With findings reminiscent of those from the Gates Foundation’s recent MET study or Chetty’s teacher-effectiveness research, this CALDER paper widens an already well-worn trail. Using a comprehensive, five-year dataset of student-test scores for beginning teachers in New York City, the authors find that early value-added results (though imperfect) are strong predictors of educators’ long-term effectiveness and that relative teacher performance (based on student test scores) remains fairly constant. Among math teachers whose performance was in the lowest quintile after their first two years on the job, 62 percent still performed in the bottom two quintiles in their third through fifth year and only 19 percent ended up in the top two quintiles. Similarly, if a school adopted a policy of firing the bottom 10 percent of new teachers (averaged over years one and two), it would rid itself of almost one third of the future lowest-performing teachers and absolutely none of the future top performers (according to years three, four, and five averages). They also find that value-added in years one and two explained 27.8 percent of the variance in average future performance (compared with only 2.8 percent explained by a number of combined “input” metrics including teacher demographics, credentialing scores, and competitiveness of undergraduate institution). The implications are clear: Cage-busting leaders should simply not keep the low performers around long enough to let them gain tenure.
SOURCE: Allison Atteberry, Susanna Loeb, and James Wyckoff, “Do First Impressions Matter? Improvement in Early Career Teacher Effectiveness” (Washington, D.C.: National Center for Analysis of Longitudinal Data in Education Research, February 2013).
