The RAND Continuous 2012 Presidential Election Poll

The RAND Continuous 2012 Presidential Election Poll (CPEP) was conducted within the American Life Panel, an Internet panel recruited through traditional probability sampling to ensure representativeness. Because the CPEP asks the same respondents repeatedly about their voting preferences, observed changes are attributable primarily to individuals changing their minds and not to random sampling fluctuations. The CPEP asks respondents to state in terms of percent chance both their preferences for a candidate and the likelihood that they will vote. Moreover, we asked the respondents about their actual voting after the election, so we can study the predictive power both within sample and out of sample (the national results). The CPEP appears to have predicted well. Our final prediction of the difference in popular vote between Obama and Romney differed about half a percentage point from the final tally, which would place it near or at the top of the polling firms. The probabilistic questions, even months before the election, were strongly related to actual voting behavior. Our approach allows us to gain insights into stability of voting preferences and the effect of events on individual preferences; for example, various shifts can be clearly related to major events.