# 2012 Presidential Election Poll

The RAND continuous presidential election poll offered a unique perspective on voter intent during the 2012 campaign between Barack Obama and Mitt Romney.

The poll was conducted within the American Life Panel, an internet panel recruited through traditional probability sampling to ensure representativeness.

It asked the same respondents repeatedly about their voting preferences, which means changes are due to individuals' changing their minds and not due to random sampling fluctuations.

The poll was also different because it asks respondents to state their preferences for a candidate and the likelihood that they will vote in probabilistic terms (percent chance), which has been shown to improve forecasts several months before the election.

The questions asked by the poll were:

• What is the percent chance that you will vote in the Presidential election?
• What is the percent chance that you will vote for Obama, Romney, someone else?
• What is the percent chance that Obama, Romney, or someone else will win?

The questions are asked daily of about 500 members of the RAND American Life Panel. The results were weighted to produce a nationally representative sample and the daily updates report averaged weekly results.

## Election Forecast

The Election Forecast provides our best forecast of the popular vote based on the responses that panelists provided in the past week. The gray band indicates if the difference between the estimates for the two candidates is statistically significant. If the lines for Obama and Romney lie outside the gray band, then with at least 95-percent confidence we can say that one candidate would win the election if on election day the citizens vote as they now anticipate. It is important to note that if the lines are within the gray band then the observed differences may be due to chance.

It is also important to note that the predictions combine the percent chance of voting for a candidate with the percent chance that a respondent will actually vote. For example, if someone says in response to the first question that he or she has a 50-percent chance of actually voting, then this person's response to the question of who they will vote for gets a "weight" of 50 percent in the calculation of our prediction (which is then further weighted by the percent chance that he or she says that they are likely to vote for their chosen candidate).

From 7/11/2012 to 11/6/2012, the forecast for the outcome of the 2012 presidential election was plotted by day, based on ALP panelists' responses. The forecast generally showed Barack Obama as the winner of the election, with the gap between Obama and Mitt Romney widening near the end of the survey as Obama moved ahead. For much of the survey period, the difference between the two candidates was small enough that it may have been due to chance.

The graphs below show the Election Forecast broken out by respondent race/ethnicity.

From 7/11/2012 to 11/6/2012, the forecast for the outcome of the 2012 presidential election was plotted by day, based on Hispanic ALP panelists' responses. The forecast consistently showed Barack Obama as the winner of the election, with between 58% and 74% of the vote.

From 7/11/2012 to 11/6/2012, the forecast for the outcome of the 2012 presidential election was plotted by day, based on Non-Hispanic white ALP panelists' responses. The forecast consistenlty showed Mitt Romney as the winner of the election, with between 52% and 57% of the vote.

From 7/11/2012 to 11/6/2012, the forecast for the outcome of the 2012 presidential election was plotted by day, based on Non-Hispanic African-American ALP panelists' responses. The forecast consistently showed Barack Obama as the winner of the election, with between 86% and 94% of the vote.

From 7/11/2012 to 11/6/2012, the forecast for the outcome of the 2012 presidential election was plotted by day, based on other ALP panelists' responses. The forecast consistently showed Barack Obama with more votes, although some of the observed differences may have been due to chance

## Predicted Winner

This graph presents the forecast by the panelists of who will win the election. This need not be the same as who one votes for. One can vote for one candidate, but still think that the other candidate will win.

From 7/18/2012 to 11/6/2012, ALP panelists were more likely to think that Barack Obama would win the 2012 presidential election. 51% to 57% of panelists thought Obama would win, while 39% to 43% of panelists thought Romney would win.

The graphs below show the predicted winner broken out by respondent gender.

From 7/18/2012 to 11/6/2012, female ALP panelists were more likely to think that Barack Obama would win the 2012 presidential election. 52% to 58% of panelists thought Obama would win, while 37% to 41% of panelists thought Romney would win.

From 7/18/2012 to 11/6/2012, male ALP panelists were more likely to think that Barack Obama would win the 2012 presidential election. 50% to 56% of panelists thought Obama would win, while 41% to 45% of panelists thought Romney would win.

## Shifts Between Candidates

This graph shows us how many panelists shifted their vote from Obama to Romney or vice versa in the past week. Notice that these are generally small percentages, which causes this particular graph to be rather "noisy".

On any given day from 7/18/2012 to 11/5/2012, the percent of ALP panelists who shifted their vote from Barack Obama to Mitt Romney ranged from about .7% to 1.6%. The percent of panelists who shifted their vote from Romney to Obama ranged from about .8% to 1.9%. Generally, the shifts were so small that they may have been due to chance.

These graphs show how many panelists, broken down by age, shifted their vote from Obama to Romney or vice versa in the past week.

On any given day from 7/18/2012 to 11/5/2012, the percent of ALP panelists under age 40 who shifted their vote from Barack Obama to Mitt Romney ranged from about .7% to 2.3%. The percent of panelists who shifted their vote from Romney to Obama ranged from about .7% to 2.4%. Generally, the shifts were so small that they may have been due to chance.

On any given day from 7/18/2012 to 11/5/2012, the percent of ALP panelists from age 40 to 64 who shifted their vote from Barack Obama to Mitt Romney ranged from about .5% to 1.5%. The percent of panelists who shifted their vote from Romney to Obama ranged from about .6% to 1.5%. Generally, the shifts were so small that they may have been due to chance.

On any given day from 7/18/2012 to 11/5/2012, the percent of ALP panelists age 65 or over who shifted their vote from Barack Obama to Mitt Romney ranged from about .2% to 2.6%. The percent of panelists who shifted their vote from Romney to Obama ranged from about .7% to 3.2%. Generally, the shifts were so small that they may have been due to chance.

## Intention to Vote

The next graph shows the "intention to vote" by candidate preference. Supporters of one candidate may be more likely to vote than supporters of another candidate and the graph shows how this differs between the Obama supporters and the Romney supporters. Recall that, as explained above, the weekly poll takes the differences in voting intentions into account.

From 7/11/2012 to 11/5/2012, supporters of Mitt Romney were more likely than supporters of Barack Obama to say that they intended to vote. 82% to 88% of Romney supporters said they intended to vote, while 78% to 85% of Obama supporters said that they intended to vote. For both candidates, the number of people who said the intended to vote gradually increased over the survey period.