Traveler Response to Transportation System Changes Handbook, Third Edition: Chapter 1, Introduction (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

1-1 1 – Introduction TCRP Report 95: Traveler Response to Transportation System Changes, is the product of combined TCRP Projects B-12, B-12A, and B-12B, “Updating the ‘Traveler Response to Transportation System Changes’ Handbook (DOT-FH-11-9579).” The overall TCRP project objective has been to provide an up-to-date and expanded sourcebook on how travel demand is affected by changes in the transportation system, program and promotional actions, and built environment options. This Chapter 1, “Introduction,” contains a four-level “Table of Contents” for the 16 published TCRP Report 95 chapter/volumes, describes the Traveler Response Handbook and its development, offers guidance to the prospective Handbook user, and includes two Handbook appendices plus known errata. It is the only source of four-level table of contents information for the published chapters, and contains important context and technical information supportive of each of the topic-specific chapters in the series. Changes/differences in urban transportation system provisions, operations and policy, and the built environment: How do they affect travel demand? The answers are vital in planning and evaluating transportation proposals, and for making effective design, operational, and policy changes to urban transportation systems. Projections of the trips by persons, mode use, and other manifestations of individual travel choices are a fundamental input to most transportation planning estimates includ- ing those of traffic and passenger volumes, congestion, revenues, costs, feasibility, travel benefits and disbenefits, economic impacts, energy use, and environmental effects. The Traveler Response to Transportation System Changes Handbook has served to assist—since 1977—as a travel demand refer- ence for transportation professionals, providing accessible, interpretative documentation of actual results obtained from various types of urban transportation actions and options. This TCRP Report 95 is the third edition. Handbook Context and PreParation In this first of the two major “Introduction” sections, the Traveler Response Handbook is placed in context with its origins, the current update, and companion reference documents. Its scope and key research steps are outlined. The second major section, “Use of the Handbook,” provides infor- mation in support of informed and effective application of the work. Genesis and Coverage of the Handbook Earlier Editions The U.S. Department of Transportation published the first and second editions of the Traveler Response to Transportation System Changes Handbook (Pratt, Pedersen, and Mather, 1977, Pratt and Copple, 1981). At the time of the first edition, energy and environmental concerns and a shift in emphasis away from capital intensive projects had caused serious consideration of transportation alternatives to constantly increasing low occupancy auto use. Regulations had been issued requir- ing Transportation System Management (TSM) as part of the urban planning process. Accordingly, low capital urban transportation improvements, both TSM and transit, were the primary system focus of the first edition.

1-2 By the time of the second edition, shrinking transportation revenues, rising construction costs, and energy scarcity had made more effective use of existing transportation systems essential. There was expanded emphasis on traffic operations improvements, ridesharing, priority for high occupancy vehicles, transit service enhancements, variable work hours, and other TSM actions. More experience had been gained in application of these types of actions. However, while the second edition reflected that advancement in practical knowledge, and substantially expanded on each topic retained from the first edition, only nine categories of system change were covered—and that was two decades prior to the new millennium. Current Update Further substantial innovations and advancements have occurred in transportation services provi- sion and management since publication of the second edition Traveler Response Handbook, given impetus by the various Transportation Efficiency/Equity Acts of Congress. Electronic media have expanded fare, fee, and toll payment options; high occupancy toll (HOT) lanes have introduced value pricing to High Occupancy Vehicle (HOV) facilities; federal interest in non-motorized trans- portation (NMT) provisions has vastly increased; and new emphasis has been placed on land use and transportation interrelationships. More has been learned from approaches new or unseasoned a quarter century ago. The concept of Travel Demand Management (TDM) has taken form and been widely applied. Connecting people with their daily activities, meeting needs for mobility, and addressing congestion in a context of continuing resource constraints, quality of life issues, and environmental concerns, calls for timely and practical information on how travel demand is shaped and altered by changes in our urban transportation systems and urban fabric. In response, the TCRP has undertaken this major update and expansion of the Traveler Response Handbook. The topic area coverage of the updated Traveler Response to Transportation System Changes Handbook has been shaped by project research on practitioner needs, including Handbook user surveys, ongo- ing outreach to the transportation planning profession, and project panel work sessions. The focus is on transportation system changes and alternatives, or land use and site design differences, that directly affect the traveler. With individual-volume publication of each chapter extending from 2003 through 2012, plus the three future volumes marked with asterisks, this third edition contains the following topic area chapters/volumes: • HOV Facilities • Transit Information and Promotion • Park-and-Ride/Pool • Transit Pricing and Fares • Busways, BRT and Express Bus* • Parking Pricing and Fees • Vanpools and Buspools • Road Value Pricing • Demand Responsive Transit/ADA • Land Use and Site Design • Light Rail Transit* • Pedestrian and Bicycle Facilities • Commuter Rail* • Transit Oriented Design • Transit Scheduling and Frequency • Parking Management and Supply • Bus Routing and Coverage • Employer and Institutional TDM Strategies Companion Handbooks The Traveler Response to Transportation System Changes Handbook has, from its inception, been part of an overall transportation systems and travel demand research effort by the U.S. Department of Transportation, and now the Transit Cooperative Research Program. Related handbooks in this effort

1-3 are Characteristics of Urban Transportation Systems (CUTS), currently in its seventh revision by the Federal Transit Administration (Cambridge Systematics, 1992), and Characteristics of Urban Travel Demand (CUTD), available in its third version as TCRP Report 73 (Reno, Kuzmyak, and Douglas, 2002). All three handbooks have received wide distribution and have become standard references in the transportation field. The three handbooks were joined in 1999 by the Transit Capacity and Quality of Service Manual, now in its second edition as TCRP Report 100 (Kittelson and Associates et al., 2003). This manual was designed to provide the public transit equivalent of the Highway Capacity Manual while also address- ing transit quality of service concepts and measurement. The fifth edition of the Highway Capacity Manual 2010 (HCM 2010) now provides multimodal incorporation of transit, pedestrian, and bicycle facility capacity and level of service (LOS) evaluation including assessment of off-road pedestrian and bicycle facilities and on-road bicycle provisions (Highway Capacity Manual, 2010). Publication timeline of this Handbook The magnitude of this third edition Traveler Response to Transportation System Changes Handbook project led to advance and staged publication in an effort to provide access to available findings pending completion of the full 19-chapter, 18-topic Handbook. The process has been as follows: • Seven topic area chapters and an initial “Introduction” were electronically published in March 2000 as an Interim Handbook—TCRP Web Document 12 (Pratt et al., 2000b). That Interim Handbook is now completely superseded by TCRP Report 95. • Print publication of the third edition Traveler Response to Transportation System Changes Handbook began in mid-2003, chapter by chapter, as TCRP Report 95. Each chapter is a separate volume. As each chapter has been printed, the electronic equivalent has been linked to from http://apps. trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=1034. Access is also available through http://www.trb.org/main/blurbs/162432.aspx. • An Interim Introduction (with Appendices A and B) was published as TCRP Research Results Digest 61 in 2003 to replace Chapter 1, “Introduction” and “Appendix A” of the Interim Handbook and accompany the TCRP Report 95 series (Pratt, 2003). That Interim Introduction is now superseded by this TCRP Report 95, Chapter 1, “Introduction.” • All TCRP Report 95 chapter/volumes are, with this 2012 publication of Chapter 1, “Introduction,” printed and electronically posted except for future Chapters 4, 7, and 8. This makes a total of 16 chapter/volumes currently available to the profession, including 15 topic area chapters. Table 1-1 lists the 18 TCRP Report 95 topic area chapters and “Introduction” in outline order, and summarizes publication and source data cutoff information. The later source data cutoff dates for the most recently published chapter/volumes have allowed some “cross-chapter” updating where there are areas of common interest. Users of Chapter 11, “Transit Information and Promotion,” will find vastly expanded information on “individual- ized marketing” programs and results (including U.S. and Australian transit ridership effects) in Chapter 16, “Pedestrian and Bicycle Facilities.” Users of Chapter 14, “Road Value Pricing,” should turn to Chapter 2, “HOV Facilities,” for additional experience with HOT lanes. And finally, users of Chapter 15, “Land Use and Site Design,” should see the “Pedestrian/Bicycle Friendly Neighborhoods” subsection in Chapter 16 for an update on elasticities of travel with respect to urban form parameters.

