The Christie administration has called for the use of student standardized test scores to evaluate teachers. But the research shows that these value-added models (VAMs) of evaluation are fraught with problems and will discourage collaboration among teachers and narrow the curriculum.
“Solving VAM's problems range from difficult to perhaps impossible,” said noted statistician Howard Wainer. “And until they are resolved the uses for VAM must be sharply limited.”
Wainer, a distinguished research scientist for the National Board of Medical Examiners, served as the moderator of a panel discussion on the use of standardized tests in teacher accountability systems hosted by the Educational Testing Service last month.
“The research vs. the rhetoric: why experts urge caution in the use of student test scores to evaluate teachers,” an article from the March NJEA Review outlines what researchers said at the symposium.
Henry Braun, Boisi professor of education and public policy at Boston College, Sean P. Corcoran, assistant professor at New York University’s Steinhardt School of Education, and Arthur E. Wise, president emeritus of the National Council for Accreditation of Teacher Education, also spoke. Each panelist identified several reasons why the sole use of student test scores to evaluate and pay teachers was not a wise course of action.
Video highlights of the symposium can be found on njspotlight.com. Links to the major works of all presenters on the topic of value-added models, teacher evaluation, and education reform are also posted. Ninth-grade English teacher Kevin Parker attended the event and blogged about it on njea.org’s Member Connection. Parker works at Washington Township High School, Gloucester County.
Richard Rothstein, a research associate of the Economic Policy Institute and former national education columnist of The New York Times, presented the opening keynote at the symposium. He cautioned the audience of 150 education stakeholders and policy makers, against focusing only on teachers in the pursuit of education reform.
“Teachers are important,” Rothstein noted, “but research shows that only 1/3 of variation in the achievement of children is attributable to in-school factors, including teachers, while 2/3 is attributable to out-of-school factors.”
Laura Goe, a research scientist in the Performance Research Group at ETS, was the symposium’s final presenter. Goe, who also serves as the principal investigator for Research and Dissemination for The National Comprehensive Center for Teacher Quality, stated that multiple measures of evaluation must be employed when determining teacher effectiveness.
“A student test score is just a snapshot,” Goe noted. “Plus, most teachers do not work in subject areas that are tested.” She went on to describe those measures that help teachers grow such as those that are explicitly aligned with teaching standards and professional development offerings.
NJEA was one of 13 organizations to sponsor the event, titled “Standardized Tests and Teacher Accountability: The Research.” Other sponsors included the N.J. Principals and Supervisors Association, the N.J. Association of School Administrators, the N.J. School Boards Association, the Garden State Coalition of Schools, the N.J. Center for Teaching & Learning, the N.J. Association of Supervision and Curriculum Development, and the Rutgers Graduate School of Education.
See your March NJEA Review for an in-depth look at the researchers’ findings.