Subgroups in NCLB
April 17, 2006
I understand, and somewhat agree with, the aspect of the No Child Left Behind Act that allows states to determine their own measures of success. Doing so allows for more local control and makes it harder for special interest groups to enshrine their preferences in a "national curriculum" of sorts (though it makes it a lot harder to compares success across states – something that researchers would certainly appreciate). However, leaving these choices to states can create loopholes for those who would prefer less effective measurement of their success. For example, witness one state’s standard for how many students is too few to constitute its own subgroup for NCLB reporting (via Joanne Jacobs):
Oklahoma lets schools exclude the test scores from any racial category
with 52 or fewer members in the testing population, one of the largest
across-the-board exemptions. That means 1 in 5 children in the state
don’t have scores broken out by race.
Other states’ quotas are much lower. It doesn’t take much examination to guess that this serves as an easy out for schools that don’t want the public to know how well (or, more to the point, how poorly) they are serving their minority populations.