Researcher Thomas Kane explains the link between good teachers and good test scores

December 17, 2010 posted by Thomas Kane

StudentsFirst scored its first Newsmaker interview to help explain the new study that finds a link between good teachers and good student test scores. Why should you care? In cities and states around the country, schools are having the debate about how to evaluate teachers. StudentsFirst and other reform groups believe a significant part of a teacher’s evaluation should be based on how his or her students demonstrate progress on tests and this study seems to support that practice.

The study’s lead author is Thomas Kane, Deputy Director of Education at the Bill & Melinda Gates Foundation and a professor of education and economics at the Harvard Graduate School of Education.

StudentsFirst: This is the first time a major study has looked at the how student test scores can predict the quality of their teachers. Why did you do this study?

Thomas Kane: We were not the first to use student achievement data to identify effective teaching. However, this is the first study to try to assemble a package of other measures — including student feedback, classroom observations, a new test of teachers’ content knowledge — to provide a more comprehensive look at a teacher’s practice. We want to make sure the other measures — such as student feedback and classroom observations — are related to student achievement gains, that they are identifying the right practices and behaviors and not pointing teachers and principals in the wrong direction.

SF: Can you simply explain “value-added” analysis of students test scores and how it relates to teachers?

TK: A teacher is said to have positive “value-added” when the students in his or her class outperform other students who had similar starting points — similar prior achievement, program participation and demographics — and had similar peers.

SF: So, why should parents care about your preliminary results?

TK: Parents know from first-hand experience that children’s learning depends heavily on the teacher leading the classroom. Research confirms that. There are often dramatic differences in student achievement gains in different classrooms in the same schools. But what many parents may not know is that teachers receive essentially no feedback for the work they do. Imagine what it would be like to come to work every day without any feedback on one’s performance. Now, imagine that you had one of the most important jobs in the world. Even President Obama got a chance to learn from the mid-term elections. Teachers don’t even get that.

The costs of that neglect are enormous, for teachers as well as students. Young teachers plateau far too early in their careers without the feedback they need to continue to learn on the job. Principals make the most important decision we ask them to make — granting tenure to a young teacher after a few years on the job — without any objective information. (Perhaps not surprisingly, they often abdicate that responsibility.) Experienced teachers have very few opportunities to get constructive feedback from peers or their principals. We need to change that.

If the evidence suggested that teachers were mostly the same, and we did not see such huge differences in student achievement gains between classrooms, we would not be doing this project. It is precisely because the evidence suggests that the quality of instruction varies so much that we need to learn more about what the effective teachers are doing.

SF: Should parents be in favor of having their students’ teachers evaluated based in large part on student test scores?

TK: Parents should request the same two principles we’re using to guide this study: first, in the grades and subjects where testing data are available, student achievement gains should represent a substantial part of annual teacher feedback; second, any other component of the evaluation—whether it be based on a classroom observation or student feedback—should be demonstrably related to student achievement gains.

SF: Some detractors have argued that if teachers are graded on how well students perform on tests they will “teach to the test” and kids will not learn critical thinking skills. Does your study shed any light on this concern?

TK: There are two results in the study which shed light on this question: First, the teachers who were successful in promoting gains on the state test also tended to be successful in promoting gains on the supplemental assessments. Second, the classrooms where students reported that “We spend a lot of time in this class preparing for the state test” were not as successful as the classrooms where students reported “We use time well in this class and we don’t waste time” or “We learn a lot in this class almost every day.”

SF: The study also looks at student evaluations and suggests that they are accurate predictors of how well a teacher is doing? Wow, so our kids have the right instincts?

TK: Like value-added data, student perceptions are not perfect. However, the results suggest that when students are asked detailed questions about their experiences in a classroom (that is, it’s not simply a popularity contest) they can provide information that is related to student achievement gains. In higher education, this is the primary way that deans and department chairs are able to assess the quality of instruction being delivered. We wanted to test whether students could help identify the conditions under which they learn more. It seems they can.

One other advantage of the student perception information is that it could help point a teacher and his or her supervisor to specific strengths and weaknesses. If a teacher is having a problem with time management or providing timely feedback on student homework, the student feedback could flag those issues for further development.