Methodology

Study findings are based on a telephone survey of 800 South Carolina residents, 400 of which are registered, likely Republican Primary voters, and 400 of which are registered, likely Democratic Primary voters. The sampling error for each party is +/- 4.9 percentage points at the 95% confidence level.  The study was conducted between September 9 and 23, 2007.

For the purposes of our branding study, we wanted respondents who were more likely to be interested in the race and better informed about the candidates.  To this end, we culled our list of registered voters so that our sample included those who had voted in one or more elections starting with the November election of 2004, as well as those who are now under 21 and registered since the last presidential election.  Further, respondents were screened so that only those who said they are "very" or "somewhat" likely to vote in the 2008 Primary were included in the study.

We asked a series of questions to assess public perception of the candidates.  Scores are derived in the following way:

• "Familiarity" is gauged by name recognition.
• For "Reputation" and "Personality," raw scores measure the percentage of respondents who rank the candidates positively on certain attributes.  To derive a net score, we subtract the percentage of negative impressions of the individual candidates from their percentage of positives.
• Finally, "Performance" and "Connectivity" scores are derived by taking the average of a series of questions designed to measure his or her brand attribute.