Teaching to the Test: Should Test Scores Rate Teachers?
September 29, 2014
Since the launch of Race to the Top funding in 2009, the Obama administration’s Department of Education has leveraged these grants and stimulus dollars to incentivize states to enact teacher evaluation systems that gauge teachers on student test score growth in addition to administrator observation.
In his latest research, the College’s Ben Superfine openly questions whether this push for test score-based evaluations is truly measuring the metrics that matter most in the classroom. In his article "The Promise and Pitfalls of Teacher Evaluation and Accountability Reform," published in the 2014 edition of the Richmond Journal of Law and the Public Interest, Superfine argues test scores are not only a poor measure of an individual teacher’s contributions to testing gains or losses but also fail to measure the more intrinsic aspects of teaching that are equally important to the profession.
“Current teacher evaluations are riddled with several sorts of interlocking problems that hamper effectiveness,” said Superfine, PhD, associate professor of educational policy studies and director of the Research on Urban Education Policy Initiative. “We may think of evaluations as tools for giving merit pay at the high end or dismissing teachers at the low end, but teacher evaluation systems actually do not effectively measure these ends that we care about the most.”
For example, Superfine says a math teacher may be rated favorably when his or her students post gains on standardized tests, but such a black-and-white approach fails to incorporate math learning that may take place in a separate science classroom. Proponents of the use of testing as an evaluation tool claim these systems improve teacher motivation, but Superfine says his research paradoxically indicates the opposite: teachers value autonomy and teach for reasons other than simple incentives. “Teach to the test” is a bit cliché, but Superfine says test-based evaluations push teachers to focus on more narrow skills instead of conceptual understanding and inquiry.
The other main area of struggles Superfine focuses on is the evaluation of teachers with heavy student populations of English language learners, students with special education needs and students from high poverty areas. He says test scores do not contextualize external factors that impact student performance and says research shows teachers who post raised test scores with general student populations may not have the ability to raise the performance of students with higher levels of need, suggesting a problem with the testing system itself.
“In a utopian world, we hope all policies are aligned and move towards similar goals,” Superfine said. “The problem with evaluation systems is they are mostly focused on accountability for teachers and largely silent on the other functions of the teacher workforce—professional development, hiring, promotion and child development.”
What’s the big fix? Superfine says the evaluations need to be re-evaluated. Policy leaders need to detail a robust conception of what an effective teacher actually is before evaluation systems are designed, in particular studying what a good teacher looks like in a difficult classroom setting.