[BrandonS]

That math doesn’t make any sense. This chart says that 45% of republican voters are at least somewhat in favor of gay marriage. If we take your numbers and say even 53% of the total population is, then to get an average of 53% with republicans at 45%, we would need democrats to be at 61%, which certainly seems more than plausible (if we assume the population is split evenly by party for simplicity).

I don’t know where you got the idea that a chart showing 45% equates to “more than half”.

[Steve_Seattle]

"I don’t know where you got the idea that a chart showing 45% equates to “more than half”."

Maybe I expressed myself poorly. Here is what I meant to say:

I think about one-third of the country identify as Republicans, and about 53% of the country support gay marriage. Assuming that 80% of Democrats and 60% of independents support gay marriage, that doesn't leave room for 45% of Republicans to support gay marriage, as claimed by the survey. To get to the 53% national figure, way less than 45% of Republicans (about 20%) would have to support gay marriage.

If the country is 1/3 Republican, 1/3 Democrat, and 1/3 independent, and if my estimates of Democratic and independent support for gay marriage are close, a 45% showing of support for gay marriage among Republicans would give a total public support for gay marriage of about 64%, not 53%, which would be a huge jump in support in the span of less than a year.

My math is basically:

80% x .33 + 60% x .33 + 20% x .33 = 52.8%, which is close to the poll figures I've seen, and very different from the poll cited in this article. If you plug 45% into the Republican part of the equation, you end up with public support of gay marriage being slightly more than 64%, and I've never seen a figure that high.

So what I was trying to say is that - unless there has been a huge, sudden change in Republican opinion - only about 20% of Republicans would support gay marriage, not 45%.