Peter Gwynn, a D.C. public school teacher who was one of 206 teachers fired this month under the IMPACT evaluation system, writes here about why he believes he was really let go. The fired teachers, my colleague Bill Turque reported, amount to 5 percent of the 4,100 teachers in the system. IMPACT was first implemented under the chancellorship of Michelle Rhee, who quit last October and was succeeded by her deputy, Kaya Henderson. IMPACT has been criticized on a range of issues, with critics calling it arbitrary and punitive and primarily a way to fire veteran teachers. D.C. schools officials reject the criticism.
By Peter Gwynn
Ambiguity can be a friend or an enemy. If you are a D.C. Public Schools teacher, it might just depend on who gets to measure the difference between words like sometimes and frequently.
I was one of the 200-plus DCPS teachers fired last week due to poor evaluations. I believe, though I cannot prove, that I was fired from DCPS, the Columbia Heights Education Campus (CHEC) in particular, because of the opinions and ideas expressed in my blog. Given the facts, this is simply what makes the most sense to me. But the more important point is that the system’s teacher evaluation system, IMPACT, is riddled with ambiguity and imprecise language such that administrators could easily manipulate teacher scores to punish or reward as suits their ends.
For most teachers, IMPACT is composed primarily of five, half-hour observations throughout the year. Three observations come from school-based administrators; two from a District master educator.
In each observation the evaluator uses a rubric to judge the teacher in nine separate performance categories. In each category, the teacher is scored a 1, 2, 3 or 4. These are averaged to produce an observation score of 1-4. The five separate observations are then averaged to produce an overall yearly score of 1-4.
The magic number is 2.5. If you score at or above this mark, you are effective or highly effective and your job is safe. Scoring below 2.5 defines you as ineffective or minimally effective and your job is lost or in jeopardy. I scored below 2.5 for two consecutive years and was fired.
I believe that my administrators wanted to get rid of me and that I was punished for my writing. In the blog I never reference people or places by name — and my name is not on it — though I do recount events from my school truthfully and in detail. It is satirical, vulgar, bombastic, and critical. If I worked for myself, I would want to fire me. Many people have expressed disapproval of the blog. That is their right, as it is my right to write it.
I believe I was targeted by the administration because that is where the evidence points.
Over two years of IMPACT, my master educator scores averaged 2.75, comfortably effective (your job is job safe). Over the same two years, my in-house evaluations averaged 2.06, comfortably minimally effective (your job is in danger). The difference is 0.69; fairly large on a scale that runs only from 1 to 4.
But the timeline is what draws my attention most. My first school-based IMPACT observation was in November 2009. I was scored effective. I started the blog one month later in December. By the end of January 2010, I had been informed that the administration was aware of my blog and reading it. This was confirmed by another source shortly after. Strategically, I suspect, the administration has never asked or confronted me about the blog.
Henceforth, my school-based evaluators scored me, without exception, minimally effective or ineffective. Meanwhile, the master educators continued to score me effective. One master educator this year noted how much I had improved since last year.
So, how could this happen?
Ambiguity in the language of IMPACT invites the capricious, perhaps subconscious, punishment or reward of teachers by administrators. Though it is cloaked in the false precision of a rubric, it is infinitely subtle and subject to manipulation. For example, in a single half-hour observation, while simultaneously monitoring and scoring eight other performance measures, an evaluator is expected to be sure if a lesson is:
Accessible and challenging to all students (score 4)
Accessible and challenging to almost all students (score 3)
Accessible and challenging to most students (score 2); or
Not accessible and not challenging to most students (score 1).
Most teachers live in the 2-3 range. If I was an administrator and I wanted to get rid of somebody, I would shade to the 2. Nobody could stop me. Nobody else was there to witness what happened. Nobody is able to check if I am consistent between teachers. It is up to me to decide what almost all means and how I will measure it in that half hour.
IMPACT is littered with language like this; subjective and inviting manipulation.
I don’t know for certain that I was targeted and that the ambiguity of IMPACT was used to illegally punish my speech. But the language of IMPACT, my IMPACT data, and the CHEC administration’s well-earned reputation for tolerating no dissent give me specific cause to suspect it.
If IMPACT is to be taken seriously in the future, this should be fixed.
Follow The Answer Sheet every day by bookmarking http://www.washingtonpost.com/blogs/answer-sheet. And for admissions advice, college news and links to campus papers, please check out our Higher Education page. Bookmark it!