129 Transportation Research Record: Journal of the Transportation Research Board, No. 2322, Transportation Research Board of the National Academies, Washington, D.C., 2012, pp. 129–137. DOI: 10.3141/2322-14 Civil and Environmental Engineering Department, Old Dominion University, 135 Kaufman Hall, Norfolk, VA 23529. Corresponding author: A. J. Khattak, akhattak@odu.edu. Survey (NHTS)] use random digit dialing associated with landlines in targeted geographic areas, university student populations are often missed. University students are more frequently mobile-phone-only users (recent surveys show that the percentage at Old Dominion University is nearly 60%). Their travel behavior is neither well understood nor properly modeled in regional travel demand models. Notably, the traditional travel demand model considers trip rates with socioeconomic information, such as household size, automobile ownership, and household income, which does not appropriately represent the travel patterns of university students. Interestingly, because of the concentration of campus buildings, university cam- puses usually represent more livable environments that are higher density, mixed use, and alternative mode friendly, offering greater access to activities nearby. As such, exploration of students’ behavior can be instructive and reveal valuable information about associa- tions of their travel (e.g., trip generation and mode choices) with the built environment, and they can also be compared with the general population in the region. For that purpose, this paper attempts to understand the travel behavior of university students from a travel demand modeling per- spective. Given the somewhat unique life-cycle stage of university students, this paper considers the special nature of their personal characteristics, their lifestyle (both studying and working), and the spatial factors of where they live and study and work. The insights gained from this study will serve as the basis for trip generation in regional travel demand models, in which university-dominated traf- fic analysis zones are treated as special generators. Also, the paper will shed light on how a university campus environment, which has a mix of land uses (e.g., offices and classes, residential, and com- mercial), is alternative mode friendly, and higher density, associated with students’ driving and walking–bicycling behavior. RELEVANT LITERATURE University students’ travel behavior is rarely investigated in national surveys, including NHTS, and only a few documented surveys focusing on university students’ travel behavior exist in the peer- reviewed literature. Even less literature was found that discussed how to model the travel demand of university students. Rodríguez and Joo have used data from a University of North Carolina–Chapel Hill commuter survey of student and staff to illustrate relationships between mode choice and spatial characteristics (5). FTA-sponsored research on transit systems in colleges and provided a relatively good representation of transit practices at selected major universities (6). Kamruzzaman et al. used the trip diary method to investigate 2 days of out-of-home travel and activities by college students (7 ). They found that the number of “unique locations” visited by students was 3.59, and female students visited more unique locations than male What Can Be Learned from Analyzing University Student Travel Demand? Xin Wang, Asad J. Khattak, and Sanghoon Son To improve regional travel demand models, transportation engineers and planners desire appropriate representation of subpopulations. University students are a relatively neglected group of the population, often missed in regional behavioral surveys and not well represented in travel demand models. Many students attending a university reside, take classes, work, and perform other activities in the university environment, which is often mixed use, alternative mode friendly, higher density, and livable. The purpose of this paper is to understand the travel behavior of university students and to model associations with their attributes that include personal characteristics, residential location (residing on cam- pus or off campus), and academic status. The data used in this study are from a unique Internet-based survey (N = 1,468) of students at Old Dominion University in Virginia. This effort was conducted in 2010 and was part of the Virginia University Student Travel Survey (USTS) sup- plement. With USTS data combined with spatial data, rigorous statis- tical models of automobile and walk–bicycle trip rates are estimated to explore associated factors. Results showed that students living on cam- pus or near campus were significantly more likely to walk and bicycle and less likely to drive automobiles and indicated the value of living in a campus environment with greater accessibility to activities and a walk- and bicycle-friendly network. The behavioral models provide helpful information that can be used to represent better the behavior of univer- sity students in regional travel demand models and to improve strategic transportation planning. College and university campuses are experiencing rapid growth associated with an increase in population and the growing desire for education. Almost every urban campus in the United States faces serious effects from automobile traffic and parking shortages. At the same time, automobile use has major effects on the local and global environment (1). Travel demand management strategy is used by many universities to alleviate the problems of limited on-campus parking, parking restrictions from surrounding neighborhoods, and financial constraints (2–4). However, few studies have documented surveys of university students’ travel behavior. Although the travel behavior of the general population is typically measured through behavioral surveys, certain subgroups can be underrepresented. This underrepresentation may be the result of a host of reasons, for exam- ple, university students may live in dormitories and, because conven- tional telephone interview surveys [e.g., National Household Travel