1-4 Scope and development of the Handbook This updated Traveler Response to Transportation System Changes Handbook responds to the need for up-to-date and comprehensive information on travel demand effects of alternative urban trans- portation policies, operating approaches and systems, and built environment options, by building upon, expanding, and selectively replacing the earlier editions to provide a contemporary assess- ment of the experience and insights gained from the application and analysis of various system changes and alternatives. The focus is on aiding transportation, transit, and land use planners—and public health practitioners in the case of Chapter 16, “Pedestrian and Bicycle Facilities”—in their conduct of travel demand and related analyses, and to inform elected officials, administrators, operators, designers, and the general public as well. U.S. DOT Publication TCRP Report 95 General Sections and Topic Area Chapters (TCRP Report 95 Nomenclature) First Edition Second Edition Source Data Cutoff Date Publication Date Ch. 1 – Introduction (with Appendices A, B) 1977 1981 1999-2012 a 2000/03/12 b Multimodal/Intermodal Facilities Ch. 2 – HOV Facilities 1977 1981 1999-2005 a 2000/2006 c Ch. 3 – Park-and-Ride/Pool — 1981 2003 d 2004 Transit Facilities and Services Ch. 4 – Busways, BRT and Express Bus 1977 e 1981 e future future Ch. 5 – Vanpools and Buspools 1977 1981 1999-2004 a 2000/2005 c Ch. 6 – Demand Responsive/ADA — — 1999-2003 a 2000/2004 c Ch. 7 – Light Rail Transit — — future future Ch. 8 – Commuter Rail — — future future Public Transit Operations Ch. 9 – Transit Scheduling and Frequency 1977 1981 1999-2004 a 2000/2004 c Ch. 10 – Bus Routing and Coverage 1977 1981 1999-2004 a 2000/2005 c Ch. 11 – Transit Information and Promotion 1977 1981 2002 2003 Transportation Pricing Ch. 12 – Transit Pricing and Fares 1977 1981 1999-2004 a 2000/2004 c Ch. 13 - Parking Pricing and Fees 1977 e — 1999-2003 a 2000/2005 c Ch. 14 – Road Value Pricing 1977 e — 2002-2003 a 2003 Land Use and Non-Motorized Travel Ch. 15 – Land Use and Site Design — — 2001-2002 a 2003 Ch. 16 - Pedestrian and Bicycle Facilities — — 2007-2011 f 2012 Ch. 17 – Transit Oriented Development — — 2004-2006 a 2007 Transportation Demand Management Ch. 18 – Parking Management and Supply — — 2000-2002 a 2003 Ch. 19 – Employer and Institutional TDM Strategies 1977 e 1981 e 2007-2009 a 2010 Notes: a Global cutoff was first year listed, but with selected information up into second year listed. b Published in TCRP Web Document 12, Interim Handbook, in 2000, without Appendix B. Updated in 2003 as an Interim Introduction and published as TCRP RRD 61. Published in this current version, as part of the TCRP Report 95 series, in 2012. c Published in TCRP Web Document 12, Interim Handbook, in 2000. Published as part of the TCRP Report 95 series on the second date shown. d The source data cutoff date for certain components of this chapter was 1999. e The edition in question addressed only certain aspects of later edition (or planned) topical coverage. f Global cutoff was first year listed, but with extensive information up into second year listed. Table 1-1 Handbook Outline Showing Publication and Source Data Cutoff Dates

1-5 Handbook Goal and Scope The goal of the Traveler Response Handbook series is to equip members of the transportation profession and others with a comprehensive, readily accessible, interpretive documentation of results and experience obtained across the United States and elsewhere from different types of transportation system changes and policy actions, and also alternative land use and site devel- opment design approaches. While the focus is on contemporary observations and assessments of traveler responses as expressed in travel demand changes, the presentation is seasoned with earlier experiences and findings to identify trends or stability, and to fill research gaps that would otherwise exist. Comprehensive referencing of additional study materials is provided to facilitate and encourage in-depth exploration of topics of interest. The Handbook is not intended for use as a substitute for regional or project-specific travel demand evaluations and model applications, or other independent surveys and analyses. They perform a vital function in that they address location-specific characteristics of travel that make blind transfer of results from other areas a risky proposition. The findings in the Traveler Response Handbook are intended to aid—as a general guide—in prelimi- nary screening activities and quick turn-around assessments. The presentation is also designed to serve as a complement to location-specific analyses, providing a basis for comparison and helping to fill gaps in model-derived demand estimates. The Handbook is further intended to enlighten transportation systems and operations specialists concerning potential or likely travel demand implications of their individual and collective policy and design decisions. It is with improved information from a variety of sources and techniques that transportation and other professionals can most effectively formulate responses to transportation needs and opportunities, and reliably predict the efficacy of their plans. Research Approach The Traveler Response Handbook findings derive primarily from reported results and analyses of real- world transportation system and policy applications and trials. Experimental or quasi-experimental empirical data have been the information source of choice, with rigorous revealed-preference (actual response) analyses accorded somewhat more weight than stated (perceived) preference experiments. Other empirical data derivations and simple accounts of outcomes have been employed as neces- sary. Forecasts and other estimates derived from travel demand model applications and similar techniques have been used, but on a very selective basis; mostly for augmenting the empirical data where gaps exist, and for providing additional insights and context. The TCRP Project B-12/B-12A handbook updating and expansion was not intended to require original research. The process has been one of synthesizing, distilling, and interpreting reports, papers, and research by others. Nevertheless, to fill major gaps, the research team has performed original assessments of key undertakings and situations, most often relying on cooperative oper- ating and planning agency staff to assemble significant data and analyses. In addition, the TCRP Project B-12B work effort involves more formalized and extensive original assessments, along the same lines, addressing Transit Oriented Development (TOD). Preparatory Tasks. The assembly and review task which began the development of this third edi- tion Handbook started with both a formal WinSPIRS based literature search and the foundational stage of extensive networking to locate papers, articles, reports, manuals, and syntheses with information on any aspect of traveler response to urban transportation system or policy changes.

1-6 The bibliographic information thus obtained, mostly from the 1981 through late 1997 period, was categorized, screened, and entered into a database. This initial phase of the literature search served to ensure that important materials only available in print, as contrasted to electronic postings, were not overlooked. The second edition had covered up through 1980. The second and longer-running phase of source material gathering continued throughout the topic development process, as the research team remained alert to new sources, contacted a num- ber of organizations and individuals known to be active in research areas of concern, and also per- formed various supplementary literature searches. Particular use was made of news sources such as the TRB Transportation Research E-Newsletter, rapidly developing electronic search capabili- ties, and the assistance of several pertinent TRB committees. A number of key documents became available during this extended period, especially in the case of highly active areas of research and application. In addition, as noted above, the research team did undertake certain of its own original assessments to supplement available literature and reportings. Table 1-1 indicates how far into the 1998–2012 period these information gathering activities continued for each individual chapter. References were selected using criteria such as the timeliness or enduring relevancy of the studies reported on, the clarity with which traveler response was measured, the apparent completeness of the work, and the extent of other literature on the same topic or event. In order to expand the base of evidence and fill research gaps, the sources employed were not limited solely to peer- reviewed papers and documents, but also included so-called “gray literature.” Government reports and other publications deemed suitable were consulted, and a variety of other sources were used when necessary and where context supported a reasonable presumption of sufficient reliability. Similar criteria were used to choose between retaining and deleting individual references and infor- mation available from the earlier Handbook editions. Old information found redundant or made suspect by the passage of time was rejected. The study team leaned toward retaining older high quality data, however, where it was still some of the best available on either presently relevant or recurring issues. This choice was in the interests of “[not losing] the message before gain[ing] the knowledge,” to rephrase the title of a Transportation Research Board Annual Meeting paper on the loss of older documentation (Rogers, 1997). Assembly and review activities were paralleled by tasks designed to solicit advice on content and presentation; to define, detail, and outline the structure of the updated Handbook; and to develop a coordinated and comprehensive approach for moving the finished product into practice. The user outreach employed a mail-out survey as the initial vehicle for obtaining the desired information. As the project progressed, volunteer review groups were formed for each topic area chapter to augment the work of the project panel in outline review, source data identification, and chapter draft review. A modified approach was used for Chapters 16, 17, and 19, wherein communication and consulta- tion with the TRB Pedestrian Committee, Bicycle Transportation Committee, Transportation and Land Development Committee, Transportation Demand Management Committee, and interested committee members, substituted for review group establishment. Preparation of first-cut literature summaries—following the literature search and related tasks— supported topic area synthesis and provided the starting point for case studies. This process was structured to define the nature of system change reported and the degree of influence of concur- rent changes, actions, and events; establish the methodology and likely reliability of the reported data sources and collection techniques; examine the nature and logic of analysis techniques used

