Conducted by telephone Oct. 16-21, 2012 by Issues and Answers Network Inc., the poll completed 650 interviews with randomly selected registered voters in Tennessee. The poll has an error margin of plus or minus four percentage points at the 95 percent level of confidence.
In order to obtain a representative sample of respondents, data were collected using a mix of landline and cell phones. Additionally, quotas were placed on landline and cell phones to ensure a proper mix of the two groups. The sample contained only registered voters and was representative of the state population.
Interviews averaged 12 minutes in length. Weights were applied to the data to match the sample's gender and race proportions to those seen in exit poll data from the 2008 election in Tennessee. However, a comparison of results with and without the weights showed little difference.
The poll defined "likely voters" as registered voters who had voted already or who had either voted in the 2010 governor's race in Tennessee and/or described themselves as "very likely" to cast a vote in the upcoming presidential election. The results reported are for the 609 poll respondents who met the likely voter criteria.
All poll results are available on the MTSU Poll website.