|
Abstract:
|
With the transition from No Child Left Behind (NCLB) to the Every Student Succeeds Act (ESSA), the landscape of teacher and school evaluation and accountability is changing. There is an increased movement toward minimizing or eliminating the impact of standardized test scores on teacher and/or school evaluation. The National Science Foundation funded RealVAMS project (DRL-1336027 and DRL- 1336265) developed a model and R package that utilizes "real-world" outcomes, such as graduation rates and college entry in addition to typical standardized tests to model teacher/school effectiveness. This presentation will compare estimates from current value-added models (VAMs) to those generated by the RealVAMS model where effectiveness is measured simultaneously with alternative "real-world" outcomes. The presentation will also include a brief discussion of the R package, RealVAMS, to generate the estimates of teacher and school effectiveness. This application of a generalized linear mixed model is currently used as a VAM in an educational setting, but can be applied to other fields to evaluate latent traits, such as doctor effectiveness in medicine and player/team effectiveness in sports.
|
