluation is not something most school principals like to do. For one thing, they have little confidence in their ability to carry out fair, consistent, and meaningful evaluation of teachers classroom performance. Second, teacher evaluation and the associated anxiety threaten the stability in the school and encapsulate the tension school administrators feel between their roles as instructional leader and building manager. Nowhere is the potential incompatibility between these two roles more apparent than in teacher evaluation. Principals tend to minimize conflict in this area by minimizing teacher evaluation.
Finally, teacher evaluation is but one of the multiple demands on a building administrators time and energy. Indeed, in terms of urgency, two broad classes of concerns eclipse teacher evaluation and performance issues: relations with parents and the community and student discipline. Until these issues central to administrator control are resolved, administrators are not inclined to turn to questions of classroom quality and teacher performance.40 Given all of these factors, it is not surprising that principals tend to spend little time on evaluation (approximately 5 percent of their time) and that the assessment of teacher performance is largely pro forma and cursory. Teacher evaluation, in short, is an activity that most principals have little interest in or capacity to carry out.
Business as usual conditions cannot promote and support teacher evaluation practices of the type discussed here. Rands teacher evaluation study points to at least five conditions essential to a teacher evaluation program that can contribute substantially to school improvement:
extensive and regular training for principals
resources for evaluators
teacher participation in program design
explicit central office support and involvement
integration with other district management activities41
TRAINING FOR PRINCIPALS
In most districts, principals receive little if any training related to their teacher evaluation responsibilities. For teacher evaluation of the type assumed here, principal training is substantial and ongoing. A weekend workshop as the program is getting underway is insufficient to give principals the requisite clinical, diagnostic, and staff-development skills.
Training of this sort requires substantial initial investment; equally as important, there must be continued attention to refreshing, refining, and building on the diagnostic skills of principals. In Lake Washington, for example, district administrators attend a two-week workshop each August. Teacher evaluation and teaching processes are always a focus of these workshops. Through simulation, role-modeling, videotapes, and other devices, administrators receive extensive and increasingly sophisticated training in clinical observation, notetaking, reporting, and conference skills. In addition to these yearly training retreats, follow-up administrator development seminars are held at least once a month. Training principals to carry out this role, in short, is not something that is finished; rather it is an ongoing, iterative activity.
In most school districts, principals have responsibility for evaluation but do not have the authority or the resources to act on their findings. This lack of effective responsibility both undercuts building principals and undermines the credibility of the evaluation activity. When principals lack the authority or resources to respond to problems identified in the process of evaluation, the activity becomes little more than a time-consuming but empty exercise. Further, the teacher frustration and alienation that can be expected to result from evaluation without appropriate follow-up may be counterproductive to improvement goals.
Resources for principals to use in response to evaluation findings are crucial if principals are to take teacher evaluation seriously and if evaluation is to support teacher improvement. For one thing, evaluators must be able to respond quickly in order to make the tie between evaluation and improvement an effective one. Teacher motivation to respond to evaluator assessment will be highest immediately following an evaluation session and the nature of improvement concerns will be freshly defined. Resources for evaluator use are also important from the perspective of the most effective support for teacher learning because evaluators can tailor a teacher-development prescription.
Districts handle this requirement for decentralized and nonstandardized resources in different ways. In Lake Washington, for example, each school has a discretionary fund that principals can use to support the in-service education activities suggested by a teachers evaluation-special workshops, a course at the nearby university, enrollment in a district in-service activity, released time for observation in another setting, and so on. Salt Lake City has a remediation team composed of central office specialists and especially identified consultants who work with teachers identified as having difficulty. Other districts use mentor teachers or teachers on special assignment to respond individually and immediately to principals request for assistance in a classroom. Evaluator resources such as these are necessary to a teacher evaluation effort that serves school improvement rather than merely accountability rituals.
