Over the past two decades, in their search for a way to measure educational effectiveness, reformers landed on Value Added measures (VAM, or sometimes VAAS). Value added is a useful concept in manufacturing; if I take a sheet of metal and turn it into a toaster, I have made that sheet of metal more valuable.

Value added is harder to compute with human children. If the sheet of metal just lies there doing nothing, no value is added. But human children will grow and “add value” on their own. So the problem, if we’re going to measure how good teachers are, is to figure out which part of the student’s growth can be attributed to the teacher.

The basic idea is to work up a prediction of how Chris would have scored in some alternate universe where all school influences are neutral. If Alternate Universe Chris would have scored a seventy-five, and real world Chris scores an eighty-five or a sixty-five, then the ten points difference must be because of Chris’s teacher.

How can we possibly know how AU Chris would have done?

Source: Over A Year Ago, A Federal Court Struck Down VAM: Why Are We Still Using It To Evaluate Teachers?