1-7 to measure or model traveler response; and assess conclusions, degree of reliability, and trans- ferability to other situations and locations. The process was adapted to the situation and source material, and ranged from highly structured to less formal, but always guided the approach to findings derivation. Classification System Development. The organizational structure and classification system for this updated Handbook not only provides a Handbook framework, but also served initially as a starting point for the topic area selection process. This process went through two primary rounds, setting priorities for the Interim Handbook and for the full TCRP Report 95 coverage. The most recent change was the addition of the Transit Oriented Development topic, introduced under TCRP Project B-12B, and entered into the Handbook outline as redesignated Chapter 17. (See Table 1-1 and also see Footnote 2 for the resulting chapter renumbering.) The transportation system change classification structure as outlined in the first iteration encom- passed 27 system change topics, each assigned to one of seven broad system change categories. In the course of the research, two of the broad categories were determined to be poorly suited for inclusion. They covered three Transportation Systems Management topics and two general purpose highway capacity topics. One of the broad categories has been split, giving the six system change categories listed in Table 1-1. Of the initial list of 27 system change topics, after subdivisions, recom- binations, eliminations, and additions, 18 remain. These form the topic area chapter/volumes of the third edition Handbook, listed earlier in Table 1-1 along with availability information. The earlier editions of the Handbook were organized to match the classification system used for the TSM actions published by the U.S. Department of Transportation. These actions were part of the regulations governing the urban transportation planning process at the time. With the pas- sage of the Intermodal Surface Transportation Efficiency Act (ISTEA) of 1991, the subsequent promulgation of regulations by federal agencies, and the reauthorizations since, the previous clas- sification system is no longer relevant. Accordingly, a new basis for a classification system was developed. A number of topical classifi- cations used in planning efforts current at the outset of this update were reviewed as part of this process. The proposed Congestion Management System strategies in the December 1, 1993, Interim Final Rule published by the Federal Highway Administration (FHWA) and the Federal Transit Administration (FTA) were considered an important source. That listing of strategies matches those outlined in the TCRP Project B-12 Research Problem Statement and the topics included in the survey of practitioners. The Congestion Management System strategies also include highway topics not covered by the TCRP Project B-12 Research Problem Statement. One limitation of the Congestion Management System strategies is that the organization of topics is somewhat awkward from the perspective of the Handbook. Two other sources were consulted for possible classification systems. These were the workbook used for the 1994 FHWA training course on Congestion Management for Technical Staff and the 1995 FHWA report, Guidebook for Congestion Management Systems: A Generic Process. Although the topic organization in these two sources dif- fered, they provided additional guidance. All of these sources, along with the initial problem statement, the expertise of the team members, and the approval of the project panel, were used to develop the organizational structure for the updated Handbook. The result is that illustrated in Table 1-1. Handbook Preparation. The heart of each topic area chapter is its comprehensive digest of travel behavior findings, backed up with case studies of selected transportation system changes, related

1-8 actions, or urban design options. As fully outlined in the next section, this digest consists of three main parts—an exposition of the traveler responses to the system changes in question, a discourse on the underlying travel behavior factors, and an overview of related information. Topical digest preparation involved comparing, distilling, combining, supplementing, and structuring the find- ings from the literature to derive relationships and concise presentations of lessons learned on the travel behavior and related effects of the transportation system changes or built environment options addressed. Empirical findings were played against modeled relationships in an effort to broaden, strengthen, and test each finding. Judgments were made as to the degree of caution that should be used in applying the various research results. The lessons thus derived have been synthesized in the form of generalized conclusions and guid- ance, including quantitative data and estimating parameters developed in accordance with con- sistency and commonality guidelines. Wherever possible the nature and sensitivity of response to each principal variable or configuration of systems change has been indicated. A comprehensive digest has thus been prepared of the literature-based travel behavior findings for each of the trans- portation system change and built environment alternatives topic areas chosen for coverage in the chapter/volumes of this TCRP update of the Traveler Response Handbook. While each topic area chapter serves to distill and present the collective observations of the relevant literature reviewed, the interpretations offered are the responsibility of the Handbook authors. Assessments are provided to aid the user in judging the degree of confidence with which the find- ings presented for each transportation change can be applied to project likely travel demand results. Handbook users should note that the findings presented are intended only as generalized guide- lines applicable under average conditions. In practical application, each individual project must be analyzed in the context of the specific site and conditions involved. The case studies selected to complement each topic area chapter’s digest of findings have been prepared according to a common approach. Each case study is designed to provide condensed case-specific traveler response and related information within the context of the system change, program action, or land use and site design option involved and the data collection and evaluation procedures originally employed. The standard approach has been to summarize the “Situation,” the observed “Actions,” the “Analysis,” the “Results,” and “More . . .” relevant insights. Case study intent has been to provide condensation while preserving the overall background, findings, and interpretations of the original sources without alteration of meaning. In some cases the Handbook authors have added calculations or interpretations, or have even carried out the basic assessment in conjunction with contributing agencies, and these instances are so noted. USe of tHe Handbook This third edition Traveler Response to Transportation System Changes Handbook consists of these Chapter 1 introductory materials and the 15 published topic area chapter/volumes, plus the three future chapters, along with their digests of traveler response findings including supportive infor- mation and interpretation. Each topic area chapter also includes case studies and a bibliography consisting of the references utilized as sources. Each topic area chapter begins with an overview and summary section that includes a descrip- tion of the types of system and policy change covered. This section also includes an “Analytical Considerations” discussion, addressing findings-reliability concerns, and closes with the chapter’s

1-9 “Traveler Response Summary.”1 The main digest which follows the “Overview and Summary” is provided in three major sections: • “Response to [the System Change],” covering traveler responses to the various specific sub- types of system change or action encompassed. • “Underling Traveler Response Factors,” addressing the travel demand mechanisms involved. • “Related Information and Impacts,” providing cross-cutting summaries of selected user, travel demand, and impact characteristics, and also overviews of associated subtopics. An identification of additional resources, the case studies, and references follow, and conclude each topic area chapter. Together, these components offer: • A consolidated, interpretive source of information on how travel demand and use of transpor- tation facilities and services are affected by the 18 different categories of transportation system change, program actions, and built environment alternatives. • Encapsulated examples and case studies. • Sources of additional information on the same and closely related subjects. Appendix A of this Chapter 1, “Introduction,” provides additional information on the deriva- tion and application of elasticity measures, and Appendix B of Chapter 1 gives historical cost of living statistics and conversions useful for interpreting transportation system changes involving user costs. Known substantive errata, pertaining entirely to Chapter 15, are addressed following Appendix B. Handbook application Applying the material in any handbook that attempts to provide useful generalizations and exam- ples must be done with care and this Handbook is no exception. The subject at hand is complicated by the many confounding factors that may influence traveler response and by the diversity of urban environments involved. Some of these considerations are expanded upon in the final four subsections of this “Introduction” under the headings “Degree-of-Confidence Issues,” “Impact Assessment Considerations,” “Demographic Considerations,” and “Concept of Elasticity.” Concerns of special importance to individual topic areas are presented within the “Analytical Considerations” subsection of each respective chapter’s “Overview and Summary.” Issues specific to individual findings are brought up as part of the presentation of those findings. Transportation planners and decisionmakers can use the information in this Traveler Response Handbook as a starting point to consider and evaluate transportation alternatives. Any specific situation must, however, in the end be examined in terms of the particular urban form, population, travel patterns, and transportation systems involved. No transportation system change can prop- erly be considered in isolation from these measures and physical expressions of local conditions. 1 Chapter 16, “Pedestrian and Bicycle Facilities,” also contains an “Adult and Child Public Health Relationships Summary,” placed at the end of the “Public Health Issues and Relationships” subsection of that chapter’s “Related Information and Impacts” section.

