TL:DR for the next paragraph: Statistics can be confusing.
I'm not sure I shared this here--I'm evaluating how different 'celebrity spokescharacters' perform for make up targeted to adults. There's a phenomenon now where make up companies are using all types of childish imagery---Hello Kitty, Disney Princesses, Unicorns--to sell make up to adult women (some call this tactic 'peterpandemonium' or 'adultescene'). An interesting article in New York magazine questioned whether this tactic would backfire as the purpose of makeup is generally to empower via confidence in one's appearance. So I looked at reactions to three ads: one featuring Hello Kitty, one featuring Miss Piggy, and one featuring Wonder Woman, and compared them to a 'control ad (no celebrity spokescharacter) and I found yes, exactly that--the celebrity spokescharacters do not perform as well (in terms of purchase intentions) as a traditional model. So I just figured out the best statistical analysis after trying to come to grips with the differences between ANCOVA and MANCOVA and MANOVA but now I don't have to because I did a regression.
Anyway--in other news, I picked up Kildalton--I haven't worked on this for a while, and it took a few minutes to get back into the Kildalton groove.