Abstract
The public record of the number of enrolled US Citizens in 2014 living on the University of Wisconsin River Falls campus is compared to the public records of the number of voters who used a residence hall address to vote in the November 2014 elections. While there was a low turnout of students voting using their campus address, a significant difference was found in the rates of voting for those with the on campus polling location compared to those with an off campus polling location. Students who had an off campus polling location showed 5.7% of students casting a valid vote compared to 10.0% of students with the on campus polling location.
Polling Location: The Impact of Off Campus Voting For Students in Residence Halls
(You Don’t Have To Go Home, But You Can’t Vote Here)
Polling Location: The Impact of Off Campus Voting For Students in Residence Halls
(You Don’t Have To Go Home, But You Can’t Vote Here)
In this study the voter turnout impact of having an off campus polling location is compared to an on campus polling location for University of Wisconsin River Falls students living on campus in residence halls. It is hypothesized that having a polling location off campus will result in reduced rates of students voting using their university address as their voting address. The study will use public records to determine the percent of eligible students who vote using their campus address for each of the dorms. The percentage turnout for voters with an on campus polling location will be compared to those with an off campus polling location.
Literature Review
Much attention in political science focuses on factors that influence voter turnout. While there is some commonality to all populations, some sub groups have factors that influence them to a greater degree than the general population. Students are one such group that is studied for specific areas of impact. General areas related to these studies include: barriers to voting, contact methods, impact of social pressure, distance to polling location, absentee balloting, and the institutional attitude toward voting.
Voting can be a more difficult process for some groups than others. College students living in residence halls face many sometimes compounding factors limiting their voter turnout. Some of these identified in research include: having recently moved, being recently new or first time voters, lacking information about process or candidates, and facing a confusing legal situation for their voting rights. Ulbig and Waggener (2011) examined voter registration and information provision impacts on turnout. They noted that while college students vote at higher rates than their cohorts not in post-secondary education they vote at a lower rate than older citizens (Ulbig). Timpone (1998) examined many factors including age and registration impact on turnout. Again age is found to be a predictor of involvement, and registration was also strongly linked to turnout (Timpone). Niemi and Hanmer (2010) found that while many of the normal factors that influence variable turnout outside of gender had little impact, distance from home was inversely related to turnout. They also found that students were likely to vote strategically, choosing to vote in a battleground state when possible (Niemi). Gimpel and Schuknecht (2003) look at distance to the ballot box in relation to turnout, but are examining the general population and not students specifically. They note the importance of knowledge of polling location, and that changes should be effectively publicized (Gimpel).
What form contacts take is a common area of study. Dale and Strauss (2009) examine some of the political theory behind voter mobilization, and compare personal with impersonal contact as a mobilization tool. Text messaging was found to be an effective low-cost method of increasing turnout. The authors noted that while personal contacts have the greatest impact on already interested voters, they bear a high cost in time and volunteer availability. They studied a group of twelve thousand voters registered by Student PIRG’s and Working Assets Group out of a pool of one hundred and fifty thousand voters registered by the groups. The impact of a single text message on the day before election to the test group was found to increase turnout 3% (Dale). While this method involved no personal contact it was low cost and involves information and technology that is likely available to universities for their students. While not limited to university students the location and targeting of the studies would have resulted in a very significant student population in the studies.
Email is readily available and a common outreach tool employed by university administrations. No longer having the novelty of the ‘you’ve got mail!’ days of the internets past these messages lack the oomph that once may have existed. It is unfortunate that these messages do not work, as the marginal cost of sending emails approaches zero. Email outreach as a tactic to promote voting has been shown to not help “Regardless of the type of email received, students turned out to vote at about the same rate. Again, it is likely that students simply ignored a text-only email that was sent by a faculty member with whom they were not taking a class” (Ulbig p. 549). Emails as a tool to promote registration and turnout have even been shown to decrease registration slightly, depressing the voter turnout (Bennion and Nickerson 2010). Centering “get-out-the-vote” policy on low cost and low effort seems to lead to low turnout. If universities want to increase turnout they must choose effective strategies “If colleges and universities really emphasized voting and political involvement the way they publicized home sports events and campus activities, many more students would probably be persuaded to participate” (Dalton and Crosby 2008 p. 3).