1-10 The Handbook is structured to point the reader progressively to more detailed information internal to the document and then on to other studies, investigations, and texts. Given a specific transporta- tion (or urban development) and travel demand question, a recommended progression within and beyond the Handbook would be: • Select the most pertinent topic area chapter, a choice process that may be aided by the compre- hensive four-level Table of Contents located at the beginning of this Chapter 1. • Review the “Overview and Summary” section to understand context and data limitations, and to obtain a preview from the “Traveler Response Summary.” • Upon finding relevant topics in the “Table of Contents,” or potentially useful information in the “Traveler Response Summary,” proceed to the “Response to [the System Change]” section of the chapter for the full traveler response presentation for each sub-type of system change, with citations. • Refer to any of the accompanying “Case Studies” which apply, for additional case example detail. Case studies are near the end of the chapter, ahead of “References.” • Consult the “Underlying Traveler Response Factors” section for interpretations useful in understanding the travel demand mechanisms and individual factors at play. • See subtopics of interest within the “Related Information and Impacts” section. This section includes compilations, relevant to the chapter topic, ranging from user characteristics and trip purpose data to prior travel mode information and system usage characteristics, and also overviews of related experience and research findings. • Check relevant cross-referenced material in other chapters, and also those updates provided in other chapters as described in connection with Table 1-1 (last paragraph of the subsection “Publication Timeline of This Handbook”). • Go outward from the Handbook—from either the main text, using the source material citations and the “References” entries linked to them, or from the “Sources” listing in each case study or the “Additional Resources” listing—to applicable papers, studies, and texts. Only the source literature can provide the breadth and detail of the original work. Finally, after making use of the material in this Handbook and other literature sources, proposals at hand that pass initial screening must ultimately be studied within their own particular context and unique environment. The degree of appropriate study, depending upon the action contem- plated, may range from design of a simple and low-cost pilot project with an accounting of the risks involved, to more detailed project designs employing user or prospective-user surveys and related analyses, to full-scale major capital investment studies inclusive of travel demand model applica- tions for forecasting site- and alternative-specific traveler responses and related outcomes. Handbook organization The remainder of this introductory chapter is devoted to additional instructive material offered to enhance use of the Traveler Response to Transportation System Changes Handbook. All subsequent Handbook chapters are devoted to the individual topic areas, each addressing one general type of transportation system change or related action except for Chapter 15, which covers “Land Use and

1-11 2 Chapter 18, “Parking Management and Supply,” was referred to in the Interim Handbook, Web Document 12, as Chapter 17. Similarly, Chapter 19, “Employer and Institutional TDM Strategies,” was referred to as Chapter 18. Site Design.” The topic area chapters are arranged according to the transportation system change classification structure of the full third edition Handbook, encompassing 18 system change topics, as described earlier. Each topic area chapter is assigned, as illustrated in Table 1-1, to one of six broad system change categories. These categories constitute general sections, which simply serve to order the system change topics. Chapter and page numbering is in conformance with the planned 18-topic, 19-chapter TCRP Report 95 third edition.2 Referrals in chapter text to one of the three “future” chapters may be encountered. These are further explanatory cross-references and do not detract from the use of the published chapter. Topic Area Chapter Format Table 1-2 illustrates the general format of each topic area chapter presentation, utilizing the “Vanpools and Buspools” chapter as an example. The “Overview and Summary” section that each topic area chapter starts with begins with a road- map to the chapter itself. Next within the overview is a statement of the generally accepted objec- tives for undertaking the types of system changes or equivalent covered by the topic. This is immediately followed by a listing of the implemented or implementable types of change or option included. Next is a discussion of analytical considerations, designed to indicate the limitations encountered in the research and the corresponding level of confidence that can be placed in the conclusions. With that background, the last item in the overview is a “Traveler Response Summary” section, highlighting the traveler response findings for the topic. (As previously detailed in Footnote 1, Chapter 16 also includes an “Adult and Child Public Health Relationships Summary,” located within that chapter’s “Related Information and Impacts” section.) The recommended approach to using either the “Traveler Response Summary,” or the material which follows, is to do so only after first reading the initial three sections of the “Overview and Summary” for background. Moreover, any consequential use of findings previewed in the “Traveler Response Summary” should be accompanied by consideration of caveats set forth only in the more detailed presentations which follow. The “Overview and Summary” of each topic area chapter is followed by the full Handbook survey of observed traveler responses for each type of system change (or program action or built envi- ronment option) addressed in the chapter. This “Response to [the System Change]” section of the chapter is accompanied by an “Underlying Traveler Response Factors” section which examines the role of both underlying travel behavior mechanisms and external factors in producing the traveler responses. Note that this sort of information is useful not only for a better understanding in general, but also for practical application in achieving an effective transportation service, facility, and urban design. Next is a “Related Information and Impacts” section which touches upon other subject areas pertinent to the particular system change topic, including environmental and cost considerations. It is important to be aware, as previously noted, that this section includes various cross-cutting compilations relevant to the chapter topic.

1-12 The “Response to [the System Change],” “Underlying Traveler Response Factors,” and “Related Information and Impacts” sections comprise what has been referred to in preceding discussion as the “digest.” These three core sections are followed by a brief “Additional Resources” section highlighting compendia, manuals, or other documents of likely special interest to a reader delving further into the topic area in question. Last in each topic area chapter, except for references, are the case studies. They provide condensa- tions of papers, reports, and other information on system change applications selected for relevancy to the topic area and usefulness in illustrating and expanding upon the travel demand and related impacts reported in the preceding topic area presentation. They focus on case-specific traveler response and related information, necessarily within the context of the particular system studied. In structuring the case studies, the intent has been to include a description of the case study context, Format (Outline/Major Headings) Example (Vanpools and Buspools) [General Category] Transit Facilities and Services [Transportation System Change Topic] Vanpools and Buspools Overview and Summary A roadmap to the chapter, plus the following: Objectives of [the System Change] Vanpool/Buspool program focus, objectives Types of Programs A listing and definition of: Vanpools—Employer-Sponsored Programs Vanpools—Third-Party Programs Vanpools—Owner Operator Buspool Programs Analytical Considerations Limitations, caveats, data interpretation alerts Traveler Response Summary Overview of traveler response to changes/actions Response to [the System Change] A survey of the traveler responses to: Employer-Sponsored Vanpool Programs Third-Party Vanpool Programs Buspools (Subscription Bus) Underlying Traveler Response Factors Exploration of traveler response mechanisms: [Travel Time Components] and Trip Distance Access Considerations Work Scheduling Implications Incentives and User Costs Preferences, Privileges and Intangibles Related Information and Impacts A survey of related special and standard topics: Extent of Vanpooling and Buspooling Demographic Characteristics of Riders Sources of New Ridership and Vanpooler Turnover Indicators of Market Potential Impacts on VMT, Energy, and Environment Revenue/Cost Considerations Additional Resources Four sources of special interest highlighted Case Studies Four case studies provided References 80-odd references listed Table 1-2 Topic Area Chapter Format Example (Vanpools and Buspools)

1-13 the system change action involved, some indication of the data collection and analysis methodolo- gies originally employed, and a highlighting of the principal findings along with related useful information. Bibliographic Format In this third edition of the Traveler Response to Transportation System Changes Handbook—issued in both print and electronic versions, with each in a separate chapter/volume—the references pertaining to each topic area are kept in a list at the end of the topic area chapter in question. Duplication results, as compared to having a consolidated reference list, but use of one chapter at a time is facilitated. Single-chapter use should not be overdone, however, because of topic and concept overlap among chapters. The citations in the text that provide the links to the list of references employ the so-called humani- ties style, giving author and date. This approach serves to alert the reader to the age of the source, and also allows the references to be alphabetically listed in bibliographic fashion. Certain of the citation protocols derive from the non-academic origins of this Handbook, and are retained in this TCRP Report 95 edition as a practical matter. In the text, citations are generally located at the end of groups of sentences or paragraphs based on the cited source material, but ahead of bulleted lists drawn from a single source. Where multiple paragraphs have been extracted from the same source, citations are repeated every so often, particularly where the continuity is not fairly obvious. Citations are not provided in the condensed “Traveler Response Summary” presentations which conclude the “Overview and Summary” section at the front of each topic area chapter. The Handbook user must look further, into the more detailed presentations which follow, for the appli- cable citations. Findings drawn from source literature and reports have been added to with original informa- tion from individual contributors. Contributors of original information are cited and listed under “References,” generally with an indication of their affiliation. Source documents and contributors are further supplemented with the practical and research experience of the Handbook authors along with their general knowledge of other studies and writings relevant to the topic. With the one primary exception of the “Traveler Response Summary” presentations, lack of a citation indicates that the information and observations offered at that juncture come directly from the professional experience and training of the Handbook authors. degree-of-Confidence issues Findings presented in this Traveler Response Handbook derive primarily from quasi-experimental and other field observations, supplemented with interpretations from additional sources such as travel demand model research, other cross-sectional studies, and stated preference experiments. Traveler response is expressed using such measures as elasticities, before-and-after market shares, and percentage changes or differences, with indicators of scale and guidance ranging from exam- ples of volumes and passenger demand to simple feasibility indicators. The confidence which can be placed in the generalizations drawn in each topic area digest con- cerning impacts of the transportation system changes varies among the types of change involved. The appropriate degree of confidence necessarily depends on the number of documented observa- tions of a certain type of change or action, the design and statistical soundness of the associated