The effective teacher evaluation practices examined in the Rand study included teachers in the development of district teacher evaluation practices.42 Teachers and administrators agreed that teacher participation was a necessary ingredient in the success of the program. Teacher involvement is important for a number of reasons. One of the most salient is the fact that teachers can maximize the transitive rewards of teaching only if they have played a role in specifying the criteria and strategies used in assessing their performance and that of their students.43
The Rand study found that teacher participation was important to building the trust between administrators and teachers necessary for the system to work; it also provides concrete evidence that the district did not intend to implement a gotcha system of teacher assessment and that improvement is a mutual goal.44 If an evaluation system is to serve teacher-improvement objectives, it is essential that teachers see it as equitable and relevant. Teacher participation is a necessary means to that end. Finally, teacher participation in design is crucial to teacher commitment to do something about evaluation outcomes.
EXPLICIT DISTRICT-LEVEL SUPPORT
Teacher evaluation is not something to which building administrators would devote substantial time or attention, all other things being equal. It conflicts with their facilitative and supportive role: it consumes already inadequate time. Express district-level commitment is essential to a strong and consequential teacher evaluation program. Principals and teachers must see teacher evaluation as a district priority and something that is taken seriously by the superintendent and central office staff. Without this support, evaluation will remain a pro forma, bureaucratic responsibility.
Central office support can be shown in a number of ways. Support will be evident, of course, in the resources made available for principals to respond to teacher performance assessment. Less tangible elements of support are required as well. Active central office oversight of principals evaluation activities conveys a strong signal about the priority afforded evaluation and the attention it should receive. To this end, some districts review principals evaluation reports for care and comprehensiveness. Indeed, in many districts where teacher evaluation is unusually effective, principals are evaluated on the quality of their evaluation. Some districts even attach sanctions to teacher evaluations in an effort to focus principals attention on the issue and halt what one teacher called the dance of the lemons. In one California district, for example, principals are penalized at salary time if a teacher they rated as competent proves incompetent when transferred to another school.45
Political support from downtown is critical. Sometimes principals will not act on observed teacher problems because they fear the political fallout. Decisions about teacher probation are inherently political; in making this recommendation, a principal risks problems with the teachers organization as well as parent or community members who may believe that a teacher has been judged wrongly. And many principals have found little if any support from downtown if a probationary placement became a heated issue. Principal confidence that the superintendent and central office staff will be supportive on tough decisions and will not, as one principal put it, leave us out on a limb while they back off for political reasons is essential to a strong teacher evaluation system.46
In short, if a district wants a strong teacher evaluation system that can contribute to improvement goals, it must demand it, support it with multiple resources, and give it the political and bureaucratic backing it requires.
INTEGRATION WITH DISTRICT MANAGEMENT ACTIVITIES
A strong and meaningful teacher evaluation system demonstrates substantive integration and strategic consistency with other district management activities. The development of technical knowledge is relatively useless in the absence of organizational structures and processes to use it. In most districts, teacher evaluation is rendered effectively inconsequential by its isolation from other district management activities. For example, Rands preliminary assessment of teacher evaluation practices across the country found that these activities typically had no connection with district planning, staff development, curriculum development, or program evaluation activities. Where teacher evaluation was effective as a school-improvement strategy, however, there were explicit interrelationships among these district activitieseach informing and reinforcing the other through common goals, expectations, and processes. In this way, teacher evaluation is a central part of an administrators responsibilities, not just a categorical and ancillary requirement.
The necessity of substantive and strategic consistency with district management practices draws attention to the fact that there is no best model of teacher evaluation. To this point, the unusually effective teacher evaluation systems examined as part of the Rand study differed along every possible design dimensionthe role of the teacher, the role of the principal, the timing and nature of the evaluation process, the resources available to evaluators, and the criteria established for teacher performance. While each of these four systems offers important lessons to inform choices in other districts, their effectiveness reflects the fact that they fit the districts particular management style and tenor.
In summary, teacher evaluation can be a potent school-improvement tool not because it puts a floor under classroom practicesthe goal of accountability-based evaluation modelsbut because it addresses the incentives central to individual development and the teachers sense of professionalism. Evaluation when seen in this light cannot be subjected to the quick fix, but requires the interaction of a host of factors that build on the norms and values central to the teaching profession.