Can we Evaluate #Teachers Without Using High-Stakes #Testing?
Late in the summer I received an email from Larry Ferlazzo, an award-winning English and Social Studies teacher at Luther Burbank High School here in Sacramento, asking if I was interested in contributing to the conversation about the evaluation of teachers for his Edweek Classroom Q&A blog. He asked several other folks across the nation also, and he recently released a three part series on teacher evaluation. The contributions in the first post Teacher Evaluations Need to ‘Support, Not Sort’ were authored by American Federation of Teachers President Randi Weingarten, California Teachers Association President Dean Vogel, and 2012 National Teacher Of The Year Rebecca Mieliwocki. The second post Using Teacher Evaluations ‘to Promote Growth’ featured contributions from myself (with Lisa Hernandez), Ben Spielberg, David Berliner and Paul Bruno. The final post in the series ‘Getting What You Pay For’ In Teacher Evaluations featured commentaries from W. James Popham, Barnett Berry, Pia Lindquist Wong, Rick Stiggins and Derek Cabrera, along with thoughts from Edweek readers.
You can also listen to a ten-minute conversation I had with Larry Ferlazzo and Ben Spielberg about evaluating teachers on Ferlazzo’s BAM! Radio Show. Click and listen— it’s interesting and it goes by quick, I promise.
Here is my response on Ferlazzo’s Classroom Q&A blog:
Some school “reformers” are determined to allocate a number to every aspect of education. No Child Left Behind accentuated the current infatuation with quantitative data in educational policy. Arne Duncan continued the quantitative love affair via teacher evaluation requirements in Race to the Top and other initiatives. As a result, many school districts throughout the United States are under pressure to meet various teacher evaluation requirements that exist due to local, state, and federally implemented policies. However, there are empirically tested qualitative alternatives to the currentquantitatively based teacher accountability and evaluation systems. These alternatives can access the expertise of master teachers instead of a number cruncher sitting in front of a computer screen.
Peer Assistance and Review (PAR) was developed in the early 1970s in Toledo, OH. Harriet Sanfordwrote that Peer Assistance and Review (PAR) is a program of structured mentorship, observation and rigorous, standards-based evaluation of teachers by teachers, and has been demonstrated to be among the strongest ways to develop great teachers. What is PAR and how does it work? In PAR, the local teacher organization and district administrators jointly manage the teacher evaluation program to improve teacher quality by developing a structure where expert teachers mentor and evaluate their peers.
The Harvard Project on the Next Generation of Teachers posited,
“PAR challenges most people’s expectations about what teachers and principals should do. It requires unusual collaboration between the union and administration. It must be grounded in a systematic approach to teacher evaluation… Increasingly, policymakers, district officials, and union leaders have pointed to PAR as a promising component of an effective human capital strategy, thus fueling interest and initiatives across the country.”
Dissimilar to relatively inexpensive teacher evaluation models that rely heavily on unstable Value-Added models, PAR is an expensive reform, costing $4,000 to $10,000 per teacher served. However, a Harvard study demonstrated that PAR affords the district “a range of financial savings and organizational benefits that offset program costs.” The benefits are accrued in two primary ways. First, the mentoring components of PAR help teachers succeed and avoid burnout, thus, increasing retention. Second, PAR, as a collaboration between the district and teachers, helps ineffective tenured teachers improve or they are subject to dismissal “without undue delay and cost because of the program’s clear assessment process and the labor-management collaboration that underpins it.”
The peer review and distributed leadership processes in PAR also empowers teachers to construct their own understanding and knowledge of the world, through experience and reflection upon those experiences. The expert consulting teachers (CTs) in PAR also creates specialized roles within the teaching profession where they are responsible for mentoring and evaluating peers which differentiates the work and future career opportunities of teachers.