1-14 analyses, confounding factors which may have affected the reported results, the extent to which impacts have been successfully modeled, and the consistency with which findings were obtained and reported. Some transportation system changes examined here have had only limited application, or have been very infrequently subjected to systematic analysis, so that it is difficult to generalize to uni- versal experience. In other instances, confounding factors or unique situations have influenced the results. Even where it has been possible to draw upon and compare numerous study findings, the validity of the inferences made is still dependent upon the quality of the data and analysis methods originally employed in deriving the reported conclusions. As already indicated, an “Analytical Considerations” section has been included in each topic area chapter to provide a discussion of factors influencing the degree of confidence with which the traveler response findings and related conclusions for that Handbook topic area can be used. Circumstances range over the entire spectrum from instances where strong empirical evidence and theoretical basis exist to support the validity and widespread applicability of the conclusions derived, to situations where little hard data exists and conclusions offered are drawn primarily by inference or from very limited applications or theoretical studies. In general, findings and conclusions provided with respect to impacts on vehicle miles of travel (VMT), energy consumption, and emissions are relatively more dependent on estimations and modeling applications than the associated traveler response conclusions. In this Traveler Response Handbook update, reduced emphasis has been placed on presenting energy conservation and air quality data, in light of the more extensive separate resources now available on the subjects of energy consumption, pollutant emissions, and sustainability. The user of this Handbook will find that absolute-value data is often presented without round- ing. This has been done as a matter of convenience, and should not be taken to always imply high levels of confidence. There is little in this Handbook with real meaning (for travel demand assessment purposes) beyond two significant figures. Reliability of elasticities is closer to one significant figure (if that) however useful elasticities may be for quick estimates. Percentages have generally been rounded. When percentages and other rounded numerical val- ues do not total exactly, for example, not adding to exactly 100 percent when representing the whole, it may be assumed that such minor discrepancies are attributable to rounding. Significant discrepancies carried over from source documents are marked with a “(sic)” notation. impact assessment Considerations The available observations of traveler response to transportation system changes are provided by assessments that run the full gamut from reports that are borderline anecdotal, to comprehensive studies carried out employing a pre-prepared and vetted study design. Full-scale travel behavior research evaluations are unfortunately limited in number for most system change sub-types, and scarcer yet are those that qualify as either true experiments or quasi-experimental before-and-after research. Moreover, even highly structured assessments do not always follow good evaluation principles (Pratt et al., 2000a, Higgins and Johnson, 1999, Committee on Physical Activity, Health, Transportation, and Land Use, 2005, Ogilvie et al., 2007, Parker et al., 2007). The result is a virtual minefield that must be negotiated carefully by the user of available observa- tions and studies, and indeed the user of compendia such as this one which must rely on available

1-15 sources. If only the best and most sound of evaluations were to be reported, the data points would be too few. The choice has been made in developing this Handbook to use more studies and obser- vations than just the most rigorous. It is therefore important that the Handbook user read caveats provided, and also pick up on available study design clues that can help determine how much reliance to put on any given finding. Three categories of problems to be alert for are highlighted here: issues of measurement and statistical significance, effects of confounding events and envi- ronments, and additional analytical concerns. Measurement and Statistical Significance Field observations of traveler response to transportation system changes are typically provided by user surveys, or by “before-and-after” counts of volume, passengers, or revenue. “Before-and- after” surveys and counts and the analyses based on them are essentially two-point-in-time lon- gitudinal studies, with one set of observations before a system-change action or program, and the other afterward. In certain cases full regional travel survey protocols have been followed in before-and-after surveys of affected areas and “control” (unaffected) areas. In other cases there may be no “before” survey at all, with reliance placed on retrospective questions designed to elicit information on behavior prior to the system change. Related issues to be alert to include (Parker et al., 2007, Pratt, 1967, Higgins and Johnson, 1999, Pratt et al., 2000a, Brög et al., 2009): • Inadequate count or survey sample size. • Insufficient survey controls with the potential, especially in the case of self-administered sur- veys, to skew results. • Incomplete system change and travel data, and data definitional problems, which may intro- duce bias or misunderstandings. • Lack of adequate determination that the outcome was statistically significant. Travel survey controls of importance include both sample selection and controlling for response rate bias. Survey response rate biases, particularly those encountered in self-administered sur- veys including post-card and internet applications, are perhaps the most troublesome of survey problems. For example, without designing in controls such as differential factoring, a transit on-board self-administered survey will tend to over-report travel by higher income riders (typi- cally downtown-oriented and concentrated in certain sectors of a city), riders making work trips (primarily peak-period riders), and riders making longer trips. Even broad categories of travel such as the proportion of AM peak-period transit riding may be off by 25 percent or more without correction (Pratt, 1967). Data completeness for a before-and-after evaluation requires a comprehensive, quantitative descrip- tion of the “before” conditions and circumstance, the “before” usage or travel demand, the “after” conditions and circumstance, and the “after” usage or travel demand. If any of these critical ele- ments are missing, then the scope of legitimate study conclusions possible is severely degraded. Travel demand data obtained in parallel for a control area where there was no change or action of the type under study can help in making adjustments for global exogenous events, such as changes in the economy or gasoline prices, giving increased confidence in analysis findings. Transportation system change actions, provisions, and programs can affect travel choices on many dimensions including travel frequency, mode, route, and time of day. On a broader level, there

1-16 may be effects on trip destination choice, and even housing and employment location choice, thereby impacting trip distribution. If data collection is limited to counts, these nuances cannot be rigorously explored. Even before-and-after surveys of travel response often have examined only one or two of these dimensions. This limitation makes difficult such assessments as examination of latent or induced demand or quantification of actual shifts among travel modes as contrasted, for example, to volume changes that simply represent shifts in route taken. Data definitions may also hamper analysis. The “unlinked trip” method of tracking transit ridership— mandated for reporting ridership to the Federal Transit Administration (FTA) and its National Transit Database—is a case in point. An unlinked trip is a passenger trip made in a single tran- sit vehicle, effectively the same as a passenger boarding. A one-way trip from home to work that involves one transfer, such as between two buses or a bus and train, produces two unlinked transit trips. Yet, those two unlinked transit trips serve only one person trip from the rider’s perspective, have the social and environmental benefit of only one transit trip, and often generate only one transit fare. To fully understand whether transit route or system changes have attracted more ridership or the converse, the before-and-after number of “linked trips” must be determined. An example of a linked trip is a one-way trip from home to work with or without one or more changes of vehicle. An increase in unlinked transit trips may reflect nothing more than the effect of more forced transfers (Pratt et al., 2000a). The “other side of the coin” in the linking of trips is seen in analysis of walking or bicycling to and from transit. Conventional regional travel survey processing assigns a single mode to “linked” multimodal trips, with mode designation done on the basis of a fairly standard hierarchy. This process typically causes certain information to be effectively lost, with non-motorized travel modes being most affected (Victoria Transport Policy Institute, 2007, Schneider, 2011). Thus, walking to a bus is subsumed into the bus trip and not counted in the walk mode, and if the bus is used as a feeder to urban rail transit, the bus trip is subsumed as well. Even in the National Household Transportation Survey (NHTS), walking and cycling trips made to access and egress transit are not entered as individual trip records in the primary trip data summarized by mode (Clifton and Krizek, 2004). Without specialized surveys and/or survey processing, use of affected modes will be undercounted. Non-motorized travel analysis offers another classic example of a definitional problem that ham- pers evaluation and comparisons. Some key surveys (such as the NHTS) count an out-and-back “loop” walking or cycling trip done for recreation or exercise as two trips, while other equally prestigious surveys and research treat such activity as one trip, twice as long. The litany of analyti- cal problems faced in particular by the non-motorized travel analyst is examined at length in the “Analytical Considerations” subsection of Chapter 16, “Pedestrian and Bicycle Facilities.” The use of full regional travel survey protocols in before-and-after surveys is an expensive trans- portation change/action evaluation approach that has been limited to date to use in assessments of individualized marketing of transit, walking, bicycling, and carpooling. The essentially volun- tary travel behavior changes involved represent incremental shifts of mode in an experimental environment easily clouded by many uncontrolled factors, making reliable discernment of the travel demand changes a challenge (Brög et al., 2009, Parker et al., 2007). Survey issues include possible under-reporting by respondents of outcomes deemed to be undesirable and other biases that are particularly of concern if the survey is not well isolated, from the perspective of partici- pants, from the transportation change/action involved (Ker, 2011). The same challenge applies, in varying degrees, to every transportation system change or intervention evaluation that might be undertaken using travel surveys. These issues are addressed in Chapter 16 as well, but within the “Individualized Marketing” discussion under “Response by Type of NMT Strategy”—“Walking/ Bicycling Promotion and Information.”

