(retirado de Kenneth D. Peterson, Ph.D., Portland State University, Portland OR)
Myth 1: The central purpose of teacher evaluation is to improve teachers and teaching. The truth is that there is scarce research to suggest that evaluation causes teacher growth. Rather, teachers will improve if you give them enough TIME to work on good ideas: uninterrupted time with students, time to plan and implement what is already known, and sufficient discretionary time to be full human beings. There are other very good reasons to evaluate: to document current good practice, reassure teachers of a needed and effective job, reassure audiences, identify good teaching practices for emulation, and prevent bad evaluation practices.
Myth 2: Better teacher evaluation is just a better rating instrument or framework of teacher behaviors. The truth is that educators do not agree on what should be included in any single catalog of teacher performances or competencies, none could encompass all of what the open-ended nature of teaching should have, teachers are effective using different sets of small numbers of behaviors, and teachers work in varied contexts which call for different competency sets. Comprehensive frameworks, descriptions, systems analysis, and lists of duties (e.g., Danielson, 1996; Heath & Nelson, 1974; Scriven, 1988) help build understanding of good teaching, but they don't cause good evaluation.
Myth 3: Excellent teaching is accomplished by strong performance of 22 (or 27 or 60) components of teaching. Rather, a good teacher performs three or four components extremely well, adequately performs some others, and (to be honest) poorly or spottily performs many other things that a teacher is "supposed" to do. Doing a few things well at the moment carries the entire performances of teaching and learning; the other possible performances simply don't matter at the given time in the real human world of a classroom. It is a misleading strategy to try to assess every possible component, duty, competency, or element of a teacher performance at a point in time in order to understand the overall quality of that teaching.
Myth 4: Specific a priori goals (unique to individual or from a general framework) are needed to evaluate a teacher. Rather, good teaching can be documented after the teaching has been done by highlighting the actual specific outcomes, performances, or preparations that played a role in that specific teacher performance.
Myth 5: A uniform system of teacher evaluation is essential: all teachers should be evaluated the same way. The reality is that teachers are good for different constellations of reasons. They work in quite different settings, with different kinds of demands and criteria for quality. Also, we just cannot get all the information we might want for each instance of teacher evaluation. Fairness demands that all teachers have an equal opportunity to document their quality in the ways most appropriate to them.
Myth 6: Pupil achievement data cannot be used in teacher evaluation, or they can be used for all teachers. Rather, we can get good pupil achievement data for some but not all teachers in a district; and the teacher evaluation system should reflect the state-of-the-art of data availability.
Myth 7: Teacher quality can be objectively measured and known by using a sufficiently accurate checklist and rating scheme, or by comparing pupil achievement test scores. Rather, all evaluation is subjective. However, there is good subjectivity and bad subjectivity. Good subjectivity is (a) based on the best objective evidence available, (b) controlled for individual bias, (c) involves the interested audiences, and (d) employs some public logic.
(o que está a negrito são destaques da minha responsabilidade)