Question:
Q29
If the 2012 election for President were held today, would you vote for Barack Obama, the
Democrat or for Mitt Romney, the Republican?
N %
Obama 336 48%
Romney 280 40%
Undecided (v) 40 6%
Would not vote (v) 21 3%
Don't know (v) 15 2%
Refused (v) 9 1%

Poll Demographics

About this Poll
The Marquette Law School Poll was conducted January 19-22, 2012. A total of 701 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 540 (77%) landline respondents and 161 (23%) 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 701 respondents is +/-3.8 percentage points. For subgroups with smaller sample sizes the
margin of error is larger. For the 309 Republicans and those who lean to the Republican Party the
margin of error is +/-5.7 percentage points. For the 322 Democrats and those who lean to the
Democratic Party the margin of error is +/-5.6 percentage points. For the difference in percentages
between pairs of candidates, the margin of error is larger. For the Walker-Barrett trial heat the
difference is 50%-44%=6 percentage points with a margin of error of +/-7.2 percentage points. For the
Walker-Falk comparison the difference is 49%-42%=7 and a margin of error of +/-7.0 percentage
points. For Walker-Cullen the difference is 50%-40%=10 with a margin of error of +/-7.0 points and
for Walker-Obey the difference is 49%-43%=6 and a margin of error of +/-7.1 points.
Post-Stratification
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 and have fewer years of formal education to exhibit higher rates of nonresponse
resulting in these groups being under-represented in the sample. To compensate for these nonresponse
effects the sample is weighted to bring sample characteristics into line with the population
values. In this sample the population values of age groups and education levels were determined by
combining the 2008 and 2010 Current Population Surveys conducted by the U.S. Census in Wisconsin
to estimate the distribution of age and education for registered voters in the state. Other demographic
characteristics including race, Hispanic origin, sex and region of the state were sufficiently close to the
population values that only age and education were used for estimation of the weights. A “raking”
procedure was used to simultaneously balance the weights so that the sample distribution closely
approximates the known population distributions for age and education. The population, unweighted
and weighted percentages and sample sizes are shown in the table below. Population values for sex,
age, 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 provides
estimates of the percentage of households that have only cellular telephone service.
Other sample demographics include partisanship, religion and marital status. 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. Marital status is based on the American Community Study
conducted by the U.S. Census and is also for all adults.