September 2001
How to Test a Teacher: A Lesson from Tennessee for Ontario
by Claudia R. Hepburn
As students return to school this month, the Ontario government has a mission to get teachers back to the books as well. The Harris government will attempt to improve teacher quality—unquestionably a worthy goal—by testing teachers. Starting this fall, Ontario plans to test teacher competency by administering written exams to each teacher every five years. Forty thousand teachers will be randomly selected for the first phase of the re-certification program, and those who fail could lose their teaching certificates (Ross, 2001).
The testing program is a brave move on the government's part. More precisely, it is brave and foolhardy. It is brave to attempt, in the face of vitriolic opposition from the teachers' unions, to ascertain which teachers are competent, and to de-certify those who are not. It is, however, foolhardy to go head-to-head with the teachers' unions over a policy that is unlikely to succeed in its mission, worthy though that mission may be. At the end of the day, the exercise is likely to be costly to the government's credibility, to taxpayers footing the bill, and to the teachers in this province who already suffer from low morale.
Ontario's teacher testing program is likely to fail because the Harris Conservatives have fallen into the same trap that governments have fallen into for generations: the temptation to measure inputs rather than outputs. Like the requirements for education degrees and courses in pedagogy which are already in place to keep incompetent adults out of classrooms, the competency test will decrease the pool of potential teachers permitted to teach, but will not guarantee the quality of those who do. The written test cannot guarantee quality because it aims to measure teaching ability through something other than teaching.
A better approach would be to test not what teachers know, but how much their students learn. This approach has been applied successfully in Tennessee for nearly a decade and is gaining recognition elsewhere. Called the Tennessee Value-Added Assessment System (TVAAS) this teacher-testing model calculates the "value" added by a teacher over the course of a school year to a student's work. Students are tested at the beginning of the year and again at the end, and teachers are graded based on the progress their students make over the course of the year.
TVAAS was developed by Prof. William Sanders, the director of the University of Tennessee Value-Added Research and Assessment Center. Using TVAAS data, collected annually by the state from every elementary classroom in Tennessee since 1992, Prof. Sanders has calculated the impact that effective and ineffective teachers have on their students. He has found that teacher effectiveness is "10 to 20 times as significant as other things" (Sanders, quoted in Marks, 2000).
Contrary to popular belief, family background is nowhere near as significant to a child's academic success as a string of good or bad teachers. Prof. Sanders found startling differences when he followed the academic progress of children who had the same level of academic achievement at the end of grade two. One child who had a succession of good teachers—those identified as being in the top 20th percentile of TVAAS effectiveness—would score, on average, at the 96th percentile on the fifth grade math test. Another child with the same second grade achievement, taught for three years by a teacher at the bottom of the TVAAS scale, would score in the 44th percentile" (Sanders, quoted in Marks, 2000).
TVAAS is fair to teachers because it excludes the influence of all pre-existing differences among students, including race, socio-economic background, intelligence, and previous learning. It therefore levels the playing field for teachers, and in doing so removes the objection put forward so often against standardized tests: that they do not account for the student's background (Stone, 1999). Teachers get no credit for having a classroom full of well-prepared students, and they are not penalized for a classroom of children testing below average. What counts is how much they improve their students' skills in the course of the school year.
Tennessee tests all elementary students and reports each teacher's performance to the district, the school, and the teacher. School districts can use the information as they see fit. Those that have chosen to ignore the information have stagnant or declining results while those that use it to work with teachers and schools are showing measurable progress (Marks, 2000).
None of us likes to be tested, but tests are worthwhile if they measure skills accurately. The flaw in Ontario's proposed test is that it cannot guarantee that good teachers will pass and bad teachers will fail. All it aspires to do is to measure which teachers know the material being tested and which do not. Certainly all teachers should have a firm command of the material they teach, but they should be tested on this before they are ever allowed at the front of a classroom, not repeatedly through their careers. Subject tests have been absent for too long from our faculties of education and that is the place to reinstate them. Once certified, teachers should be tested not on what they know, but on how well they convey their knowledge to students. A value-added assessment such as TVAAS would do just that.
Though the teachers' unions would certainly fight value-added assessment as vehemently as they are fighting the government's current proposal, I suspect that many good teachers might privately see the value of it. Good teachers could receive a deserved pat on the back and weaker teachers could be helped to improve. As a result, morale would improve for the whole teaching profession, something that has not happened for a long time in Ontario.
As students go back to their books this month, so should government policy makers. They should study up on policies such as the Tennessee Value Added Assessment system that have proven they can assist schools in identifying and improving the quality of their teachers. Only then will teachers have good reason to work for their "A" rather than being tempted to cut class altogether.
References
Dave Ross, spokesman for Janet Ecker, Minister of Education (2001). Quoted by Mohammed Adam in "Teachers set to defy re-testing order." National Post, August 9, p. A4.
Marks, Marilyn (2000). "The Teacher Factor," New York Times, Education Life Supplement, January 9.
Stone, J.E. (1999). "What Is Value-Added Assessment and Why Do We Need It?" Education Consumers Clearinghouse. Digital Document: http://education-consumers.com/TVAAS-web.htm.
Claudia R. Hepburn (claudiah@fraserinstitute.ca) is Director of Education Policy at The Fraser Institute, and a former teacher. She has a B.Ed. and an M.A. from the University of Toronto. She works in Toronto.
info@fraserinstitute.ca
4th Floor, 1770 Burrard Street, Vancouver, BC, Canada, V6J 3G7
Tel: (604) 688-0221 Fax: (604) 688-8539 Book Orders: 1-800-665-3558 ext. 580
You can contact us at the above email address for any comments or information requests. Please report any dead links or technical problems.