USF Libraries
USF Digital Collections

A comparative analysis of travel time expenditures in the United States

MISSING IMAGE

Material Information

Title:
A comparative analysis of travel time expenditures in the United States
Physical Description:
Book
Language:
English
Creator:
Toole-Holt, Lavenia Anne
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
budget
discretionary travel
socio-demographic
Nationwide Personal Transportation Survey
National Household Travel Survey
Dissertations, Academic -- Civil Engineering -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Literature on transportation planning and modeling is replete with the concept of a travel time budget. According to this concept, average daily travel times tend to be relatively constant. However, evidence from the 1983 Nationwide Personal Transportation Survey and the 2001 National Household Travel Survey suggest that the average daily travel time has increased by 1.9 minutes per year. Understanding travel time expenditures is important for forecasting travel demand, especially future vehicle miles of travel. Historically, travel demand models considered vehicle availability and income as limiting factors for travel, but going forward time may be the constraint. As individuals spend more time devoted to travel, less time will be available for other activities. Therefore, future travel demand is dependent on people's willingness to spend time traveling. Growth of travel demand has been per capita based not just population based. This has been enabled by several cultural trends, including fewer children to care for; specialization of activities; multitasking during travel, for example, cell phone use can add value to travel time; seeking socialization away from home; and increases in real income enabling more activity participation. This study will report the increase in average daily travel time expenditures and analyze the increase by various demographic segments of the population. Travel time expenditures are also related to activity participation, the characteristics of the area, and many other interrelated factors at the person level. Aggregate values will be used to investigate the general relationships between daily travel time expenditures and socio-demographic characteristics. Careful consideration of the implications of the increase in travel time, as well as the changes in society that have contributed to these changes will be explored. The increase in travel time expenditures is likely to play a significant role in future travel demand growth in the United States and will impact the performance of the transportation system going forward. If travel time expenditures continue to grow, the hope for slowing VMT growth may not materialize. Understanding the mechanics of why people are traveling more will aid planners and modelers in estimating future travel demand.
Thesis:
Thesis (M.S.C.E.)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Lavenia Anne Toole-Holt.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 138 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001478748
oclc - 56389680
notis - AJS2438
usfldc doi - E14-SFE0000390
usfldc handle - e14.390
System ID:
SFS0025082:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001478748
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 040811s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000390
035
(OCoLC)56389680
9
AJS2438
b SE
SFE0000390
040
FHM
c FHM
090
TA145 (ONLINE)
1 100
Toole-Holt, Lavenia Anne.
2 245
A comparative analysis of travel time expenditures in the United States
h [electronic resource] /
by Lavenia Anne Toole-Holt.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Thesis (M.S.C.E.)--University of South Florida, 2004.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 138 pages.
520
ABSTRACT: Literature on transportation planning and modeling is replete with the concept of a travel time budget. According to this concept, average daily travel times tend to be relatively constant. However, evidence from the 1983 Nationwide Personal Transportation Survey and the 2001 National Household Travel Survey suggest that the average daily travel time has increased by 1.9 minutes per year. Understanding travel time expenditures is important for forecasting travel demand, especially future vehicle miles of travel. Historically, travel demand models considered vehicle availability and income as limiting factors for travel, but going forward time may be the constraint. As individuals spend more time devoted to travel, less time will be available for other activities. Therefore, future travel demand is dependent on people's willingness to spend time traveling. Growth of travel demand has been per capita based not just population based. This has been enabled by several cultural trends, including fewer children to care for; specialization of activities; multitasking during travel, for example, cell phone use can add value to travel time; seeking socialization away from home; and increases in real income enabling more activity participation. This study will report the increase in average daily travel time expenditures and analyze the increase by various demographic segments of the population. Travel time expenditures are also related to activity participation, the characteristics of the area, and many other interrelated factors at the person level. Aggregate values will be used to investigate the general relationships between daily travel time expenditures and socio-demographic characteristics. Careful consideration of the implications of the increase in travel time, as well as the changes in society that have contributed to these changes will be explored. The increase in travel time expenditures is likely to play a significant role in future travel demand growth in the United States and will impact the performance of the transportation system going forward. If travel time expenditures continue to grow, the hope for slowing VMT growth may not materialize. Understanding the mechanics of why people are traveling more will aid planners and modelers in estimating future travel demand.
590
Co-adviser: Pendyala, Ram M.
Co-adviser: Polzin, Steven E.
653
budget.
discretionary travel.
socio-demographic.
Nationwide Personal Transportation Survey.
National Household Travel Survey.
0 690
Dissertations, Academic
z USF
x Civil Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.390



