Originally posted by SharkD
Originally Posted by sagegrouse
Sorry guys, but the unsupported quote, "scientific survey," makes the hair stand up on the back on my neck. Uh, how was the sample chosen? Random calls to listed numbers (or all phone numbers in NC)? How were respondents chosen? What was the non-response rate? How was the question teed up -- you know, introductory material? Were there other questions in the survey pertaining UNC and Butch Davis? Were there other questions in the survey pertaining to politics or other subjects?
Public Policy Polling has the survey results, questions posed, sample break down and cross-tabs in the full PDF:
PPP surveyed 597 likely North Carolina voters from October 15th to 17th. 159 of those respondents identified themselves as UNC fans. The margin of error on that subgroup is +/-7.8%. Other factors, such as refusal to be interviewed and weighting, may introduce additional error that is more difficult to quantify.
PPP is considered one of the more accurate political polling firms and, as yet, the most accurate founded since 1990. During the 2008 election cycle, in both local and national races, they had the second-smallest margin of error between predicted and actual results. (
http://online.wsj.com/article/SB122592455567202805.html http://www.fivethirtyeight.com/2010/...0-results.html )
That said, the relatively small sample and the potential self-selection bias may have resulted in an exaggerated degree of inaccuracy in these results.