I don't have high hopes for this system working out. The ATR situation is a mess. We have a different ATR in my school every week. I really liked last week's ATR; I even saw him teach part of a class, and thought he was pretty good. He certainly seemed competent to me. But this seemingly solid, knowledgable, hard-working man has been replaced by a different one this week.
Here's a thought: As the ATRs rotate through the system, why not let them stay in one place if a principal likes them? The principal could make a phone call or send an e-mail letting someone know that the ATR teacher is working out well and that s/he would like to retain the ATR teacher as a long-term substitute. That ATR doesn't have to wonder where s/he is headed each week; the school gets a reliable long-term substitute that works well in that particular school; the DOE has one less teacher to find a random place for every week. Problem solved. I'm off to go work for Tweed now and enjoy my BlackBerry/latte/town car budget.
But back to the issue of evaluation. If, as the city claims, one or two formal observations by a principal is insufficient to rate a regularly appointed tenured teacher, then how on Earth do they propose to complete 4, or 6, or whatever the magic number is, observations for a teacher who isn't in the same place, the same subject area, or in front of the same kids for longer than 5 days at a clip? I've never been observed, even informally, while doing a class coverage, and I'd never in a million years want to be. "Survival" is a fine goal for most coverage lessons if you don't know the kids. I can't imagine my job security turning on how well I taught a lesson for which I knew none of the kids, didn't know the subject area, and had been in the school less than a week. (Although if a teacher could actually pull that off, they should be hired, for any vacancy, from wood shop to AP physics to French, immediately, because clearly s/he can do anything.)
Well, good luck, ATRs. It looks like you're going to need it.