> "[A]ccording to the Pew Research Center for the People and the Press,
> people who attend church more than once a week vote Republican by 63
> percent to 37 percent; people who seldom or never attend vote Democratic
> by 62 percent to 38 percent"
ATTENTION: NUMERICAL SLEIGHT OF HAND IN PROCESS. Offhand, I'd say the number of people who attend church an *average* of *more* than once a week is very small -- and the number of people who seldom or never attend is very big. Especially when you that consider "seldom" includes all the Easter/Christmas Christians, which is the overwhelming majority of Christians I've ever met.
It'd be nice to have hard numbers about those two groups, but offhand it looks to me as if these figures prove exactly the opposite of what Wallis thinks he's proving. They suggest the current set-up favors the Democrats, and that most people are perfectly happy with their lack of emphasis on religion -- probably because they think it's a private matter.
It's also worth pointing out that *neither* group is the one Wallis spends his time presuming to speak for: fairly religious people who go to church with fair regularity, i.e, once a week except when something comes up. Their voting behavior is completely undocumented in this column. So Wallis has zero evidence to back up these arguments. He's just talking out of his behind.
Sadly, I think that's SOP for constructing a newspaper column: write up the argument, then hunt and fish for numbers that look like they back it up. And don't spend any time in trying to honestly crunch them a bit. Just use them as ornaments.
Michael