Noted in brief – 10/19/2009 Posted on 10/19/2009 by Jon Stahl Split-Testing: Are yours statistically valid?