Here's a good article titled Closing the Racial Learning Gap on the La Griffe du Lion web site. While I'm not statistically sophisticated enough, having forgotten nearly all my college statistics, to follow all of the math involved in the appendices the articles on this site are always entertaining and lead to non-intuitive conclusions.
In this article the author shows that decreasing gaps in black-white scores on standardized tests do not necessarily show that the gap is actually decreasing but may instead mean only that the threshold for passing has changed. Given a couple assumptions the author's model then points to the likelihood that claimed gap reductions in North Carolina and Texas over the past few years are only artifacts of the testing procedures themselves.
Thursday, April 22, 2004
Racial Testing Gap
Posted by Dan Agonistes at 9:48 AM
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment