Research

1. The 2012 electorate contained the smallest share of white voters and the largest share of nonwhite voters in American history. White voters constituted 72 percent of the electorate, down from 74 percent in 2008, 77 percent in 2004, and 81 percent in 2000. African-Americans made up the next largest share at 13 percent, the same as 2008, and up from 11 percent in 2004 and 10 percent in 2000. Hispanics constituted 10 percent of the electorate in 2012, compared to 9 percent in 2008, 8 percent in 2004, and 7 percent in 2000. Asian voters made up 3 percent of the 2012 electorate, an increase from 2 percent each in 2008, 2004, and 2000.
2. Mitt Romney won a larger share of the white vote than either John McCain or George W. Bush. Romney defeated Obama by 59 to 39 percent among whites, compared to McCain winning whites by 55 to 43 percent over Obama in 2008, while Bush won whites by 58 to 41 percent over John Kerry in 2004 and by 54 to 42 percent over Al Gore in 2000.
3. Mitt Romney won white voters in almost all demographic groups, usually by substantial margins. Romney's campaign was extremely successful at appealing to white voters across the board, and won almost all white groups except Jewish voters. In every group listed below, Romney's percentage in 2012 surpassed McCain's percentage in 2008.

   

   
    
4. Romney lost among African-Americans by roughly the same margin as John McCain and by a greater margin than George W. Bush. Obama won African-American voters by 93 to 6 percent in 2012 and 95 to 4 percent in 2008. John Kerry defeated Bush among African-Americans by 88 to 11 percent in 2004, and Al Gore won among African-Americans by 90 to 9 percent in 2000.
5. Romney lost Hispanic voters by a greater margin than either John McCain or George W. Bush. Obama defeated Romney among Hispanics by 71 to 27 percent in 2012, and he defeated McCain among Hispanics by 67 to 31 percent in 2008. Kerry defeated Bush among Hispanics by 53 to 44 percent in 2004, the high water mark for a Republican presidential candidate. Gore won the Hispanic vote over Bush in 2000 by 62 to 35 percent.
6. Romney lost Asians by the greatest margin in recent history, and by a greater margin than he lost Hispanics. Obama won among Asians in 2012 by 73 to 26 percent, compared to 62 to 35 percent over McCain in 2008. Kerry defeated Bush among Asians in 2004 by 56 to 44 percent, while Gore defeated Bush among Asians in 2000 by 55 to 41 percent. Bob Dole was the last Republican presidential candidate to carry the Asian vote, winning 48 percent to 43 percent for Bill Clinton and 8 percent for Ross Perot in 1996.
7. The partisan makeup of the electorate in 2012 was closer to 2008 than either 2004 or 2000. In 2012 Democrats outnumbered Republicans by six percentage points, 38 to 32 percent. That was close to the seven-point Democratic advantage in 2008 of 39 to 32 percent, and more than the even partisan breakdown in 2004 (37 percent each), or the four-point Democratic advantage of 39 to 35 percent in 2000.
8. Romney won among Independents, but not by enough to overcome the Democratic advantage in the electorate. Romney won Independents by five points, 50 to 45 percent. But given the Democrats' six-point advantage in the electorate, his margin among Independents fell just short of what he needed to win a majority of the popular vote.

1. Obama won among early voters, Romney won among absentee voters, and the candidates split evenly on Election Day. Among people who voted in person early, Obama won 58 to 41 percent. Among those who voted by absentee ballot, Romney won by almost as large a margin, 55 to 45 percent. Among people who voted on Election Day, the candidates split evenly at 49 percent each.
2. Unlike most incumbent reelection campaigns, Obama won late deciders. Voters who are undecided late in a campaign usually break disproportionately for the challenger. For example, in 2004 voters who made up their minds in the last week (11 percent of all voters) voted for John Kerry over George W. Bush by 52 to 46 percent. But Obama broke that pattern this year. This post-election poll shows voters who decided "during the last week before the election, or on the day of the election" voted for Obama by 48 to 40 percent. Exit polls show the same pattern, with voters who decided "just today or in the last few days" (9 percent of all voters) voting for Obama by 50 to 44 percent.
3. President Obama's handling of Hurricane Sandy appears to have had a modestly positive effect on late deciders. Obama's handling of the hurricane led 18 percent of all voters to say they were more likely to vote for him versus 8 percent who said they were less likely to support his reelection, with 73 percent saying it had no effect. But that result is not relevant for those voters who made up their minds before the hurricane struck on October 29, eight days before the election. While the number of respondents is small, in this survey among voters who made up their minds during the last week before the election, or on the day of the election, 15 percent said Obama's handling of the hurricane made them more likely to vote for him, 6 percent said less likely to vote for him, and 77 percent said it made no real difference in their vote. Obama's handling of Hurricane Sandy, and the publicity it garnered for him as a bipartisan leader in its aftermath, is the most likely explanation for why Obama broke the pattern of late deciders breaking for the challenger.
4. By a two-to-one margin, Romney voters say they voted for Romney rather than against Obama. Fifty-five percent of Romney voters say they were voting primarily for Romney rather than against Obama.
5. By an eight-to-one margin, Obama voters say they voted for Obama rather than against Romney. Eighty percent of Obama voters say theirs was a positive vote for Obama rather than a negative vote against Romney.

