Lutherology: What affects the way we vote?
In the spirit of the quickly approaching election, this Lutherology report focuses on the factors that shape how Luther students vote. While a previous Lutherology article focused on societal factors that “push†us towards being liberal or conservative, this one examines factors that influence why students prefer either Barack Obama or John McCain.

Our data, collected from a random sample of 245 students, was gathered during Oct. 7-9 by this year’s Sociological Research Methods course. We found that 76% of students prefer Senator Obama, 19% prefer Senator McCain and 6% prefer “other†(margin of error is +/- 6%).
So why does Obama have a 57% lead among Luther students while only a 7% lead in national polls? One primary reason is that nationally, Obama has amassed a 30% lead among 18-29 year olds. We can see that Luther is definitely part of this national trend among young voters.
Aside from this specific statistic, have you ever wondered what factors lead a Luther student to vote one way or another? Let’s examine some likely hypotheses that may explain voting behavior.
We hear a lot about the “gender gap†in politics, with women being more liberal than men, but our data shows a weak link between gender and presidential preferences. A slightly higher percentage of women (78%) preferred Obama, compared to 73% of men, but this difference is within the margin of error of our survey (and therefore are not statistically significant).
Do students become more liberal while attending Luther? Not really, at least not in voting preferences. First-years, sophomores and juniors all showed similar support for Obama (around 74%) as compared to 81% of seniors.
Perhaps the size of one’s hometown gives us some insight into student presidential preferences. Well... somewhat. While students from mid-size towns were most likely to choose Obama (87%), individuals from very small towns or large suburbs were the least likely to prefer Obama (69%). Although size of town may influence voting preferences more than gender, the relatively narrow margin indicates it is not a major predictor of voting behavior for Luther students.
Is religious affiliation an important predictor in choosing a particular candidate?

Our data showed that about 3/4 of Lutheran and Catholic students preferred Obama. Those with no religion affiliation were the most likely to vote for Obama (87%) while the most pro-McCain group (27%) were non-Lutheran/non-Catholic Christians. Students with the same non-Lutheran/non-Catholic affiliation demonstrate the lowest support for Obama (62%). This pattern may be caused by the adoption of evangelical Christians’ ideals of traditional moral values (e.g., pro-life, protecting heterosexual marriage) within the Republican Party, as opposed to the progressive secular values supported by liberals. Students who favor legalizing gay marriage strongly support Obama (87%) while those who oppose it slightly favor McCain (48% to 44%).
In the 2004 Bush-Kerry presidential election, a clear pattern emerged: the higher a person’s income, the more likely it was they would vote for Bush. This pattern is not surprising, considering that liberals want to “tax the rich†to re-distribute wealth to all and conservatives want to cut taxes on high income earners to reward success and promote job growth (insert your favorite “Joe the Plumber†quote here). Does that pattern hold for Luther students? No. There is almost no correlation between a student’s family income and their candidate preference. All income groups favored Obama between 74% and 79%.
If gender and income have no effect on Luther students’ voting preferences, and religion and town size only have a small to moderate impact on candidate preference, what factors really help predict Luther students’ voting behavior? Two factors: parents and one’s own political identity.
Not surprising to sociologists, parents play a central role in the political socialization of Luther students. If a Luther student has two “Obama parents†they are 98% likely to support Obama themselves. If parents are split, a student is 80% likely to support Obama. Having “McCain parents†seems to be the only force strong enough to withstand Obamamania. If a student has two “McCain parents,†they are 57% likely to support McCain - by far, the most pro-McCain group in the sample.
The second important factor in predicting the voting preferences of Luther students is political identity. 97% of self-identified liberals support Obama and 81% of conservatives support McCain. While there are twice as many liberals as conservatives at Luther (which accounts for much of Obama’s lead at Luther) the big surprise is that 3/4 of moderates support Obama while only 14% support McCain. Obama also has the support of 3/4 of those who “don’t know†their political ideology. While 90% of our sample said they were planning to vote, we’ll have to wait until election day to see how many young adults visit the polls and which Presidential candidate wins the vote of today’s youth.







