Yesterday I wrote that comment on an opinion column in Ed Week in which the author Justin Snider opposed merit pay based on student test scores. He said you could not attribute a student's score solely to the that year's teacher of that subject.
He dismissed the "value added" approach and described it:
"When it comes to school accountability, today’s favorite catchphrase is “value added” assessment. The idea is that by measuring what students know at both the beginning and the end of the school year, and by simply subtracting the former from the latter, we’re able to determine precisely how much “value” a given teacher has “added” to his or her students’ education."

Now Justin Snider is a professor at Columbia and he either does not understand "value added" or he's disingenuous.

The Houston Schools are using the value added approach to teacher evaluations but it's quite different from Snider's description.
"Put simply, the value-added analysis by North Carolina statistician Bill Sanders looks at a student's test score history to project his or her scores the next year. Teachers are rated on whether their students met, exceeded or fell short of expectations."

Sanders has repeatedly defended his method, which he pioneered in Tennessee in the 1990s."

If Professor Snider wants to question the validity of this "value added" approach, I'll listen, but to mischaracterize it and then dismiss it, diminishes the rest of his arguments.

Here's another professor, Daniel Willingham, who mischaracterizes value added. He produced a cute YouTube video to make his point.
Video: Six Reasons Value-Added "Growth Model" Teacher Evaluations are Unfair.
FYI-Willingham is all over ed journals this week.