>The reason testing is good isn't because it is somehow super accurate. It is good because it keeps teachers honest. Without testing, how do you measure teacher performance at all? How can you tell if someone who is capable of teaching well isn't just being lazy or getting distracted?
The other issue is that value-added testing should be used in evaluating teachers, but only a single component of a larger evaluation process. On its own, it's almost as pernicious as the system we have now.
>It's easy to point out problems. It's useful and important that people find flaws and fault.
The largest issue isn't identifying problems—it's that the people who benefit most from the status quo (teachers, unions) derive very concentrated benefits, while the solutions to the status quo are diffuse (students, parents, society). One sees this pattern in many fields (farm subsidies are the canonical example), and there's an extensive economics literature on it.
The other issue is that value-added testing should be used in evaluating teachers, but only a single component of a larger evaluation process. On its own, it's almost as pernicious as the system we have now.
For more on these subjects, see
* "What Makes a Great Teacher?": http://www.theatlantic.com/magazine/archive/2010/01/what-mak...
* "Why Kids Should Grade Teachers": http://www.theatlantic.com/magazine/archive/2012/10/why-kids...
* The collection of articles I listed here: http://jseliger.wordpress.com/2009/11/12/susan-engel-doesnt-... .
>It's easy to point out problems. It's useful and important that people find flaws and fault.
The largest issue isn't identifying problems—it's that the people who benefit most from the status quo (teachers, unions) derive very concentrated benefits, while the solutions to the status quo are diffuse (students, parents, society). One sees this pattern in many fields (farm subsidies are the canonical example), and there's an extensive economics literature on it.