1. The powerful relationship between an incumbent's job approval and his standing on the reelection ballot is borne out again in this election, with Obama's ballot standing almost identical to his overall job approval. Obama's job approval in the 2012 electorate was 51 approve and 47 disapprove, extremely close to his ballot victory of 50 to 48 percent. Ninety-four percent of Democrats approved of his job performance; 92 percent of Democrats supported his reelection. Forty-six percent of Independents approved of his job performance, compared to 45 percent of Independents who voted for him. Six percent of Republicans approved of his job performance, and 6 percent of Republicans voted for his reelection. Obama's job approval hovered in the upper 40s throughout the fall, right on the bubble for incumbent presidents seeking reelection. Among those who turned out to vote, his job approval was just enough to eek out a victory in the popular vote.
2. Based on the difference between Democratic and Republican approval of his job performance, Barack Obama is the most polarizing president at the time of his reelection in recent history. Ninety-four percent of Democrats approve and 6 percent of Republicans approve of Obama's job performance, a difference of 88 percentage points. In the last Gallup poll before the 2004 election, 93 percent of Republicans approved of George W. Bush's job performance and 11 percent of Democrats approved, a difference of 82 percentage points. In the last Gallup survey before the 1996 election, 86 percent of Democrats approved of Bill Clinton's job performance and 19 percent of Republicans approved, a difference of 67 percentage points.
3. A powerful relationship also exists between Obama's job approval on handling the economy and his ballot standing. Resurgent Republic has argued throughout the fall that Obama's job approval on the most important issue in the election would be an excellent predictor of his vote, and that turned out to be the case. Forty-nine percent of the 2012 electorate approved of Obama's handling of the economy compared to his 50 percent standing on the ballot.
4. A majority of the 2012 electorate holds the Obama Administration at least somewhat responsible for the current state of the American economy. Seventy-two percent of 2012 voters say Obama is at least somewhat responsible for the state of the economy, 36 percent very responsible and 36 percent somewhat responsible. Fifteen percent say his Administration is not too responsible, and 12 percent say not at all responsible. Republicans hold Obama most responsible for the economy – 55 percent very and 35 percent somewhat responsible. Thirty percent of Independents say Obama is very and 45 percent somewhat responsible. Even a majority of Democrats say he is at least somewhat responsible – 25 percent very and 30 percent somewhat responsible.

1. Romney drew a split on key characteristics of presenting a credible plan for the economy and better able to work with both parties, but Obama led on helping the middle class. As a general rule, Republicans associated all the positive characteristics with Romney, as Democrats associated positive characteristics with Obama. Distinctions between the two are driven by differing perceptions among Independents. Romney had slight overall edges on presenting a credible plan to fix the economy and being better able to work with both parties, and a wider edge for thinking America is an exceptional nation. Obama had the largest lead on any characteristic tested, however: would help the middle class. His advantage on that characteristic is particularly pronounced in the battleground states, with a 56 to 39 percent margin.
   

   
   The following table ranks various characteristics from those most associated with Romney to those most associated with Obama:
   

   

   
    
2. Romney did pass the threshold of being a qualified candidate, with two-thirds of voters saying they “view Mitt Romney as someone who has the qualifications and experience to be president.” The overall figure of 67 percent includes virtually unanimous agreement by Republicans, and 73 percent of Independents, with the overall figure limited by the agreement of just 35 percent of Democrats.
3. As seen in our final pre-election survey, both presidential nominees and their running mates ended the election with net positive personal ratings. President Obama held a 56 to 43 percent rating overall (53 to 45 percent among Independents), virtually the same as Mitt Romney’s 56 to 42 percent rating (62 to 35 percent among Independents). Paul Ryan’s rating of 50 to 39 percent slightly exceeded that of Joe Biden’s (50 to 43 percent), with a notable edge among Independents (55 to 36 percent for Ryan versus 44 to 49 percent for Biden).

1. For the first time in our polling dating back to April 2009, voters say the country is headed in the right direction. The slim 50 to 48 percent overall margin of voters saying the country is heading in the right direction is buoyed by a high percentage of Democrats saying the country is headed in the right direction (90 to 8 percent), and a shift among Independents (from 30 to 64 percent in late October to 46 to 51 percent now).
2. The top concerns of the 2012 electorate are consistent with pre-election polling, with economy and jobs as the top issue. While it comes as no surprise that the economy and jobs are the top issue overall and among Independents, there is a striking difference between the overall electorate and Independents regarding Medicare: while a majority of voters overall say Medicare was extremely important, that figure drops to just 39 percent among Independents (48 percent among Republicans but 68 percent among Democrats). Only social issues (abortion and gay marriage), immigration, and the war in Afghanistan were less important issues than Medicare to Independent voters.
   

   
   The following table ranks various issues according to the percentage of voters who gave each issue a five (extremely important) on a one to five scale:
   

   

   
    
3. One likely explanation for why Medicare was relatively less important to Independent voters is that this survey shows Republicans won the Medicare argument. Voters were given some basic information about Paul Ryan’s Medicare plan, and asked their view of the plan:
   

   
   Republican vice-presidential candidate Paul Ryan has proposed making significant reforms to Medicare. Under Ryan’s plan, Americans under 55 years of age would receive premium support from the government, which would allow them to receive traditional Medicare or could be used to pay for part or all of a private insurance plan. Meanwhile, Medicare would remain the same for Americans 55 and older. Which of the following comes closer to your opinion?
   
   Paul Ryan's Medicare plan would preserve and protect this program, saving us trillions of dollars over the long run and keeping Medicare from going bankrupt while giving seniors more choices and control over their own health care.
   
   Paul Ryan's plan would end Medicare as we know it, leaving senior citizens at the mercy of greedy insurance companies and forcing them to pay for part of their own health insurance which could lead to cuts in life-saving medical treatment.

   Voters say the Ryan Medicare plan would preserve and protect by a 52 to 35 percent margin, including a 50 to 34 percent margin among Independent voters (Republicans agree by a 79 to 13 percent margin, while Democrats disagree by a 31 to 56 percent margin).

   Seniors are persuaded that the Ryan plan would preserve and protect Medicare along with younger voters, by a 51 to 31 percent margin among those 65 and older, a 54 to 35 percent margin among voters between the ages of 30 to 64, and a 46 to 41 percent margin among voters under 30.

4. Voters continue to support repealing and replacing the 2010 health care reform law. By a 54 to 38 percent margin, with a nearly identical margin among Independents (55 to 38 percent), voters support repealing and replacing the President’s primary legislative achievement. Just a narrow majority of Democrats oppose repealing and replacing the law (51 percent, while 39 percent support repealing and replacing the law), while Republicans continue to support repealing and replacing it, now by a 70 to 24 percent margin.

1. A majority of the electorate favors a role of government based upon a conservative definition of opportunity rather than a liberal viewpoint of fairness. By 54 to 41 percent, voters believe “government policies should promote opportunity by fostering job growth, encouraging entrepreneurs, and allowing hardworking people to keep more of what they earn” instead of “government policies should promote fairness by narrowing the gap between the rich and poor, making the rich pay their fair share, and reducing income inequality.” Independents are closer to Republicans on this question, 57 to 37 percent and 80 to 17 percent, respectively. Democrats side with the fairness message 64 to 30 percent.
   

   
   There are also cautionary signs when looking at key cohorts of President Obama’s winning coalition. Women voters favor the conservative opportunity message (53 to 42 percent), and suburban women do so by a slightly wider margin (55 to 41 percent). The opportunity theme also garners majority support among young voters 18-29 (52 to 44 percent), and Hispanic voters split evenly (tied at 47 percent).
   

2. A plurality of voters is hesitant to support a more activist role of government. By 49 to 46 percent, the electorate is more likely to support the viewpoint that “government is doing too many things better left to businesses and individuals” compared to “government should do more to solve problems and help meet the needs of people.” As we have seen before, Independents (53 to 40 percent) look more like Republicans (80 to 17 percent) when considering the role of government, while Democrats believe government needs to do more (75 to 19 percent).
   

   
   Once again, there are signs that conservatives can make inroads with President Obama’s winning coalition over the debate of the proper role of government. Women voters overall slightly favor the conservative message (48 to 46 percent), and a majority of suburban women do so as well (52 to 44 percent). Majorities of young voters 18-29 and Hispanic voters side with the more liberal statement, but do so at a smaller margin than their support of President Obama (55 to 42 percent and 62 to 32 percent, respectively).
   

