From a reader:

Jonah,
This is my first time writing; I appreciate the site and love your work.
I have the politics/genetics article in front of me (the actual one,
in APSR, not the NYT one). In response to the questions that one of
you other readers brought up:
–Yes, the researchers have data on parents (one of their controls is
for “shared environment,” basically if the twins were raised together
or separately–thereby controlling for parental influence)
–I don’t see info about whether the parents “talked politics” at home
or were partisan. This data was unavailable.
–It does not discuss if parents were of different partisanship. It
is unclear (to me, at least) what relationship, if any, this would
have on the empirical findings. Would we expect genetics to be
stronger in split households? Weaker/nonexistent? Unless the reader
is making an argument about the sample (which is large–enviably so),
I’m not sure about this point.
As to his big criticism, the authors are quite clear. Fraternal twins
are more likely than identical twins to differentiate. This is,
basically, the whole point of the article. Rather than dismissing the
study as “pretty silly,” a good social scientist would actually read
the journal article and examine the findings. Causality is difficult,
yes, and this article does not “prove” that genetics causes policy
preferences. It does, however, provide evidence for that point and
shoudl be confronted on its own terms and not because we tend be
averse to genetic explanations for social phenomena.
(Note also that the article concedes that the attitudes in question
are of the “rough outline” variety. They explicitly state that Party
ID is weakly explained by genetics, if at all.)
Thanks for the great work,