This survey is a collaboration between Minnesota Public Radio and the Center for the Study of Politics and Governance at the University of Minnesota's Humphrey Institute of Public Affairs. The survey was analyzed by the Center.
The research team was Lawrence R. Jacobs (director) and Joanne M. Miller (Department of Political Science). Melanie Burns was the team's research and data analyst.
The survey was fielded by the Center for Survey Research and Analysis (CSRA) at the University of Connecticut, which has extensive national and state experience conducting non-partisan surveys on politics and government policy.
CSRA called a sample of telephone exchanges that was randomly selected by a computer from a list of active residential exchanges within the State of Minnesota.
Within each exchange, random digits were added to form a complete telephone number, thus permitting access to both listed and unlisted numbers. The samples were designed to represent different regions of the state in proportion to the number of adults living in each region.
Within each household, one adult was selected to be the respondent for the survey.
Results are based on a model which adjusts responses according to the likelihood of a respondent voting.
Likelihood to vote is based on the following factors: self-reported probability of voting in the upcoming election, voting in previous elections as reported by the respondent, enthusiasm for the 2008 election, and incorporates differential turnout levels in urban, suburban, and rural areas.
In addition, the results have been weighted to reflect the number of adults in a household and the number of telephone lines reaching the household as well as the demographic characteristics of adults in Minnesota based on region, sex, age, education, and race. Our analysis indicates that 70% percent of Minnesotans who are 18 or over are likely to vote in November.
In addition to sampling error, the practical difficulties of conducting any survey of public opinion may introduce sources of error into the poll. Variations in the wording and order of questions, for example, may lead to somewhat different results.