Thanks for the article, Robert. It was also written by one of the Freakonomics guys!
Yes, sorry if I wasn't clear. It is the actual study that Freakanomics itself is referencing in the portions about abortion.
I confess my eyes glaze over when I see something like "ln(ARRESTSstb) = b1ABORTsb + ãs + ëtb + èst + åstb)" or "ln(CRIMEst) = b1ABORTst + XstÈ + ãs + ët + åst".
That's why there are two authors of Freakanomics, Steven Levitt is the guy who did the hard thinking and Stephen J. Dubner is the guy who can help write it in an engaging and accessible way.