PAGE 1

A Comparative Analysis of Travel Time Expenditures in the United States by Lavenia Anne Toole-Holt A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Civil Engineering Department of Civil and Environmental Engineering College of Engineering University of South Florida Co-Major Professor: Ram M. Pendyala, Ph.D. Co-Major Professor: Steven E. Polzin, Ph.D. Jian J. Lu, Ph.D. Xuehao Chu, Ph.D. Date of Approval: July 6, 2004 Keywords: national household travel survey, nationwide personal tr ansportation survey, budget, discretionary trav el, socio-demographic Copyright 2004, Lavenia Anne Toole-Holt

PAGE 2

ACKNOWLEDGEMENTS I would like to express my sincere gratitude to my co-major professors, Dr. Steve Polzin and Dr. Ram Pendyala, for their continuous support and encouragement throughout this process. Their ideas, advice, and availability were critical to my performance on this thesis. I would also like to thank Dr. Xuehao Chu and Dr. Jian Lu for serving on my committee. Dr. Chus input and assistance was especially invaluable. I am also grateful to Dr. Steve Polzin for his guidance and support during my employment at the Center for Urban Transportation Research. The experiences and opportunities that I have enjoyed as a graduate research assistant have meant a great deal to me and helped shape my future career goals. I would like to acknowledge the Center for Urban Transportation Research for enabling students to become actively involved with projects and for providing an environment that not only fosters education, but also develops friendships. Words can not express how thankful I am to have had the support of my family and friends throughout my education. Without their love, understanding, and encouragement, this experience would have been much more difficult.

PAGE 3

TABLE OF CONTENTS LIST OF TABLES iii LIST OF FIGURES v ABSTRACT ix CHAPTER 1. INTRODUCTION 1 1.1 Background 1 1.2 Motivations 2 1.3 Objective and Scope of the Study 3 1.4 Outline of Thesis 4 CHAPTER 2. LITERATURE REVIEW 5 2.1 Travel Time Budgets 5 2.2 Socio-demographic Characteristics 8 2.2.1 Age 9 2.2.2 Gender 10 2.2.3 Employment Status 10 2.2.4 Household Size 11 2.2.5 Income 11 2.2.6 Private Vehicle Ownership 11 2.2.7 Life Style Group 12 2.3 Area Characteristics 13 2.4 Activity Time Allocation 17 2.5 International Trends in Travel Time 20 CHAPTER 3. SETTING THE STAGE 23 3.1 Aging Population 24 3.2 Household Size 26 3.3 Labor Force Participation 29 3.3.1 Labor Force Specialization 30 3.4 Income 33 3.5 Vehicle Availability and Use 35 3.6 Time Allocation Shifts 40 3.6.1 Time-Deepening 44 3.6.2 Technology 45 i

PAGE 4

CHAPTER 4. DESCRIPTION OF DATA 47 4.1 Survey Description 47 4.1.1 1983 Nationwide Personal Transportation Survey 48 4.1.2 2001 National Household Travel Survey 51 4.2 Survey Comparability 54 CHAPTER 5. DESCRIPTIVE ANALYSIS 57 5.1 Daily Travel Time Expenditure Analysis 57 5.1.1 Travel Mode 63 5.1.2 Trip Purpose 69 5.1.3 Average Daily Travel Time by Travel Day 76 5.1.4 Average Daily Travel Time by Gender 79 5.1.5 Average Daily Travel Time by Age Group 85 5.1.6 Average Daily Travel Time by Worker Status and Gender 92 5.1.7 Average Daily Travel Time by Life Cycle 94 5.1.8 Average Daily Travel Time by Household Income 97 5.1.9 Average Daily Travel Time by Vehicle Availability 100 5.1.10 Average Daily Travel Time by Driver Status 103 5.1.11 Average Daily Travel Time by Education Level 105 5.1.12 Average Daily Travel Time by Race 107 5.1.13 Average Daily Travel Time by Home Ownership 109 5.1.14 Average Daily Travel Time by MSA Size 110 5.2 Conclusions 112 CHAPTER 6. DICUSSION OF QUANTITATIVE ANALYSIS 113 6.1 Share Analysis 113 6.2 Multivariate Model 115 6.3 Travel Time Forecast 115 CHAPTER 7. CONCULSIONS AND FUTURE RESEARCH 117 7.1 Summary of Findings 117 7.2 Future Research 120 REFERENCES 122 ii