1-17 Confidence in survey findings can be increased not only by use of control groups, but also by employing “external evidence.” If transit revenues, passenger counts, and individual-mode vol- ume counts move in the same direction and with the same general magnitude as travel demand shifts identified in “before” and “after” surveys or retrospective studies, evidence of validity is provided. A basic question in impact evaluations is, “Did something really change?” It is not enough to know that the final measurement of carpoolers, vehicles parked, transit riders, pedestrians, or vehicle emissions was some percentage more or less than in the “before” condition. It is also important to know the fluctuation around the average, so that the significance of the change relative to normal variation can be judged. There are time-honored statistical tests for determining significance, but they are all too often not applied, or perhaps applied only to some variables. For many “changes” or “differences” reported in the literature, it is simply not known whether they are meaningful or not (Higgins and Johnson, 1999). Confounding Events and Environments The other key question in evaluations of transportation changes besides “Did something change?” is “What caused it?” (Higgins and Johnson, 1999). The analyst who evaluates system change impacts must be especially cautious when interpreting results that may have been affected by confounding events or unique circumstances. Such events or circumstances may adversely affect the transferability of results from one location or time period to another (FHWA, 1974). Alternative explanations for travel changes that might otherwise be attributed to a transportation system change of interest may include (Higgins and Johnson, 1999): • External Transportation Events—transportation system changes unrelated to the change of interest, such as new or closed facilities, altered levels of enforcement, transit service changes, transportation labor actions, construction delays, and altered parking charges, highway tolls, transit fares, or employer involvement in parking/commute cost subsidies. • Other External Events—changes in the local/national economy or unemployment rates, changes in government regulations including taxes, social unrest, natural disasters, and special events. • Effects of Testing—impacts of the experimental process itself on travel behavior, as when travelers learn more about their travel options as a result of initial surveys. • Maturation Effects—gradual changes in the circumstances of travelers, most pertinent when a given sample is surveyed at two or more points in time, or regression toward the average of subjects chosen for their above- or below-average behavior. • Measurement Inconsistencies—changes in the measurement process, specifically including alterations in survey procedures and questions, and drop-off in survey response rates and associated reliability. An example of external events is provided by the two periods of fuel shortages which occurred during the 1970s, in 1973–74 and in 1979. Because of the drop-off in quasi-experimental “before- and-after” research following cessation of the Systems and Methods Demonstrations of the Urban Mass Transportation Administration (now Federal Transit Administration) in the early 1980s, and the uniqueness of some of the demonstrations, it remains necessary and useful to refer back to

1-18 findings from 1970s investigations. The observed traveler responses to ridesharing and transit programs instituted concurrently with these fuel shortages may not be fully applicable, however, to times of normal gasoline supply (Pratt and Copple, 1981, Pratt et al., 2000a). In the same way, unique circumstances in a particular city may influence system change results, sometimes in a markedly atypical manner, and other times by simply reinforcing or dampening impacts. The size and geography of cities serve to concentrate or disperse travel, affecting facil- ity volumes or ridership. Cities bordering oceans or lakes have their radial travel concentrated in perhaps five major sectors or broad corridors, as in Chicago, or fewer in a mountainous urban area like San Diego. In extreme contrast, a flatlands pair of cities such as Minneapolis and St. Paul may have 15 separate broad corridors (eight for each city with one shared) over which radial travel is distributed (Deen and Pratt, 1992). An example of results reinforcement is provided by New York, where it can be reasonably certain that modifications to transit fares during the 1990s were accompanied by especially positive results because of the then booming economy in New York City proper. Conversely, statistical analysis shows that patronage losses in the late 1980s in Dallas during a period of fare increases and service reductions were exacerbated by a local recession (See Chapter 12, “Transit Pricing and Fares”). Frequently, a particular system change is directly linked with other simultaneous changes which may affect the results. For example, transit fare reductions have often been accompanied by pro- motional campaigns and increases in transit service frequency and coverage. TDM programs inherently involve multiple actions. Under such circumstances it is difficult if not impossible to separate the impact attributable to each of the individual actions involved. The fact that certain exogenous events can have impacted outcomes, for example earthquake destruction or collapse of a major bridge, may seem overly obvious. However, if not documented in connection with data presentations, they may raise questions or be overlooked in later uses and interpretations of the results affected. Factors such as weather conditions or facility maintenance clearly are subject to being overlooked if they have not been explicitly recorded as part of the data- gathering record. The time required for use of a new service or facility to become established creates difficulties in before-and-after analysis design and in interpreting “after” survey results. Table 16-113, “Motorized and NMT Facility Usage Maturation Experience,” in Chapter 16, summarizes examples of usage stabilization rates. This table shows that while new urban rail facility ridership appears to reach equilibrium in less than 2 years, High Occupancy Vehicle (HOV) facility usage may take 4 years on average, and non-motorized travel facility usage has been seen to take 7 or more years to mature. Among the eight cases set forth in the table, the average usage seen at the end of the first 12 months is only 55 percent of stabilized usage. Thus an “after” count or survey taken, say, 2 months after introduction of a new transportation service or facility may pick up only a modest proportion of the ultimate effect. On the other hand, a survey taken too long after a system change will produce results more likely to be confounded by exogenous events such as those enumerated above. Additional Analytical Concerns Some of the traveler response interpretations presented in this Handbook are based in part on travel demand model results, produced either by research models or forecasting models, as con- trasted to observed results. The travel demand modeling efforts utilized are sometimes based on stated preference survey data but are more frequently based on revealed-preference cross-

1-19 sectional survey data from a single point in time. Use of such survey data relies on the assump- tion that impacts over time of transportation system changes can be inferred from the response at a single point in time to differing system characteristics as observed in different locations in an urban area. This is an assumption that has neither been satisfactorily proved nor disproved (Mayworm, Lago and McEnroe, 1980). Other interpretations are based on cross-sectional and other comparative analyses of one sort or another that involve no travel demand modeling per se. With or without models, additional con- cerns with cross-sectional or comparative analyses include (Deakin, 2002): • Reliance of some aggregate studies on a small number of cases, effectively placing reliance on a very small sample. • Existence of problems of scale and aggregation level, to the point where averages mask varia- tions in characteristics. • Substantive definitional inconsistencies affecting key variables, especially across different research agencies and study areas. • Confusion of correlation and causality in interpretation of results, overlooking the fact that just because two or more parameters move in parallel, the one is not necessarily causing the other(s). With respect to the final critical point above, correlation does not prove causality (Committee on Physical Activity, Health, Transportation, and Land Use, 2005). Despite the benefit of structured analysis, certain of the more complex research studies are actually among the more naïve in terms of assumptions made and relationships accepted as meaningful. Concerns such as these listed here, along with presence of unaccounted for confounding events and local area individualities, can make validity of findings—to say nothing of comparison and transferability of findings among locales—both problematic and demanding of special care on the part of the analyst. Depending on circumstances, deficient analytical approaches may result in either overstatement or understatement. Public opinion and oversimplified preference-survey-based evaluations, when inappropriately used as a forecasting tool, are notoriously associated with overestimates. For example, Madison, Wisconsin, 1970s bicycle commuting estimates made in this manner were only half fulfilled (Zehnpfenning et al., 1993). Even supposedly sophisticated forecasting studies of pro- posed toll roads and urban rail systems have an international historical record of overestimation, labeled “optimism bias,” and attributed to institutional factors (Committee for Determination of the State of the Practice in Metropolitan Area Travel Forecasting, 2007). Similar institutional factors may pertain to “smart growth” land development and TDM traffic mitigation impact estimates carried out for private developers and employers. On the other hand, a systematic review of interventions to promote walking suggests that analyti- cal inadequacies can be associated with implemented-action impact underestimation. Of 25 stud- ies, all seven ranked highest for validity produced statistically significant findings, while under one-half of the 13 studies given intermediate rankings had statistically significant outcomes, and none of the five studies with the lowest rankings did. Moreover, the magnitude of the observed effectiveness that was reported declined with lesser analysis validity (Ogilvie et al., 2007). In the case of transportation changes evaluated with before-and-after studies, as already discussed, “after” evaluations done too soon after implementation may underestimate the long-run effects of the action.

