Am I correct in thinking that they basically only have nine data points? I get that they broke it down by year, but because the values in consecutive years aren't really independent, that doesn't actually increase the power that much, does it?
Not that there aren't a whole bunch of other problems with this methodology, of course.
8 comments:
You need a new category. May I suggest "There will always be an Ann Arbor"?
Perhaps when the Democrats aren't in office they are out killing babies.
Am I missing something or is the paper not available? We just cite the abstract and trust the paper really backs it up?
I did learn one useful thing from the reference: the probably value of any future paper by Rodriguez, Bound, & Geronimus.
;-)
Whenever I read a study like this, I always want to know "What specific policies caused this?"
Am I correct in thinking that they basically only have nine data points? I get that they broke it down by year, but because the values in consecutive years aren't really independent, that doesn't actually increase the power that much, does it?
Not that there aren't a whole bunch of other problems with this methodology, of course.
Claim: Use of the word "ascertain" in the abstract is a sufficient condition for the paper to be of laughable quality.
Bonus: Spelling U.S. as "Us"
Does this mean people tend to kill their babies when a Republican is elected? I'd like to see a suicide correlation study as well.
Post a Comment