Policymakers rely on transportation statistics, including data on personal travel behavior, to formulate strategic transportation policies and to improve the safety and efficiency of the U.S. transportation system. Data on personal travel trends are needed to examine the reliability, efficiency, capacity, and flexibility of the nation's transportation system to meet current and future demands; to assess the feasibility and efficiency of alternative congestion-alleviating technologies (e.g., high-speed rail, magnetically levitated trains, and intelligent vehicle and highway systems); to evaluate the merits of alternative transportation investment programs; and to assess the energy use and air quality impacts of various policies.
To address these data needs, the U.S. Department of Transportation (USDOT) initiated an effort to collect detailed data on personal travel. In 1969, the first Nationwide Personal Transportation Survey (NPTS) was conducted. The survey was repeated in 1977, 1983, 1990, and 1995. In 2001, an expanded survey included both daily and long-distance travel. In essence, the 2001 survey combined the NPTS and the 1995 American Travel Survey and was renamed the National Household Travel Survey (NHTS).
Three USDOT agencies sponsored the 2001 survey: the Federal Highway Administration, the Bureau of Transportation Statistics, and the National Highway Traffic Safety Administration. The primary objective of the survey was to collect trip-based data on the nature and characteristics of personal travel so that the relationships between the characteristics of personal travel and the demographics of the traveler could be established. Commercial travel was not part of the survey.
For the 2001 survey, the national sample consisted of 26,000 households. Four states and five planning areas purchased over 43,000 additional samples. These "add-ons" increased the number of sample households in these state/planning areas so that trip rates and travel statistics could be estimated more reliably at that geographic level.
Interviews were attempted with all members of sample households. Each was asked to provide detailed information on their daily and long-distance travel. They were also asked to provide information about their household, its members, and vehicles. Data about every one-way trip taken by each household member during a designated 24-hour period (the household's designated travel day) and data describing every roundtrip of 50 miles or more from home taken by each household member during a four-week period (the household's designated travel period) were collected.
This special issue begins with Erlbaum's comprehensive analysis of the quality of the 2001 NHTS data and comparisons between 2001 NHTS data and data from other sources (e.g., traffic count programs and administrative records). Although Erlbaum's analysis was based primarily on the New York add-on sample, it helps illuminate how 2001 NHTS data can be used and/or integrated with other data sources.
Understanding the interactions between land use and travel is critical to designing balanced transportation systems. With the wealth of information in the 2001 NHTS, Scuderi and Clifton used a Bayesian approach to improve, in relation to previous research, complex quantitative relationships between land use and transportation. Specifically, the authors apply Bayesian belief networks (BBNs) to analyze complex spatial systems (e.g., the urban environment). They demonstrate that this approach does not rely on ad hoc statistical models or assumptions. As such, there is no need to characterize variables as independent or dependent. Although limited results are presented, this paper identifies future opportunities where BBNs could provide insights to effectively address complex transportation issues.
Polzin and Chu analyze 2001 NHTS data and other data sources to take a closer look at the national trends in public transportation mode share. Their research shows that as the growth in overall national travel has slowed, transit use appears to have fluctuated. This was the case in terms of both absolute trips and transit's share of overall travel. The research also identifies the shortcomings and differences of the various data sources for determining transit use and mode share trends. The authors recommend synthesizing multiple measures of mode share to identify data needs and to provide a knowledge base to inform national policy deliberations.
Special attention was given in the 2001 survey to prompt the responders about their walk and bike trips, contributing to a 60% increase in reported walking trips from 1995 to 2001. Using 2001 NHTS data for the Baltimore metropolitan region, Targa and Clifton present an empirical analysis of the effects of several land-use, urban form, and neighborhood-level design attributes, as well as traveler attitudes/perceptions of the urban system, on the frequency of walking and the share of walking trips relative to total trips. Their results suggest that people who walk more frequently live in neighborhoods with higher densities, more diverse land-use mixes, better street connectivity, and better access to bus transit lines.
While the NHTS is an invaluable resource to estimate household travel statistics at the nationallevel, it is not advisable to use NHTS data to estimate travel statistics for geographic areas smaller than a Census Division. To meet the data needs for areas smaller than a Census Division, Mei et al. used a Bayesian updating approach to estimate travel statistics at the statewide level and for counties and rural areas.
At a time when budgets do not allow the collection of adequate sample sizes, Stopher and his colleagues demonstrate that Monte Carlo simulation with Bayesian updating could be a reasonable alternative to a full-scale household travel survey. Furthermore, the preliminary results obtained by the authors suggest that 2001 NHTS data could be suitable for Monte Carlo simulation of household tours.
The last paper in this Special Issue, by Sharp and Murakami, suggests some methodological considerations for future household surveys. In light of technology trends (e.g., cell phones and web utilities) and sociodemographic changes, the authors offer considerations to ensure data quality, increase response rates, and sustain survey efficiency. These considerations can assist the transportation community in designing and implementing future surveys in a more effective and efficient manner.
Space limitations in this Special Issue precluded comprehensive coverage of the 2001 NHTS. A few examples of core data we could not include are the amount of time Americans spend in their vehicles on a typical day, how Americans view the quality of our nation's transportation services, how these views have changed over time, and how and what different subpopulations require of the nation's transportation services. For example, Americans considered road rage as the most serious transportation problem in 2001, followed by distracted drivers and high gasoline prices (figure 1).
Americans felt that the quality of our nation's transportation services had not improved since the 1995 survey. For those quality aspects included in both the 1995 and 2001 surveys, pavement conditions and highway congestion were the top two concerns (figure 2). The growing discontent over highway congestion was partially substantiated by the increasing number of minutes spent driving, which rose to more than an hour a day for a typical driver (figure 3).
Readers can explore the potential of the 2001 NHTS by visiting its website (http://nhts.ornl.gov) and by sharing research findings with others. Conference papers are available at http://www.trb.org/Conferences/NHTS/Program.pdf.
The articles in this Special Issue were among the many presentations made at the National Household Travel Survey Conference: Understanding Our Nation's Travel,held at the National Academy of Sciences in Washington, DC, November 12, 2004. This conference served as a forum for the various data users of the national survey to discuss analysis and findings from the recently released 2001 NHTS. We wish to thank the Transportation Research Board for organizing, and the Bureau of Transportation Statistics and other agencies for providing funding and support to, the conference.
This issue would not have been possible without the dedication, expertise, and objectivity of the numerous referees. Their insight and careful reviews helped advance the quality of every article. We also wish to thank the journal's Editor-in-Chief, Peg Young, and the publishing staff: Marsha Fenn, Alpha Glass, and Dorinda Edmondson. Their advice, commitment, and professionalism made it much easier to undertake the editing and compilation of this issue.
Center for Transportation Analysis
Oak Ridge National Laboratory
Lee Giesbrecht and Joy Sharp
Bureau of Transportation Statistics
Research and Innovative Technology Administration
U.S. Department of Transportation