1-20 Finally, attention must be paid to the effect of inflation whenever traveler response has been affected by user cost changes, such as transit fare, highway toll, and parking fee modifications. This is particularly important in the case of older data from periods before or during high inflation. The effect of inflation is pertinent whether traveler responses are observed or estimated.3 Absolute changes in user charges should be interpreted in constant dollars. Unless a specific base year is pro- vided for the value of dollar costs, the best assumption is that costs (and foreign currency conver- sions if applicable) are for the time of the event studied, if given, or otherwise for the time period preceding publication of the source material, the date of which is available from the citation. The Handbook user may wish to convert costs to present day dollars, in order to have a better understanding of the scale involved. Appendix B, “Inflation and CPI Conversions,” provides a table of conversions to 2011 dollars along with accompanying guidance. The more severe of the analytical problems introduced by inflation can be largely avoided by describing user cost changes in terms of relative change. The concept of elasticity, discussed further on, is particularly useful in this regard, as it is a relative indicator. demographic Considerations Although the focus of this Handbook is on traveler response to transportation system changes, it is important to recognize that a number of factors external to the transportation system are crucial determinants of travel behavior. Among these factors, the most important are income, auto own- ership, and organization of the built environment. The location and concentration of residences, employment, commercial activity, recreational areas, and other land uses are primary determi- nants of the number, purpose, and orientation of trips in an urban area (the trip generation and distribution). Lower worker or family income and lower auto ownership are associated with lower than average trip generation rates and higher than average transit usage (Pratt and Copple, 1981). The concept of “captivity” is sometimes used to help explain the impact of low incomes and low auto ownership on mode choice. A traveler is considered to be a “captive” to a particular mode if he or she effectively has no alternative means of transportation. Captive transit riders are those who have no automobile available for their trip, and must therefore use transit or forego the trip. System changes designed to attract more transit riders are directed toward “choice” riders—people whose auto availability allows them to choose freely between transit and auto (Curtin, 1968). At the other end of the scale are captive auto users, those trip makers who for some reason must use an automobile. A traveler may be an auto captive because of a need for the car at work, because a side trip requiring an auto is to be made, or for other reasons not necessarily well understood. The auto captive is not expected to be attracted by transit service enhancements and may be deterred from ridesharing, walking, or bicycling. The concept of captivity suffers from some theoretical and practical difficulties. One problem is that it fails to adequately address the mode choice option of traveling as an auto passenger. Another concern is that the condition of captivity, being highly related to auto ownership, is often a matter of choice. Nevertheless, captivity is sometimes referred to in this Handbook, because it is a familiar term and is used in certain Handbook sources. 3 When using travel demand models, it must be presumed unless otherwise stated that the cost expressed in the model pertains to the value of the dollar in the year of the survey upon which the model was based.

1-21 4 A negative sign used with an elasticity value simply indicates that the cause and effect operate in opposite directions. Price elasticities are almost always negative: when the fare, toll or fee goes up, usage normally goes down, and vice versa. Note that for arc elasticities (discussed next) and other closely equivalent elas- ticity formulations, the qualifier “calculated in infinitesimally small increments” applies to examples and applications such as the one given here. 5 This more simplistic computational method has shown up again in domestic road value pricing investiga- tions, but with different nomenclature such as “approximated point elasticities” or even simply “elasticity.” An advancement over the concept of captivity is the approach of distinguishing between “mobil- ity choices” and “travel choices.” If a person does not have a particular mode available for a trip, it is usually because of long-term mobility choices which were made in the past. Mobility choices include the choice of residential location (including proximity to transit service), number of auto- mobiles owned, and employment location. The usual mode for the work trip can even be thought of as mobility choice, as it may affect auto availability for other travel. Short-term travel choices cover the day-to-day decisions concerning trip frequency, distribution, time of day, mode, and route. In this framework, short-term travel choices are always made in the context of longer-term mobility choices (FHWA, 1974). The concept is particularly relevant to examining traveler responses to transportation system changes, because it recognizes not only the short-term impacts of changed travel options, but also the longer-term impacts related to residen- tial location, workplace location, and auto ownership. Concept of elasticity Elasticity measures provide a convenient tool for summarizing quantitative information about over- all travel demand changes in response to certain types of system changes and are used extensively in this Handbook. For elasticity measures to be applicable, the transportation system change must generally be a relative one. In other words, it must involve a quantifiable percentage increase or decrease in the system parameter involved. (There are, however, procedures for addressing free transit, covered further on.) There are a number of elasticities of interest with respect to demand for transportation, including elasticities for changes in the overall amount of transit service, transit frequencies, transit fares, vehicular tolls, parking charges, and gasoline costs (Kemp, 1974). Elasticity estimates are available as well for comparative degrees of presence of certain built environment characteristics (see, for example, Ewing and Cervero, 2010). Transportation elasticities are informally adopted from the economist’s measure “price elastic- ity.” Loosely speaking, price or service elasticity may be defined as the percentage change in the quantity of a commodity or service demanded by the public in response to a one percent change in price or service. For example, if transit service is measured by the number of bus miles operated, the transit service elasticity indicates the percentage change in patronage observed or expected in response to a one percent change in the number of bus miles. If the transit service elasticity is +0.6, a six-tenths-of-one percent patronage increase is indicated for each one percent increase in service.4 There are several different methods for computing elasticity. An expanded discussion of these methods and their application, along with comparative examples and illustrations, is provided in Appendix A, “Elasticity Discussion and Formulae.” The most frequently used form of elasticity in transportation analyses is arc elasticity. There is also a more simplistic form encountered in mostly older transit fare analyses, known as a shrinkage ratio or factor.5

1-22 Arc elasticity is defined by a logarithmic formulation and, except for very large changes in price or service (P) and demand (Q), is closely approximated by a midpoint formulation which makes use of the average value of each independent variable (Webster and Bly, 1980, Mayworm, Lago and McEnroe, 1980, Litman, 2012): log arc elasticity: η = ∆ ∆ = − − log log log log log log Q P Q Q P P 2 1 2 1 mid-point (or linear) arc elasticity: η = ∆ +( ) ÷ ∆ +( ) = ∆ +( ) ∆ +( ) = Q Q Q P P P Q P P P Q Q1 2 1 2 1 2 1 22 2 Q Q P P P P Q Q 2 1 1 2 2 1 1 2 −( ) +( ) −( ) +( ) where h is the elasticity, Q1 and Q2 are the demand before and after, and P1 and P2 are the price or service before and after. Arc elasticity is based on both the original and final values of demand and price or service. All elasticities involving free fares are of mathematical necessity calculated with the mid-point for- mulation. Otherwise, the logarithmic formulation has been used wherever elasticities have been calculated directly from available data in this updated Handbook, and the same was the case in the second edition. Certain values carried over from the first edition Handbook were computed using the mid-point formulation, and are so identified. The shrinkage ratio or shrinkage factor has historically been used as a means of reporting response to transit fare changes, primarily fare increases. It is defined and discussed in detail in Chapter 12, “Transit Pricing and Fares,” under “Response by Type of Strategy”—“Changes in General Fare Level.” There are certain conceptual problems with shrinkage ratios (discussed in more detail in Appendix A) which have led to the predominant use of arc elasticities in this Handbook and most contem- porary works based on quasi-experimental data. Shrinkage ratios and equivalent computations of importance are reported, but arc elasticity conversions are given wherever possible. Figure 1-1 illustrates the differences between the two arc elasticity formulations and the shrinkage ratio using transit fare elasticity as an example. It is essential to take into account that arc or linear elasticities and shrinkage ratios, being formu- lated differently, are not applied the same. Formulae to apply elasticities are provided in Appendix A. It is also important to note that there are inconsistencies and overlap in elasticity nomenclature that hold potential for significant confusion. Appendix A addresses this situation with an illus- trated discussion of definitional differences and misinterpretations.

