The Council for Exceptional Children has issued a new position paper that rejects the use of “value-added” systems that link evaluations of teachers — especially those in the field of special education — to student standardized test scores and says they are cause “for concern.”
Here’s part of the position paper:
The most commonly discussed method used to incorporate student growth into teacher evaluations is a statistical model known as “value added.” In its simplest form, a value-added model aggregates student test scores, generally from state-mandated assessments, with a few other factors such as school and student demographics and produces a score for the teacher which purports to describe the teacher’s impact on student growth (Lipscomb, Teh, Gill, Chiang, & Owens, 2010). The research that exists about these models indicates they are only reliable over time (i.e., based on several years of data) with larger student populations and when the underlying assessment instruments are fair, accurate, and reliable.
Moreover, research has determined that value-added calculations are invalid for two teachers in a co-teaching environment, as the statistical model cannot determine
which or by how much each teacher impacts student learning (Steele et. al., 2010). Additionally, most state data systems are not sophisticated enough to account for innovative models of instructional organization (Watson, 2012). While value-added models cause some concern for all teachers, they raise specific concerns for any educator who works with students with disabilities.
Because of these concerns and the lack of proven methods, there is increasing recognition that evaluating the effectiveness of teachers of students with disabilities
needs singular attention. Currently, there is no consensus and almost no research about how these teachers might be evaluated (Holdheide et. al., 2010). Indeed, very few states and districts are addressing the unique challenges associated with evaluating special education teachers, and this is an area where much work remains.
The position paper says that teacher evaluation systems should include multiple measures and “must involve teacher input and expertise and help them develop throughout their career.”
It includes examples of some teacher evaluation systems that use multiple measures, including, ironically D.C. public schools, whose IMPACT teacher evaluation system has come under criticism for being so heavily weighted on standardized tests through value-added formulas. These formulas purport to be able to figure out how much “value” a teacher has added to a student’s learning based on the students’ test scores.
You can read the whole position paper here.