Social pressure has been found to be a major factor in voter turnout. Gerber, Green, and Larimer (2008) tested several message with varying degrees of social pressure to evaluate effectiveness. This large scale study involved a control group of 100,000 households and 4 test groups of 20,000 households each. Each of the test groups were sent postcard reminders. Two held generic reminders on the importance of civic participation. One reminded voters’ that if they voted or not was public record, and sent them a copy of their voting record. The last card also sent the voting history of others in the neighborhood, and promised to follow up with another after the election. The mailing with the voters own information was markedly more successful and the one with the neighbors’ information by far the most successful (Gerber). This method involves social pressure, and may not be within the range of activities seen as appropriate by university administrators. However informing students about the nature of public records and that their information is available is not only appropriate; but also related to the obligation already in law regarding informing students about their public records.
Distance to voter location has been found to be a factor impacting turnout. Researchers have found “even relatively minor differences in the distances to polling places in cities significantly affects voter turnout” when looking at distances of 2-6 miles (Timpone). This finding has been confirmed in Atlanta where even a small change of distance was found to impact turnout. (Haspel and Knotts 2005). Fixing this barrier is simple “reducing the travel costs associated with voting can mobilize more voters” (Dyck and Gimpel 2005). It should be noted the Dyck was looking at costs in terms of both time and expense. In a study of college students there was found to be an inverse relationship between distance to the polls and turnout (Niemi). Cost-benefit analysis for a voter is often an explanatory reason. As costs in terms of time increase the value of being a voter is lower in comparison. It can be expected that these pressures of distance will not be lesser for first time voters, or those who have not established patterns of voting.
Combining polling locations, thus increasing the travel time, has also been found to reduce turnout significantly (McNulty Dowling and Ariotti 2009). Splitting precincts and reducing travel time has the opposite effect and serves to increase turnout (Haspel 2005). While these studies focused on turnout for the general public, it can be expected that this factor will also be important for students.
Placing polling locations on campus can result in “outstanding” turnout. (Carpini and Frishburg 2005). Carpini was basing his observations off of voter turnout being higher than expected at two Iowa universities where on campus voting was instituted, but did not numerically quantify the difference. The observation from the other side has found that a lack of a campus polling location limits the student vote (Kolasky and Wondolowski n.d.). Kolasky also looked at redistricting being used to reduce student voter power (Kolasky). These references to campus polling location are not measured or quantified and only presented as anecdotal information.
Despite the knowledge of the impact of campus polling location, it has not been studied extensively. In some cases all students will not have a campus voting location, in others all will. The cases where both polling locations exist provide for a quasi-experimental condition for study. Depending on the degree of confidence required, populations could be more or less closely matched for study. Considering the large number of campuses in the United States that may have multiple polling locations, it should be possible to determine if this pattern holds true in many institutions. The degree to which this impact holds true has important policy implications for university students. Students may have a right to vote at home or by absentee ballot, but those choices are not always the most desirable for the voter. By reducing barriers to voting at a university address overall turnout for the age group could be increased.
While absentee balloting might reduce the impact of long travel, Wisconsin voters cannot go online to request an absentee ballot (Kolasky). In Los Angeles County the result of polling location changes were to reduce voter turnout, about 1/3 of the loss was compensated with increased absentee balloting (Brady and McNulty 2011). Unfortunately youth were more inclined to not vote than increase their absentee balloting rate (Brady). While the Brady study did not focus on students as a group the results can be expected to apply to this population also. In Wisconsin requests for an absentee ballot must be received by the Municipal Clerk in writing by 5pm the Thursday before an election, so this option will not be viable for those who decide to vote after that time. It is likely that without a focused effort to increase its use absentee balloting will not be a major factor of compensation for voters living on campus.
The importance of Democracy to our society is well recognized “Free and open elections are among the most sacred privileges in a democratic society” (Clayton 2004). The University of Wisconsin River Falls even recognizes the value of citizenship in its focused mission statement “Our mission is to help prepare students to be productive, creative, ethical, engaged citizens and leaders with an informed global perspective” (University of Wisconsin River Falls n.d.). Engaged citizenship forms from a basis of participation in our democracy, but it is appears to not be something that happens in the absence of effort toward the cause.