1-23 referenCeS Brög, W., Erl, E., Ker, I., Ryle, J., and Wall, R., “Evaluation of voluntary travel behaviour change: Experiences from three continents.” Transport Policy, Vol. 16, Issue 6 (2009). Cambridge Systematics, Inc., “Characteristics of Urban Transportation Systems.” Prepared for the Federal Transit Administration, Washington, DC (1992). Clifton, K. J., and Krizek, K. J., “The Utility of the NHTS in Understanding Bicycle and Pedestrian Travel.” Prepared for National Household Travel Survey Conference: Understanding our Nation’s Travel. Washington, DC (November 1–2, 2004). Committee for Determination of the State of the Practice in Metropolitan Area Travel Forecasting, “Metropolitan Travel Forecasting—Current Practice and Future Direction.” TRB Special Report 288. Transportation Research Board of the National Academies, Washington, DC (2007). Committee on Physical Activity, Health, Transportation, and Land Use, “Does the Built Environment Influence Physical Activity?—Examining the Evidence.” TRB Special Report 282. Transportation Research Board and Institute of Medicine of the National Academies, Washington, DC (2005). Curtin, J. F., “Effect of Fares on Transit Riding.” Highway Research Record 213, Highway (now Transportation) Research Board, National Research Council, Washington, DC (1968). Note: Source: Mayworm, Lago and McEnroe (1980). The term “point elasticity” as used in the figure title refers to the derivative of the assumed underlying demand curve—it is not used here Figure 1-1 Elasticities of Different Types Calculated from a Demand Curve with an Initial Point Elasticity of –0.30

1-24 Deen, T. B., and Pratt, R. H., “Chapter 11—Evaluating Rapid Transit.” Public Transportation, Second Edition. Grey and Hoel, Eds. Prentice Hall, Englewood Cliffs, NJ (1992). Deakin, E. A., University of California. Email attachment to the Handbook authors. Berkeley, CA (March 28, 2002). Ewing, R., and Cervero, R., “Travel and the Built Environment—A Meta-Analysis.” Journal of the American Planning Association, Vol. 6, No. 3 (Summer, 2010). Federal Highway Administration, “Design of Procedures to Evaluate Traveler Responses to Changes in Transportation System Supply.” Conference Summary and White Papers. U.S. Department of Transportation, Washington, DC (1974). Higgins, T. J., and Johnson, W. L., “Evaluating transportation programs: Neglected principles,” Transportation, Vol. 26, Issue 4 (1999). Highway Capacity Manual 2010. Transportation Research Board of the National Academies, Washington, DC (December, 2010). Kemp, M. A., Reduced Fare and Fare-Free Urban Transit Services—Some Case Studies. The Urban Institute, Washington, DC (1974). Ker, I., “Too True to be Good? A Response to Morton and Mees (2010).” World Transport Policy & Practice, Vol. 17, No. 1, http://www.eco-logica.co.uk/pdf/wtpp17.1.pdf (May, 2011). Kittelson & Associates, Inc., KFH Group, Inc., Parsons Brinckerhoff Quade and Douglas, Inc., and Hunter-Zaworski, K., “Transit Capacity and Quality of Service Manual, Second Edition.” TCRP Report 100. Transportation Research Board of the National Academies, Washington, DC (2003). Litman, T., Understanding Transport Demands and Elasticities—How Prices and Other Factors Affect Travel Behavior. Victoria Transport Policy Institute, Victoria, BC, Canada, http://www.vtpi.org/ elasticities.pdf (May 25, 2012). Mayworm, P. D., Lago, A. M., and McEnroe, J. M., Patronage Impacts of Changes in Transit Fares and Services. Prepared for the Urban Mass Transportation Administration (now Federal Transit Administration) by Ecosometrics, Incorporated, Bethesda, MD (1980). Ogilvie, D., Foster, C. E., Rothnie, H., Cavill, N., Hamilton, V., Fitzsimons, C. F., and Mutrie, N., “Interventions to promote walking: systematic review.” BMJ (British Medical Journal), doi:10.1136/ bmj.39198.722720.BE; Vol. 334, No. 1204; and Data Supplement (June, 2007). Parker, J. H., Harris, L., Chatterjee, K., Armitage, R., Cleary, J., and Goodwin, P., “Making Personal Travel Planning Work: Research Report.” Prepared by Integrated Transport Planning Ltd for Department for Transport, London, United Kingdom (December, 2007). Pratt, R. H., How to Obtain Transit Travel O-D Data. Paper presented at the Highway (now Transportation) Research Board conference on New Concepts in O-D Studies, Washington, DC (September 14–15, 1967). Pratt, R. H., “Traveler Response to Transportation System Changes: An Interim Introduction to the Handbook.” TCRP Research Results Digest 61, Transportation Research Board, Washington, DC (September, 2003).

1-25 Pratt, R. H., and Copple, J. N., Traveler Response to Transportation System Changes. Second Edition. Prepared for the Federal Highway Administration, Washington, DC (July, 1981). Pratt, R. H., Pedersen, N. J., and Mather, J. J., Traveler Response to Transportation System Changes—A Handbook for Transportation Planners [first edition]. Prepared for the Federal Highway Administration, Washington, DC (February, 1977). Richard H. Pratt, Consultant, Inc.; J. Richard Kuzmyak, Transportation Consultant; Texas Transportation Institute; Parsons Brinckerhoff Quade & Douglas, Inc.; SG Associates, Inc.; Herbert S. Levinson, Transportation Consultant; McCollom Management Consulting, Inc.; Gallop Corporation; K.T. Analytics, Inc.; Cambridge Systematics, Inc., “Research Needs Identified in the Initial TCRP B-12 Work Effort.” Draft Report. Unpublished (April, 2000a). Richard H. Pratt, Consultant, Inc.; Texas Transportation Institute; Cambridge Systematics, Inc.; Parsons Brinckerhoff Quade & Douglas, Inc.; SG Associates, Inc.; and McCollom Management Consulting, Inc., “Traveler Response to Transportation System Changes—Interim Handbook.” TCRP Web Document 12 (Project B-12): Contractor’s Interim Handbook. http://onlinepubs.trb.org/ onlinepubs/tcrp/tcrp_webdoc_12.pdf (March, 2000b). Reno, A., Kuzmyak, R., and Douglas, B., “Characteristics of Urban Travel Demand.” TCRP Report 73. Transportation Research Board, Washington, DC (2002). Rogers, L. H., “Have We Lost the Message before We Gain the Knowledge?” Transportation Research Board 76th Annual Meeting, Preprint Paper No. 971382, Washington, DC (1997). Schneider, R. J., Understanding Sustainable Transportation Choices: Shifting Routine Automobile Travel to Walking and Bicycling. PhD Dissertation. University of California Transportation Center, University of California, Berkeley, CA (Spring, 2011). Victoria Transport Policy Institute, “Evaluating Nonmotorized Transport: Techniques for Measuring Walking and Cycling Activity and Conditions.” Online TDM Encyclopedia. http://www. vtpi.org/tdm/tdm63.htm (Webpages updated August 27, 2007). Webster, F. V., and Bly, P. H., The Demand for Public Transport. Transport and Road Research Laboratory, Crowthorne, England (1980). Zehnpfenning, G. H., Design Ventures, Inc., Cromar, J., and Maclennan, S. J., “Case Study No. 4— Measures to Overcome Impediments to Bicycling and Walking.” National Bicycling and Walking Study. Federal Highway Administration, U.S. Department of Transportation, Washington, DC (1993).