1. Republicans makeup a 3-point Democratic edge on the generic congressional ballot by winning Independents by 7-points. Independents favor the Republican candidate on the generic ballot by 47 to 40 percent. This margin is 2-points greater than the presidential exit polling and the primary reason why Republicans maintain majority control of the House of Representatives today. Both political bases broke evenly for their respective candidate: Republicans (93-3) and Democrats (94-2).
2. Independents are more likely to hold an unfavorable view of congressional Democrats than congressional Republicans. For Democrats in Congress, the upside-down margin widens to 10-points among Independents when considering their favorable/unfavorable image (39/49). Republicans in Congress are perceived more favorably by Independents (46/45). Among voters overall, Democrats in Congress have a positive favorable/unfavorable image (47/42), while the perception of Republicans in Congress also splits evenly (44/45).

1. Independent voters were contacted evenly by both the Obama and Romney campaigns. The two campaigns reached out to critically important Independents at equal percentages and frequencies: 36 percent of Independents were personally contacted by the Obama campaign and 37 percent of Independents were personally contacted by the Romney campaign. When considering frequency, the Obama campaign contacted Independents once (7 percent), two to four times (11 percent), and five or more times (18 percent). The Romney campaign contacted Independents once (6 percent), two to four times (11 percent), and five or more times (20 percent).
2. While more Democrats were reached by the Obama campaign than Republicans by the Romney campaign. The Obama campaign strategy of turning out its base is reflected by the fact that 57 percent of Democrats were personally contacted, while 45 percent of Republicans were personally contacted by the Romney campaign. The margin is smaller when considering those who were contacted five or more times, 28 percent of Democrats by the Obama campaign and 25 percent of Republicans by the Romney campaign.
   

   
   Just 43 percent of the voters were not contacted by either campaign. That drops to just 23 percent among battleground voters. Nationally, 21 percent were contacted by both campaigns, 14 percent were contacted by Romney but not Obama, and 18 percent were contacted by Obama but not Romney. In the battleground states, 37 percent were contacted by both campaigns, 5 percent were contacted by the Romney campaign but not Obama, and 19 percent were contacted by Obama but not Romney.
   

3. While 8 in 10 voters talked to family and friends about the election, significant campaign activity took place on social media platforms and campaign websites. The national electorate visited both candidate websites evenly: the Obama website (21 percent) and the Romney website (20 percent). Yet more voters took action on social media platforms like Facebook or Twitter: 32 percent shared political comments and 44 percent read political comments. When considering partisan breakouts, Republicans were more active than Democrats. Thirty-five percent of Republicans shared political comments, compared to 30 percent of Democrats. This margin also corresponds with our findings on those who read political comments on social media (45 percent of Republicans and 41 percent of Democrats). Perhaps more noteworthy, Independents had the highest response on this category at 47 percent.
4. Voters still view cable and broadcast television news as their main sources of political information, followed by a combination of online platforms. Seventy-six percent of the electorate considers television programming their main source of political information: 44 percent cable and 32 percent broadcast news. Online platforms equal a combined 60 percent: 41 percent Internet news sites, 11 percent political blogs and 8 percent social media.

Mitt Romney did a superb job of winning the support of white voters, sweeping almost every white demographic group and rolling up a landslide margin among whites. But that was not enough to win a presidential election in the America of 2012. While many other factors affected the outcome – the Obama campaign's tactical superiority in turnout in battleground states, strong turnout among young voters, a gradually improving economy, Hurricane Sandy's assist for Obama among late deciders, unfortunate comments from some candidates that offended many women – nothing was as important as the shifting demographics of the electorate. Romney's performance among white voters would have been sufficient to put him in the White House in any election before 2008.

The handwriting is on the wall. Until Republican candidates figure out how to perform better among non-white voters, especially Hispanics and Asians, Republican presidential contenders will have an extraordinarily difficult time winning presidential elections from this point forward.

This post-election survey of one thousand 2012 voters nationally was conducted November 6-8, 2012. Respondents were selected randomly from a random-digit-dialing sample including both cellular and landline telephone numbers, and were contacted by live interviewers. Thirty percent of respondents were reached using the cellular sample; of all respondents, including those reached on both cellular and landline telephones, 20 percent of respondents say they receive all their personal telephone calls on a cell phone, 17 percent receive almost all their calls on a cell phone, 29 percent split their calls between cell phones and landline phones, 13 percent receive almost all their calls on a landline phone, and 18 percent receive all their calls on a landline phone.

All respondents confirmed that they are registered to vote in the county in which they live. Quotas were set for state, age, and race based on state registration and previous turnout. Results were weighted to reflect the election results of 50 percent Obama and 48 percent Romney.

The margins of error for responses with an even split – 50 percent for one response and 50 percent for another response – are ±3.10 percent for the full sample, ±5.71 percent for Republicans (295 respondents), ±5.39 percent for Independents (330 respondents), ±5.19 percent for Democrats (357 respondents). The margin of error is smaller when one response receives a higher level of support. For example, the margin of error is ±2.68 percent when 75 percent of respondents in the full sample choose one response and 25 percent choose another response.