The Big Test was upon us, and during the next three weeks of testing, my special education students felt like complete failures—due to the fact that they were complete failures. They cried. They screamed and threw chairs. They stared at text that may as well have been written in Swedish and conjured up an answer. Many of them bubbled in the same letter each time, or simply rewrote the question in the "extended response" boxes, painstakingly recopying words that held no meaning for them.

The 18-plus-hour test was many grade levels beyond their abilities, yet the state of New Mexico forced my students to take it anyway so that number-crunchers could gauge whether or not I was an effective teacher working for an effective school. The state has little choice: If it wants to receive federal education dollars (roughly 18 percent of New Mexico's education budget), then it must administer the test to everybody.

Days after my students completed the test, Gov. Susana Martinez announced her new teacher accountability plan. The details haven't been hashed out yet, but the cornerstone of the plan is that, starting with the 2013-2014 school year, half of a teacher's evaluation will be based upon student academic achievement as measured by the very test through which my students had just suffered.

Currently, principals evaluate teachers annually based on their "professional development plans," a formality that few teachers take seriously, mostly because they're busy writing lesson plans and grading papers. It works like this: Teachers set a goal for the year, explain how they plan to achieve it, and write a reflection at the end. The principal observes the classroom a few times, then meets with the teachers individually to discuss the plan.

It's not hard to pass this evaluation, and the process weeds out only the worst teachers. Anyone able to earn a college degree can make sure that the basics are in place—lesson plans matched to state standards and posted on the wall, for example—and many principals notify teachers about upcoming observations to ensure their success. The system is easy to game and tends to reinforce the status quo.

Martinez claims that her plan focuses on rewarding good teachers while ensuring that "ineffective" teachers receive professional development tailored to their weaknesses. On paper, this makes sense, for a rigorous annual evaluation would prevent experienced teachers from resting on their laurels while motivating new ones to implement the skills they learned in their licensure programs. Incentives, financial or otherwise, can help push teachers to the next level of performance, and struggling teachers can certainly benefit from targeted support.

I am cautiously open-minded about the proposal, but I urge the planners to take four particular issues into consideration.

First, a successful classroom includes these elements: engaging lesson plans tied to specific learning goals, a positive behavioral support system designed to encourage on-task behavior and academic achievement, and a clear plan for dealing with misbehavior. Omit any of those pillars, and a substantial portion of students will fail to thrive. Thus, the first step in any evaluation should be ensuring that some variation of each of these elements is in place and being utilized on any given day. This would improve student performance more than adopting the latest overhyped reading program, and it would be cheap and easy to evaluate.

Second, not all school populations are created equal, and evaluators must take socioeconomic indicators into account. Out of my 22 students, only one lives with Mom and Dad. Seven have been adopted after being removed from abusive homes. Six more live with grandparents. The rest live with single mothers, all of whom struggle to put food on the table. Twelve of my students have been sexually molested at some point in their young lives—and those are just the incidents that have been officially documented.

These kids get very little support at home due to the fact that Dad is dead or in jail, Mom is in rehab for heroin, and Grandma is simply too tired to help out with a spelling lesson. I'm not overdramatizing the situation or trying to make excuses for myself as a teacher—I'm simply sharing what I see every day. Any evaluator who fails to take the socioeconomic context into account will end up punishing teachers who have chosen to serve the state's neediest children.

Third, if you want high-quality teachers, pay them accordingly. After paying taxes and covering my portion of my health care costs, my family and I are left with $400 per week. That's $1,600 a month to cover everything: rent, utilities, groceries, gasoline, insurance deductibles, student loans, clothing and the ever-mounting credit card payments. Many of the best and brightest college students take one look at our teaching salaries and choose another career—particularly if they'll be subjected to a stringent teacher evaluation process.

Finally, there's the matter of the Big Test. I think the test is a fair assessment of what a student should learn by the end of a given year, but only if a student entered the grade ready for the new knowledge. Without exception, my students began this school year with academic skills far below their peers. Many have been passed along simply because retaining (aka "flunking") them for a lack of reading skills would mean that, by the time they read at a third-grade level, they'd be 14 years old. If the evaluators expect these students to suddenly become proficient readers, they'll be disappointed—and I'll be labeled as ineffective.

Since special education students all have Individualized Education Plans that lay out specific learning goals, special educators' evaluations should examine students' progress toward meeting these goals rather than scores on tests they'll never pass. Academically, these goals are modest compared to the standards of the official test—a fourth-grader might be focusing on adding two-digit numbers. Such goals might seem silly, but meeting them requires all the willpower my students can muster, and any evaluation of my teaching skills must take such situations into account.

New Mexico teachers are in for some big changes. My sincere hope is that the new evaluation process will help build up schools and teachers rather than punish them for failing to meet unrealistic expectations on a single test.