Blog

I recently came across an opinion piece in the Washington Post in which the author wrote about a prominent school board member in Florida who took the 10th grade portion of the FCAT –Florida’s Comprehensive Assessment Test—and, perhaps surprisingly, failed miserably (and openly admitted doing so). The author used this story to highlight the shift in education to use the results of assessments such as this (which an educated adult struggled with) in more and more high-stakes decisions about students and teachers, and brought up the current shift in the evaluation of teachers and principals in the state of New York as an example.

Teachers and principals in New York are now being evaluated on a 100 point scale, which will group personnel into one of four performance categories: Ineffective, Developing, Effective, and Highly Effective. The ratings will be based on value-added measures of student learning (40%) and a principal/supervisor evaluation (60%), and will be used, in part, for decisions about hiring, firing, and compensation.  And, you probably guessed this….teachers and principals in NY aren’t happy.

Do they have reason to be upset? That question is perhaps one of the most polarizing in education today. Teachers are there to teach, so why shouldn’t we hold them accountable for the impact they have on their students, most notably on how well their students perform on standardized tests?

While evaluation of student performance and evaluation of teacher performance seem to go hand in hand, there are a number of reasons policymakers might ease up on the evaluation gas pedal. A number of leading researchers on value-added measures noted the following concerns, among others, in response to the implementation of the New York evaluation model:

• There could be an overemphasis on “teaching to the test”, or only teaching those things that might be covered on a standardized test;
• Teachers could be discouraged from working with those students who are less likely to perform well on these tests (i.e. the students most in need of additional academic support) as they might negatively impact a teacher’s evaluation;
• Different value-added models produce different results….if that is the case, which value-added model is the “right” model?;
• The tests used for a large portion of this evaluation were not designed to measure how much a student has learned over a school year, but instead, were designed to measure if students have met the state’s proficiency standards or not.

Wherever you stand on teacher accountability, whether you think all decisions about teachers should be based on measures of performance, or whether you believe that standardized tests are the bane of education today (though if you are reading this blog, chances are you likely don’t fall into this camp), there are real stakes being attached to the results of these assessments. And in more and more states, the term “high-stakes” is becoming less about the performance of students (as the term was intended), and more about the performance of teachers.

So what is my take on evaluation, testing, and accountability? First, I’m all for holding teachers accountable, especially for the impact they have on student learning. However, and that is a big however, if we are going to hold teachers accountable, I think we have to also consider the following:

• Give teachers better information about their student’s performance during the school year so teachers can make mid-year adjustments to better meet the academic needs of their students;
• Use assessments that accurately measure student growth; in this way, we can capture how much students have learned during a year (and not discourage teachers from working with the lowest performing students who simply won’t be able to meet proficiency no matter how much growth they show);
•  If you are going to evaluate teachers using test scores, make the evaluation process as transparent and understandable as possible; telling a teacher that she needs to raise the performance of her classroom by 2 standard errors is likely less helpful than setting reasonable benchmarks of growth for each student at the start of the year;
• Finally, test scores should be a part of the evaluation, not the evaluation; from year to year, test scores might vary for a number of reasons, so making a decision about a teacher’s employment based on one year of testing data may result in the wrong teacher being let go.

Fortunately or unfortunately, depending on your stance, these types of teacher evaluations aren’t going away; in fact, they are likely going to soon become the rule in education, rather than the exception. Thus, perhaps my take away point from all of this is if teachers are going to be evaluated in this way, then policymakers and education leaders need to approach this process in a thoughtful, deliberate manner to ensure a fair assessment of teacher performance, year in and year out.

Of course, these are just my thoughts on the role of testing in the evaluation of teachers…what are yours?