Question:
If the election for U.S. Senate were held today, would you vote for (the Democrat Tammy Baldwin) or for (the Republican Tommy Thompson)?
1 Baldwin
2 Thompson
(DO NOT READ)
3 Undecided (VOL)
4 Would not vote (VOL)
8 Don't Know
9 Refused

About this Poll
The Marquette Law School Poll was conducted October 11-14, 2012. A total of 1,007 Wisconsin registered voters and eligible voters who said they would register by Election Day were interviewed by a combination of landline and cell phone using random digit dialing (RDD). Interviews were completed with 703 (70%) landline respondents and 304 (30%) cell phone respondents. The data collection was managed by LHK Partners Inc, Newtown Square, PA. The margin of error for a single percentage in a sample of 1,007 respondents is +/- 3.2 percentage points. For subgroups with smaller sample sizes the margin of error is larger. For the 870 respondents who are likely to vote in the presidential election in November, the margin of error is +/- 3.4 percentage points. Likely voters are any voters who indicated they were "absolutely certain to vote" or had "already voted either by absentee ballot or early in person voting" in the November election.

Post-stratification, or weighting, compensates for patterns of non-response that shift sample characteristics from known population values. In telephone surveys it is common for potential respondents who are younger to exhibit higher rates of non-response resulting in these groups being under-represented in the sample. To compensate for these non-response effects, the sample is weighted to bring sample characteristics into line with the population values. In this sample, the population values for age, sex, and education were determined by combining the 2008 and 2010 Current Population Surveys conducted by the U.S. Census in Wisconsin. The distribution of population by media market (DMA) was provided by Claritas, a media research firm. A “raking” procedure was used to simultaneously balance the weights so that the sample distribution closely approximates the known population distributions for age, sex, education, and region. Other demographic characteristics including race, Hispanic origin, and marital status were sufficiently close to the population values that age, sex, education, and region were used for estimation of the weights. The population values, unweighted percentages, weighted percentages, and sample sizes are shown in the table below.

Population values for sex, age, marital status, race, Hispanic origin, and education are based on the Current Population Survey for 2010 and 2008. The distribution of population by media market (DMA) was provided by Claritas, a media research firm. The Centers for Disease Control and the National Center for Health Statistics provide estimates of the percentage of households that have only cellular telephone service. The Gallup organization provides estimates of party identification and religious affiliation based on their polls in the state over a six to twelve month period. The “leaned” party identification classifies independents who say they are closer to a party as supporters of that party. Gallup’s data is based on all adults while the Marquette Law School Poll samples registered voters.