the source of information you need to answer the question is the exit poll data, and the exit poll data has been used as evidence of fraud.
I just ran the crosstabs for 2004:
With weights applied, Bush won the catholic vote (52.1% of the catholic vote), but without weights applied, he lost it (45%).
So what you conclude depends entirely on whether you think the post-stratification reweighting is valid. We can't assume that
a priori, but, as you may know, I do think (having gone into the question rather extensively) that it almost certainly was - that the poll over-sampled Kerry voters relative to Bush voters, and that would include Kerry catholic voters relative to Bush catholic voters.
And if so, then Bush seems to have won the catholic vote in 2004, whereas he lost it in 2000 (47%)
http://www.cnn.com/ELECTION/2000/results/index.epolls.htmlAs a catholic, that makes sense to me, although it also appalls me. There was a certainly pressure on catholics not to vote for Kerry, and too many catholics do what they are told. Frankly I blame people like Archbishop Chaput for Bush's win:
http://www.nytimes.com/2004/10/12/politics/campaign/12catholics.html?ex=1255233600&en=0ad90158e24ec98f&ei=5090&partner=rssuserlandand the people who paid attention to him, of course.