By CHRISTINE MATTHEWS
    
WASHINGTON - Republican primary voters in Indiana say they want a U.S. Senator “to focus first on trying to solve many of our country’s problems, even if that means working with elected officials across the aisle to do it,” (60%) versus a U.S. Senator “to focus first on standing up for conservative principles, even if that means not working with elected officials across the aisle to solve problems” (30%).
        
Yet, on the cusp of the May 8th election, they appear poised to elect someone who believes “We don’t need bipartisanship as much as we need the application of principle.” (Richard Mourdock, Evansville Courier & Press, April, 2012).   Challenger Richard Mourdock leads incumbent Senator Richard Lugar by a 43%-35% margin which expands to 48% to 38% when initial undecideds leaning toward a candidate are included.
    
So, what to make of this apparent contradiction?  
    
Actually, Mourdock supporters are about evenly split on this question:  48% want a problem solver vs. 46% want conservative principles first.  Lugar supporters, however, overwhelmingly want someone to solve problems first (78%) vs. putting principles ahead of bipartisan problem solving (17%).
    
However, while voters may differ on what brand of conservative they want to send to Washington, and even express a preference for bipartisan problem solving, the race really boils down to this: Close to three-fourths of voters supporting Richard Mourdock say it’s because of Lugar’s longevity or age or the fact that he lives in Virginia. These voters think Indiana’s senior senator has served too long.  The comments relating to Lugar being too liberal, or Obama’s favorite Republican, or voting for his Supreme Court Justices account for just about 15% of the reasons given for supporting Mr. Mourdock.  In fact, while the Senator’s support for the DREAM Act is one of the main arguments Tea Party groups make against Senator Lugar, the majority of GOP primary voters say they support it (54%), while only about one-third oppose it.  And Lugar’s support of the DREAM Act did not register as a reason, specifically, for supporting Mourdock.
    
What’s happened since our last survey in late March is that Mr. Mourdock is now capturing a majority of the vote among those who identify as Republicans (51% - 36%) whereas Senator Lugar has remained very close to the level of support he got in March (38%) when the two candidates were tied.  All the voters we interviewed for this survey have a past history of voting in a Republican primary, but roughly one-fourth identify initially as independent.  Among these self-identifying independent voters, Senator Lugar has a narrow edge (41%-38%).  He needs a strong showing among independent-leaning primary voters to compensate for Mr. Mourdock’s lead among Republicans.  However, because the race appears to hinge as much, if not more, on longevity as ideology, adding independents may less dramatically accrue to Senator Lugar’s benefit. [1]  
    
Unlike a general electorate, the majority of Republican primary voters tend to be male and they are breaking toward the challenger by a 52%-36% margin.  Among women, the race is closer with 44% for Mourdock and 39% for Lugar.  Nearly six in ten of the remaining undecided voters are women.
    
A plurality (44%) of primary voters identifies themselves as “very conservative” and, not surprisingly, they overwhelmingly favor Mr. Mourdock (63%-26%).  So the balance of the electorate is somewhat conservative (33%) or moderate (19%) and, to win, Mr. Lugar would have to dominate among these voters. The problem for Senator Lugar, is that Mr. Mourdock is also winning the less conservative voters (43%-41%), so that the senator carries only moderate voters (60%-26%).
    
Generally, the rule of thumb is that undecided voters break two or three to one for the challenger.  That may be true here, but the demographics of the undecideds give Lugar a shot.  Nearly six in ten are women with whom he fares better. They are less conservative and more independent leaning than voters overall.  The endorsements of Governor Mitch Daniels and the Indiana Right to Life carry the most weight with them, as with voters overall. They are much less likely to have an opinion of either candidate (33% favorable – 24% unfavorable for Lugar and 21% favorable – 17% unfavorable for Mourdock), but have become more unfavorable toward both over the past few weeks – mostly likely as a result of all the negative television ads.
    
About one third of voters say their opinion of Senator Lugar has become less favorable over the past week or so, compared with about a quarter trending negative for Mr. Mourdock.
    
While Lugar has a slight edge over Mourdock on general electability, enough voters think both would fare equally well against Democrat Joe Donnelly that this argument doesn’t help much.  And Mourdock has a similar edge on what may be a more important voting consideration, which is, “shares my values.”  
    
We conducted our polling on Monday (April 30) and Tuesday (May 1), ending a full week ahead of the May 8 primary.  Of course, the last week before a primary campaign can be very fluid; things can change.  But, without any new data to conclude otherwise, it certainly looks as if the momentum has shifted toward Richard Mourdock.

Footnote: [1]  Of course, one could argue that this primary will attract some voters with no history of voting in a Republican primary.  In addition to the 700 interviews we conducted, we also dialed a separate list of voters with no primary vote history but who voted in the 2008 and 2010 general elections.  We did not dial known Democratic primary voters.  We stopped calling after the first night, after obtaining 50 interviews because:  1.) the calls are expensive due to the lower number that qualify as likely GOP primary voters and 2.) they were breaking for Mr. Mourdock in a similar fashion to our 700 sample of past GOP primary voters.  Those 50 additional interviews are not included in our results. Our thought was to obtain these interviews in case we needed to model what an expanded electorate might look like in the event of a very close contest. While we only did 50 interviews, if the voters were breaking strongly for Senator Lugar and the race were virtually tied, an argument could be made that this could be a factor, but only with a race that is close enough that it could be decided by non-traditional voters.

Matthews is president and CEO of Bellwether Research, is a graduate of Indiana University and a native of Kokomo.