The overwhelming factor that will drive the outcome of the election will be beyond Romney’s control: the state of the economy in the run-up to the election. What is sometimes also a key factor – whether we are at war, and how the war is going – similarly so. Most of what will determine who wins or loses in the general election, in other words, will have nothing to do with Romney as a nominee, or the state of Romney’s campaign. Nonetheless, whether Romney wins or loses, analysts will try to discern whether that campaign did anything to make a difference.

Some of that analysis will center on demographic factors. These were fairly important in 2008, though obviously (given the margin of victory) not decisive. Barack Obama’s candidacy drove higher African-American turnout than usual, which probably made a difference in winning states like North Carolina; it’s also pretty clear that antipathy to his candidacy was higher than the Democratic norm among Appalachian whites, resulting in a worse relative performance than John Kerry’s in states like Kentucky and West Virginia. Analogously, an enthusiastic Mormon vote should help Romney at the margins in must-win-but-vulnerable states like Arizona, Colorado and Nevada, but if white evangelicals show up in decreased numbers in the general election, and Romney risks losing a state like Virginia or North Carolina, many observers will attribute this to an aversion to a Mormon President. (Conversely, if they show the same enthusiasm they usually do for Republican candidates, those same observers will conclude that Romney’s Mormonism was only a handicap in the primaries – partisanship trumps sectarian rivalry.)

But demographic factors relate only tangentially to issues. What about Romney’s actual campaign?

From most perspectives, the right thing to expect from a Mitt Romney general election campaign is generic Republican-ness. He’s going to call for less regulation and lower taxes on business as the solutions to our economic problems. He’s going to take firm stands against “appeasement” and for “American exceptionalism.” He’s going to promise to appoint judges who “interpret the Constitution” rather than “making law.” And mostly he’s going to criticize the President from every possible vantage. Inasmuch as Mitt Romney runs a “generic” Republican campaign, there’ll be very little to learn from it.

There is one issue, though, where Romney has distinguished himself from recent Republican nominees: immigration. John McCain, the last Republican nominee, was well-known to be a strong proponent of liberal immigration reform. George Bush similarly. Romney, on the other hand, has made a point of endorsing restrictionist positions in the primaries, and of attacking his opponents (Perry and Gingrich, most notably) for “apostasy” on this question. Romney is certainly not a restrictionist’s dream candidate – his overall pro-business orientation makes it unlikely he would propose or support immigration measures that business opposes, and he has said many times that he supports increased legal immigration. But I would argue that, based on the positions he’s taken in the primaries, he would be positioned further in the restrictionist direction than any recent nominee of either party.

It remains to be seen, of course, whether Romney sticks to those positions, or “evolves” for the general election. On the assumption that he does run a relatively restrictionist campaign, the interesting question to me is: how will we know whether it has any effect on the election?

One place to look – and the place where reporters are already looking – is the Latino vote. President Obama’s approval rating has dropped substantially among Hispanic voters since 2008 – but that’s true of his standing among the electorate generally, mostly because of perceptions of an inadequate response to the economic crisis. A variety of Latino leaders have criticized Obama for his own immigration stands – for the substantial uptick on the enforcement front and for the failure to pass the DREAM Act. If, relative to his overall national performance, Mitt Romney underperforms among Latinos relative to John McCain, that reasonably will be taken as evidence of a negative reaction among Hispanic voters to his immigration stances. If he doesn’t change his positions, and doesn’t underperform, that reasonably will be taken as evidence that immigration politics isn’t as important to Hispanic voters as commentators often suggest, at least not when unemployment is above 8%.

But what about the other side of the equation? If Romney does run a relatively restrictionist campaign, that’ll be because he thinks it’s a winning position to take with some set of voters that he needs to keep in his column. Can we identify who these voters are?

The difficulty with identifying immigration-restriction-voters is that a very large number of voters express contradictory preferences on the subject – for example, simultaneously reacting negatively to amnesty and negatively to deportations or to enforcement mechanisms aimed to “encourage” self-deportation. Moreover, I would expect the strongest immigration-restrictionist voters are already reliable anti-Obama voters. If Romney does better among, say, Appalachian whites than McCain did in 2008, is that an indication that immigration-restriction was a “winning issue” with these voters, or is it evidence of Obama’s deepening unpopularity in this segment of the population? Confounding the issue further is the fact that, based on his performance in the primaries, Romney starts out a bit behind with this core demographic (because of his wealth, his tenure as governor of Massachusetts, his Mormonism, etc.) – so perhaps if he runs merely as strong as McCain (again, relative to his national numbers) with this demographic, that’s evidence that he did “something” to earn more support than he would otherwise have obtained, and that “something” could be immigration restriction.

As it happens, while I don’t expect it to be an important point of emphasis, I also don’t expect Romney to change his immigration stance for the general election. If I’m right, I’d be very interested to see a sophisticated read of the data to try to determine whether the issue turns out to make a difference, positive or negative, and where.

Perhaps Nate Silver and Steve Sailer could collaborate on this one?