In summary, when creating teacher evaluation systems, the primary data should be from veteran teachers that provide a variety of expert perspectives. PAR is not experimental and has been demonstrated successfully in districts across the United States (See the NEA Foundation’s Online PAR Training module #11 here). This constructivist approach identifies and fosters high quality educators by producing useful qualitative and quantitative data for efficacious teacher evaluation. A primary reason so many districts are flummoxed by issues relating to accountability and teacher evaluation systems is due to the fact that they are being forced to rely upon “junk science” statistical modeling that utilizes non-generalizable data from test scores. As an alternative, PAR is a locally-based approach in which veteran teachers create qualitative and quantitative data that is actually meaningful for sound teaching practice.
See also Politicians v. Experts: The Latest on “Value-added” Modeling
Please Facebook Like, Tweet, etc below and/or reblog to share this discussion with others.
Want to know about Cloaking Inequity’s freshly pressed conversations about educational policy? Click the “Follow blog by email” button in the upper left hand corner of this page.
Click here for Vitae.
Please blame the “student error term” for any typos.
Interested in joining us in the sunny capitol of California and obtaining your Doctorate in Educational Leadership from California State University Sacramento? Apply by March 1. Go here.
Pingback: New York Times Editorial Board on Annual Testing: “PREECCCIIOOOUUUUSSS!” | Daniel Katz, Ph.D.
Pingback: Community-Based Reform and Accountability Measures for States and Communities | Cloaking Inequity
Pingback: Will you stink up the privatization garden party with me? | Cloaking Inequity
Pingback: Axe is grinding: Is PAR teacher evaluation discriminatory? | Cloaking Inequity
Pingback: Podcast Series: Conversation with Loyola Students about Education Reform | Cloaking Inequity
This is a very timely and important conversation. I appreciate the thoughtful dialogue. The topic of teacher evaluations is often prone to very emotional and intense feelings and responses. In light of this, it’s important to keep in mind several key points.
1) Very few thoughtful educators are advocating for evaluating teachers on a single variable. Discussions around evaluations can be very misleading, Most practitioners, researchers, and politicians agree that evaluations should be comprehensive and include a variety of qualitative and quantitative data. The really intense discussion comes down to: how much weight should be given to each factor?
2) Assessment data (local, national, summative, standardized, formative, statewide, norm-referenced, etc.) is not inherently bad or good. What’s important is to use the data from each source strategically and appropriately. Teaching without assessing frequently is like wandering around in the dark hoping students will find the exit on the other side of the room.
3) Just as statistical models may be prone to instability, so too are perceptual data, self-report data, survey data, rubric scoring, and other qualitative data sources. Both quantitative and qualitative data have their place.
4) It is a mistake to assume that experienced more effective than teachers early in their career. In order to use teachers as mentors/evaluators it is absolutely necessary to distinguish between teachers that have a track record of exceptional achievement and those that merely have a lot of experience.
5) It is also a mistake to assume that we know who the good teachers are just by our perceptions and observations. It is often very difficult to tell the difference between teachers that are entertaining/engaging and teachers that consistently increase students proficiency in course content.
6) Teacher evaluations must be focused on what system and/or criteria of evaluation will lead to the most meaningful, effective learning experiences for students. All other priorities, including how palatable the process is for teachers themselves or how arduous it is for administrators comes a distant second.
In any responsible model, VAM is only a starting point (or at the very least one of many weighted components). Even the most hardened quantitative methodologists in educational research recognize that eventually someone has to evaluate and interpret the coefficients, p-values, confidence intervals, standard error, and effect sizes in the context of prior research to make an appropriate inference. The same will always be true of teacher evaluations. The point is to ensure a thoughtful, comprehensive evaluation system that provides the best possible education for students.
>I find your article over PAR vs VAM insightful, timely, and providential. My only question: How do we get everyone on board?!
>PAR has all the beneficial qualities of a progressive, small-is-beautiful, constructivistic approach. I believe it is the way the teaching profession evolved in this country– teachers as workers; principals as managers– that got us in trouble. Education needs to modern-up to join medicine and law where it is responsible to peer review and its own professional board. Not to the government or corporate oversight by a simplistic VAM quotient.
Other teachers always know who the good teachers are and it never has anything to do with test scores.