National Journal joins Reuters and Pew in showing Barack Obama back to a 50% or better rating on job approval.  The new level in the NJ/Allstate poll is Obama’s best in the series since May 2011, taken after the killing of Osama bin Laden, but he still trails on his re-elect numbers:

Boosted by growing optimism about the economy, President Obama is showing signs of reassembling the coalition of support that powered him to his 2008 presidential victory.

In the latest Allstate/National Journal Heartland Monitor poll, Obama’s approval rating rose to 51 percent, up from only 44 percent in each of the previous polls taken in October and December. That’s Obama’s highest approval rating in the Heartland Monitor since the survey taken immediately after the killing of Osama bin Laden in May. Other than that, the new survey marks the first time that Obama’s approval rating in the poll has crossed the critical 50 percent threshold since September 2009.

The survey suggests the overall verdict is split—Obama’s support is strengthening but hardly yet secure, and the country remains divided closely enough on his performance and agenda to virtually ensure a competitive general election against the Republican nominee. Indeed, just 44 percent of registered voters surveyed said they intend to vote for the president’s reelection, while 49 percent said they will likely or definitely vote for someone else. A series of economic measures also shows that Obama is continuing to receive equivocal ratings, especially from whites.

But the most revealing measure of electoral strength for an incumbent is his overall approval rating, and on that front the survey, conducted from March 3 to 6, showed Obama recording clear progress. Compared with the December survey, the new poll found his approval rating rising by 11 percentage points among independents; 8 among nonwhites; 6 among all whites; 7 among both college-educated white men and women; and 9 among the so-called waitress moms—white women without a college degree. Only among noncollege-educated white men did Obama remain stuck in neutral with virtually no gain from December.

You know what else in this poll is similar to Reuters and Pew?  The composition of the sample.  NJ didn’t provide a link to the data in its article, but supplied it to me on request — and the D/R/I is about what you’d expect.  The sample gives Democrats an eight-point advantage, 34/26/35, which both overstates independents and vastly underrepresents Republicans.  Even in 2008, when Democrats surged to the polls after eight years of George W. Bush, the exit polls showed a seven-point advantage for Democrats, 39/32, which mirrored Obama’s seven-point victory in the popular vote.  In 2010’s midterms, exit polls showed a 35/35/30 split, which means that either the poll undersampled Republicans by six or nine points, depending on which turnout model one presumes this general election will most closely resemble. On top of that, the approval numbers are based on general-population adults, not the subsample of registered voters.

The re-elect numbers numbers come from the registered voter sample, though, and they’re not good at all for an incumbent — especially one whose opposition is in the middle of a tough primary fight.  Even with a skewed sample with an eight-point advantage for Democrats, only 27% of men and 32% of women say they will definitely vote for Obama in 2012, and only 31% overall.  That’s actually lower than the percentage of women in the sample who are Democrats (37%) and men as well (32%).  Both are also lower than the percentage that believe Obama increases opportunity for people like them to “get ahead” (30% among men, 32% among women, and 31% of overall respondents).  Obama isn’t even convincing his own party’s voters, let alone anyone else.

In comparison, 39% in this skewed sample say they will vote against Obama in 2012, thirteen points higher than the GOP representation in the sample.   The same is true in the gender breakdowns, with 38% of men decided against Obama and 36% of women, both higher than those who have decided to vote for Obama, and much higher than Republican representation in the gender demo (27% of men, 26% of women).   This poll is bad news for Obama’s re-election chances even with the skew — and probably would have been much worse with proper sampling.

Update: The 2008 exit poll data can be found here, and the 2010 exit poll data here.