On April 8, 2014 the American Statistical Association issued a statement on the Use of Value-Added Models (VAMs) for Educational Assessment. (http://www.amstat.org/policy/pdfs/ASA_VAM_Statement.pdf.) The purpose of VAMs is to “estimate effects of individual teachers or schools on student achievement while accounting for differences in student background.” In other words, VAMs attempt to determine the extent to which a teacher “adds value” to a student’s learning independent of external factors such as home life, family income, prior knowledge, exposure to information, parental involvement and support, and basic intelligence. Using standardized tests, VAMs attempt to rule out all influences on student learning except instruction and then quantify the extent to which instruction alone results in academic growth. North Carolina has adopted VAMs as part of the teacher evaluation instrument. VAMs also yield data that is factored into school and district rankings.
Not surprisingly, many educators are skeptical of the reliability of VAMs. Apparently, that skepticism is shared by the America Statistical Association. Their report says, “VAMs typically measure correlation, not causation: Effects – positive or negative – attributed to a teacher may actually be caused by other factors that are not captured in the model…Most VAM studies find that teachers account for about 1% to 14% of the variability in test scores…” yet some school systems are using VAMs for high stakes purposes, including “compensation, evaluating and ranking teachers, hiring or dismissing teachers, awarding tenure, and closing schools.”
The public sees numbers – teacher rankings, school rankings, district rankings – that seem data-driven and accurate. We hear that “numbers don’t lie” and teachers who resist being judged in this way just don’t want to be held accountable. This is simply not true. Most teachers work hard and care deeply about their students. We shouldn’t need statisticians to tell us what common sense should: that human learning is the net result of multiple causes that are very difficult to measure in isolation. Personally, I’m ready for some common sense in Raleigh – though statistics will serve.