How stable are value-added measures of teaching effectiveness?
Value-added scores for a teacher can vary relatively widely from one year to the next. This is due in large part to the fact that class sizes are relatively small—for example, an elementary school teacher typically has between 15 and 25 students in a class, and with a sample size that small, a few unusually rowdy or studious students can make a relatively big difference in value-added scores from year to year. However, variations in value-added scores do not make them unreliable predictors of future teaching effectiveness. This is because beliefs about a teacher’s effectiveness are cumulative; they are not based on only one year of data. Averaging value-added scores from two or more years creates a cumulative estimate of teacher value that is a very reliable predictor of future effectiveness. Rankings based on these cumulative estimates vary little from year to year, change very little over time, and predict large future performance differences among teachers.
Related Questions
- The teachers put so much hard work into their classrooms. What measures will be taken to stop misuse of teaching space?
- What are the challenges of collecting VAM (Value-added Measures) given the high number of transient students?
- How do value-added measures differ from adequate yearly progress (AYP) measures?
