This terrific graph below plots Romney's advantage over Santorum on a weekly basis, among men and women.
Note that both lines have gone up and down since December. The big plunge in both lines came in and around the time of Santorum's winning performance on Feb. 7 in voting in Minnesota, Missouri, and Colorado. The rise in the lines after that reflects Romney's recovery of his lead.
But, and this is the important point, note how closely the two lines -- one representing the Romney lead among men and the other representing the Romney lead among women -- track each other. The big tidal forces affecting Republican voters' choice for their party's nominee are affecting both men and women about equally. And, the closeness of the lines to one another shows us that Rick Santorum is not, as some have speculated, losing support disproportionately among women compared with his support among men.
So, no matter what comments and policy positions Santorum has taken in recent weeks relating to women's issues, nothing seems to have affected his relative position among women in any way that ithasn't affected his relative position among men.
Now, religiosity among Republicans is another thing. We do see a growing differential relationship between church attendance and support for Romney over Santorum.
As you can see in the graph, the line representing the Romney advantage among weekly church-going Republicans began in early January to drop below the lines representing Romney's advantage overSantorum among the two groups that attend church less frequently.
In other words, religion increasingly began to matter as the GOP race moved into January and February.
Now, in our latest weekly average through March 11 (the farthest right-hand point on the graph), 15 percentage points separate the Romney advantage between weekly church-attending Republicans and among Republicans who seldom or never attend church. To be more specific, Romney leads Santorumby only 3 points, 34% to 31%, among weekly church attenders, but he leads by 18 points, 35% to 17%, among those who seldom or never attend church. In fact, Gingrich does as well as Santorum among this latter group.
Santorum has obviously positioned himself in a very advantageous way among highly religious Republicans. The interesting thing about this, of course, is that Romney -- a faithful Mormon -- is far from unreligious himself. In fact, taken as a whole, Mormons in the U.S. are much more religious on average than are Catholics, Santorum's religion.
Santorum does particularly well among weekly church-going Protestants. In fact, he beats Romney among this group by a 35% to 25% margin. Romney wins among less religious Protestants.
Among weekly church-going Catholics, Romney wins 35% to 26% -- and wins among all other Catholics as well.
Taken as a whole, Romney leads Santorum among all Catholics by 18 points (last week) -- a fascinating figure given that Santorum himself is a prominent Catholic. For that matter, Gingrich is a Catholic as well, and he does even worse than Santorum among those who identify with his own faith.
It appears to me that highly religious Protestant Republicans, the ones often called "evangelicals," are searching for a candidate who fits with their view of the world. Santorum has two characteristics that apparently fit this view. First, he has emphasized strongly his commitment to typical family values issues -- opposition to abortion, concern about contraception, and the sanctity of the traditional family unit. Second, he is not a Mormon - a religious identity that still causes concern among highly religious Protestants.