Some universities have taken official effective action to increase turnout, such as Hofstra in New York “Attending a multi-perspective, one-day undergraduate conference on issues facing the nation held three weeks before the presidential election produced a significantly higher voter turnout rate” (Bogard, Sheinheit, and Clarke 2008). The Bogard study focused specifically on student voting. Knowing what issues an election encompasses has been found to increase turnout, and in this case the applied knowledge resulted in the expected results.
Other universities have neglected their role “if colleges and universities took the requirements seriously, we would see increases in student voter participation” (Kolasky). The solution is action “colleges must allocate resources and staff time for mobilization efforts. Administrators should make voter registration — including absentee applications — a part of fall orientations, move-in events, and class registration” (Carpini). The lack of attention of the role of higher education in supporting society appears to be a developing downfall of both Democracy and the role of the university in society “both local boards of elections and university and college administrations could engage in much greater outreach activity to get students to register and vote” (O’Loughlin and Unangst 2006).
Research Design
Institutional review Board Approval
As this study involves human subjects, Institutional Review Board approval was sought, the research protocol was approved on December 9 2014 and assigned protocol number W2014 – T203. The study population was University of Wisconsin River Falls students who lived in residence halls on campus during the Fall Semester of 2014 and are US Citizens and over the age of 18. This resulted in a study population of 2128 individuals.
Background
University student voting has implications for the success of the United States democratic experiment. Funding for universities and students needs can also be linked to electoral participation by students. The past studies of student voting have identified many factors that influence turnout. Many suggest that an on campus polling location results in the best turnout, but no studies of a campus split into multiple polling locations have identified to what degree this is a factor. Since 2012, the University of Wisconsin River Falls has had two polling locations for the approximately two thousand two hundred students who live on campus. One polling location is on campus, in the same building students use for dining and student recreation activities. 1533 students living in seven residence halls were eligible to use the on campus polling location during the 2014 November elections. The other polling location is located about 1.5 miles road distance from the residence halls that use it for voting, at the River Falls public high school. 595 UW – River Falls students living in three different residence halls were eligible to use the High School voting location for the November 2014 elections.
Method
The count of US Citizens over 18 who lived in residence halls was requested from Institutional Research at the University of Wisconsin River Falls. Seven residence halls used campus as a polling location, voting in the University of Wisconsin River Falls ‘University Center’ building. These 1533 potential voters lived in the residence halls: Crabtree, Grimm, Hathorn, McMillan, Parker, South Fork Suites, and Ames Suites. Three residence halls used the River Falls High School as a polling location. These three dorms accounted for 595 potential voters in Johnson, May, and Stratton halls. As the name implies the University Center is a much more central place for university students than the River Falls High School
The Wisconsin Government Accountability Board is in charge of maintaining the state voter file. From this board the records of votes cast in the November 2014 election were received in an excel file. This record was sorted by address, then the number of voters using each address was counted. Votes cast in person early at the City Clerk’s office, through an absentee ballot using the residence hall as primary address, or in person on Election Day are counted through this method. The polling books from the election were also examined, votes found to have been cast in the wrong precinct were determined as these votes would not count as a valid vote during a recount. In Wisconsin there cannot be more ballots counted than valid votes cast, when this situation exists it is called an overvote. The result of an overvote is a “drawdown” of the ballots, removing one from the ballots cast by random draw until the number of ballots does not exceed the number of eligible voters. (Plunkett 2011) Two residence halls, Ames Suites and South Fork Suites, changed their address over the summer of 2014. Votes using both the new and old address were counted as valid if they were cast in the correct precinct as they would likely be allowed to be counted during a recount.
Evaluation of Findings
Results will be compared of the on and off campus groups and a Z-ratio determined for these groups as a whole. Results will also be shown for each residence hall, and for the two groups. A z-ratio in excess of 2.33 for a greater than 2% average difference in voter turnout will be assumed to confirm the hypothesis.
