Understanding the attitudes and perceptions of a community's members towards public transportation is valuable when designing and delivering service. In this paper, the design and descriptive statistics for the first two waves of a longitudinal study of students enrolled at North Dakota State University are presented. The use of a longitudinal as opposed to a traditional cross-sectional survey instrument provides the ability to identify individual changes in attitudes and behaviors instead of relying on aggregate results, which often mask important differences that occur over time. Analysis of first-year data using a binary probit model finds that vehicle access, on-campus residence, prior transit use when traveling, and use of transit by family and friends significantly influence ridership behavior. Ridership by members of the survey cohort increased from 33% during their freshman year to 45% during their sophomore year.