Several previous studies have demonstrated that student exit polling has educational value and promotes civic engagement (Berry and Robinson 2012, Evans and Lagergren 2007, Lelieveldt and Rossen 2009, and others). The authors of this paper have created assignments and an instructor's manual on running student exit polls in undergraduate courses. Three institutions used these assignments during the fall 2012 semester. Working together, these instructors created an opportunity for their students to participate collaboratively with others in survey design and data analysis. This effort further provided assessment data on the effectiveness of this pedagogical approach for student engagement outside of the classroom in different communities and course contexts. Student surveys measured the impact that this experience had on their understanding of their own community, their relationship to the national community, their understanding of survey methodology, and descriptive statistics. Do students learn more about their community or the scientific process? Does it matter whether the course is designed primarily around politics, statistics, or public opinion? This paper addresses these questions and how these effects vary across different types of students and classrooms.