Implications of Research
Voter turnout and electoral participation are markers of the health of a Democracy. Barriers to electoral participation should be evaluated for their impact so that they can be properly weighed when compared to other factors such as accessibility of a polling location, ward and district boundaries, legislatively mandated district size, post census redistricting, and the cost of staffing multiple polling locations during elections. This research should add valuable factual information about student voter turnout to help in these evaluations.
Findings
153 valid votes were cast by the 1533 students who vote on campus, resulting in a 10.0% turnout. 34 valid votes were cast by the 595 students who vote at the River Falls High School, resulting in a 5.7% turnout. This results in a z-value of 3.19 corresponding to a less than 0.1% chance these results are not significant, with an overall turnout difference of 4.3% of the eligible voter pool studied.
The results for each dorm are shown in Figure 1. Figure 2 shows the results if invalid votes are included. Table 1 show the results for all the dorms, and includes numerical count of attempted votes in addition to the percent of attempted and valid votes.
The unexpected findings related to the number of votes that would have been ruled invalid during a recount. Nearly ¼ of the students who attempted to vote, and should have used the off campus polling location, instead voted at the on campus polling location. None of the students who should have voted on campus voted in the wrong location. This level of failure in both the process checking and in voter information indicates a need for tighter checks at the polling location and better voter education.
Despite the high statistical significance and the disparity in voting between the groups the results should be approached with caution. Confirmation of finding from prior years should be checked. Voting in other potential locations should be check and ascertained to not replace votes lost on campus. There are significant limitation to this study, some of these are noted in the following section.
Limitations of Study
This research only determines the likelihood of casting a vote using the campus address as the primary voting address. It is possible that people knowing in advance that their polling location is not convenient will chose to vote using their parents address either in person or through an absentee ballot. A more robust design would account for total likelihood of voting but would be much more complex as voting records for each of the home states would need to be examined. Self-reporting of voting is found to have a large desirability bias so would not be a choice method for study.
Outside of the number of US Citizens over 18 years of age in each dorm there was no evaluation of other demographic considerations. Known confounding factors of the sample include generally older students in Ames Suites and South Fork Suites. Other factors could include different sex balance in the halls, differing policies taken up by Resident Advisors or hall staff, local actions by outside groups, and the impact of one or several activists on their neighbors in the residence halls. By looking at large populations many of the confounding factors will be lessened in impact.
The year 2014 was studied. Other years of data are available and would make for a more complete picture. Changing elections laws, uncertainty over Voter ID requirements, and new proof of residency and student status requirements may have resulted in a more suppressed student vote than in prior years. It is likely that in future elections these changes will happen in a more timely fashion, or will rely on rules already in place from prior elections so that information will be disseminated in a more timely fashion to voters. Changes in application of Wisconsin law regarding voter ID were ongoing up until a month before the election, and there was a great amount of confusion about how the rules would be applied even among trained voting registrars and poll workers. In addition two residence halls changed their address in the summer prior to the election and this information was not well publicized, with some sources continuing to show students their old address. This uncertainty likely added to the confusion of students and suppressed the student vote more than it would be in a normal year. Votes cast using the incorrect address but in the correct precinct were counted as they would likely be ruled as valid voters in the precinct during a recount.
The study does not account for other methods of voting. All university students eligible to vote may choose to vote in person at their home address or their university address. They may also be able to vote by absentee ballot. Further study is needed to confirm that the depression in voting caused by off campus polling location is not made up by increased voting from home or with absentee voting. Prior years should be examined to confirm the impact found for 2014. Information from 2012 for the split campus voting and 2006, 2008, and 2010 when all students in residence halls voted on campus is available and would add to the depth of this study.
Conclusion
Having a polling location off of the University of Wisconsin - River Falls campus reduced the turnout of voters in the November 2014 election compared to having an on campus polling location. Among the potential voter pool of US citizens living in the residence halls the absolute difference in turnout was 4.3%, with 5.7% of off campus voters compared to 10.0% of on campus voters using their university address to vote. The absolute difference will likely vary from election to election but it is reasonable to expect the same trend.
While low in absolute difference this results shows a significant increase in voting percent, and may show a much larger absolute difference in higher turnout elections. This is the first study of this type, and many potential problems exist with the methodology employed. Further research should be completed to confirm these findings.
