Rethinking Value-Added Models in Education: Critical Perspective on Tests and Assessment-Based Accountability by Audrey Amrein-Beardsley

In order to keep pushing VAMs and VAM products, producers, politicians, and lobbyists, use some key assumptions. It is assumed that VAMs will yield more accurate statements about the causal effects of educational inputs like teachers, administrators, schools, and districts. A large majority of our population readily buys the assumption that the best way to measure school performance is to measure the gains posted by students over time. Many also believe that stronger accountability can eliminate the root causes of poor student performance. The big assumption is that teachers are largely responsible for student performance, but only 10%-20% of the variance in test scores can be attributed to teachers. The rest is due to other school factors and out-of-school factors. Another assumption is that schools have too many bad teachers. The assumption states that if you get rid of the bad teachers, there are lot of good teachers waiting to take their place. There is no reason to believe that this is the case as most people looking for teaching jobs are relatively inexperienced.
The norm-based nature of VAMs assures us that half of the teachers will be on the left side of the bell curve. Many ineffective teachers also voluntarily purge themselves from the profession. Most people know when they aren’t doing a good job or enjoying their work. Since teaching is very complex, so is determining teacher effectiveness. Thus, we have no idea of how many ineffective teachers there are. The assumption that VAMs will improve what we have is not supported by the evidence. Even renown businessman Warren Buffett warned about the misuse of data to evaluate people and other things. The final assumption in the chapter is the when the VAM data is feed back to teacher it will somehow add value and allow teacher to teach better. The problem is that VAM data does not have nearly enough depth to be diagnostically or instructionally useful. Not only is it difficult for teachers to understand, it often comes after students have moved on to the next grade.
VAM supporters believe that the rewards and penalties they bring will result in teachers teaching better. This also assumes that many teachers are teaching in an intentional, sub optimal manner. Assumptions, by their nature, appeal to people’s emotion. While it is sometimes necessary to make assumptions, it is important to realize when one is dealing with assumptions as opposed to established facts. In the case of the VAM assumptions, they are certainly doubtful at best.