The low turnout of voters in this election, and the large number of invalid votes cast, point to a need for much greater effort from the University and the City Clerks in providing election information in a method that is effective. That non-university students of the same age vote at even lower rates indicate a true failure in American democracy, and one that should be rectified.
The youth of today not only inherit the largest nuclear arsenal on the planet they also will come to possess the history and future of America. The sooner this generation comes to understand the magnitude of their role, and starts to participate and contribute, they better they will be prepared for the inevitable day when they are in control. Presuming that voting is socially desirable it is surprising that civic involvement is not a higher priority for todays students.
Appendix
http://images.dailykos.com/...
Figure 1
http://images.dailykos.com/...
Figure 2
2014 US Citizens per dorm and voter turnout
Dorm Count Attempted Vote Percent Valid votes percent
May 156 11 7.1 5.8
Ames 235 26 11.1 11.1
Crabtree 118 15 12.7 12.7
Grimm 207 21 10.1 10.1
Hathorn 315 34 10.8 10.8
Johnson 286 21 7.3 5.2
McMillan 197 15 7.6 7.6
Parker 232 17 7.3 7.3
South Fork Suites 229 25 10.9 10.9
Stratton 153 10 6.5 6.5
Grand Total 2128 195 9.2 8.8
Table 1
Bibliography
Bennion, E., & Nickerson, D. (2010). The cost of convenience: An experiment showing e-mail outreach decreases voter registration. Political Research Quarterly, 64(4), 858-869.
Bogard, C., Sheinheit, I., & Clarke, R. (2008). Information they can trust: Increasing youth voter turnout at the University. PS: Political Science & Politics, 41(3), 541-546.
Brady, H., & McNulty, J. (2011). Turning out to vote: The costs of finding and getting to the polling place. American Political Science Review, 105(1), 115-134.
Carpini, M., & Frishberg, I. (2005, December 1). Colleges should foster growth in young-voter turnout. Retrieved December 10, 2014, from http://repository.upenn.edu/...
Clayton, Dewey M. (2004). A funny thing happened on the way to the voting precinct: A brief history of disenfranchisement in America. The Black Scholar, 34(3), 42-52.
Dale, A., & Strauss, A. (2009). Don't forget to vote: Text message reminders as a mobilization tool. American Journal of Political Science, 53(4) 787-804.
Dalton, J., & Crosby, P. (2008). Student voting and political engagement in college: Should higher education be doing more to promote civic agency? Journal of College and Character, X(1).
Dyck, J., & Gimpel, J. (2005). Distance, turnout, and the convenience of voting*. Social Science Quarterly, 86(3), 531-548.
Gerber, A., Green, D., & Larimer, C. (2008). Social pressure and voter turnout: Evidence from a large-scale field experiment. American Political Science Review, 102(1), 33-48.
Gimpel, J., & Schuknecht, J. (2003). Political participation and the accessibility of the ballot box. Political Geography, 22(5), 471-488.
Haspel, M., & Knotts, H. (2005). Location, location, location: Precinct placement and the costs of voting. The Journal of Politics, 67(2).
Kolasky, E., & Wondolowski, L. (n.d.). Not home, not welcome: Barriers to student voters. Retrieved December 10, 2014, from http://goodspeedupdate.com/...
McNulty, J., Dowling, C., & Ariotti, M. (2009). Driving saints to sin: How increasing the difficulty of voting dissuades even the most motivated voters. Political Analysis, 17(4), 435-455.
Niemi, R., & Hanmer, M. (2010). Voter turnout among college students: New data and a rethinking of traditional theories. Social Science Quarterly, 91(2), 301-323.
O'Laughlin, M., & Unangst, C. (2006, January 1). Democracy and college student voting (Third Edition). Retrieved December 10, 2014, from https://www.salisbury.edu/...
Plunkett, B. (2011, April 28). WI State Senator under recall hammer caught in Supreme Court recount. Retrieved from http://www.dailykos.com/...
Timpone, R. (1998). Structure, behavior, and voter turnout in the United States. The American Political Science Review, 92(1), 145-158.
Ulbig, S., & Waggener, T. (2011). Getting registered and getting to the polls: The impact of voter registration strategy and information provision on turnout of college students. PS: Political Science & Politics, 44(3), 544-551.