PAGE 5

LIST OF TABLES Table 4.1 Sample Size and Weighted Population for 1983 NPTS 51 Table 4.2 Sample Size and Weighted Population for 2001 NHTS 54 Table 5.1 Summary of Travel Time Expenditures 57 Table 5.2 Sample Size and Weighted Population by Travel Group 58 Table 5.3 Summary of Travel Time Expenditures by Mode 63 Table 5.4 Summary of Travel Time Expenditures by Purpose 71 Table 5.5 Summary of Travel Time Expenditures by Purpose in 2001 74 Table 5.6 Summary of Travel Time Expenditures by Travel Day 76 Table 5.7 Average Travel Time Expenditures per Person by Travel Day and Trip Purpose 78 Table 5.8 Summary of Travel Time Expenditures by Gender 79 Table 5.9 Summary of Travel Time Expenditures by Purpose for Males 83 Table 5.10 Summary of Travel Time Expenditures by Purpose for Females 83 Table 5.11 Summary of Travel Time Expenditures by Age Group 85 Table 5.12 Summary of Travel Time Expenditures by Age Group for Males 87 Table 5.13 Summary of Travel Time Expenditures by Age Group for Females 89 Table 5.14 Summary of Travel Time Expenditures by Worker Status 92 Table 5.15 Summary of Travel Time Expenditures by Life Cycle 95 Table 5.16 Summary of Travel Time Expenditures by Household Income 97 Table 5.17 Summary of Travel Time Expenditures by Household Income and Travel Mode 99 iii

PAGE 6

Table 5.18 Summary of Travel Time Expenditures by Number of Vehicles in Household 100 Table 5.19 Summary of Travel Time Expenditures by Number of Vehicles and Travel Mode 102 Table 5.20 Share of Population by Income Quartile and Number of Vehicles 102 Table 5.21 Summary of Travel Time Expenditures by Driver Status for Persons Over 16 Years of Age 103 Table 5.22 Summary of Travel Time Expenditures by Education Level 105 Table 5.23 Summary of Travel Time Expenditures by Race 107 Table 5.24 Summary of Travel Time Expenditures by Home Ownership 109 Table 5.25 Summary of Travel Time Expenditures by MSA Size 110 Table 6.1 Average Minutes of Travel per Person Based on Population Distributions 114 iv

PAGE 7

LIST OF FIGURES Figure 1.1 Average Daily Travel Time Expenditure per Person 1 Figure 3.1 Median Age of the United States Population, 1970 to 2000 25 Figure 3.2 Age Distribution of United States Population 25 Figure 3.3 Average Household Size, 1930 to 2000 27 Figure 3.4 Distribution of Households by Size, 1940 to 2000 27 Figure 3.5 Average Number of Own Children under 18 per Family 28 Figure 3.6 Share of Population by Marital Status, 1950 to 2000 28 Figure 3.7 Civilian Labor Force Participation Rate, 1948 to 2002 29 Figure 3.8 Share of Total Food Expenditures Spent on Meals Away from Home 31 Figure 3.9 Educational Attainment of the Population 15 Years and Older 32 Figure 3.10 Median Income in the United States, 1980 to 2002 34 Figure 3.11 Income and Travel Cost per Vehicle Mile of Travel 34 Figure 3.12 Registered Motor Vehicles, 1970 to 2002 36 Figure 3.13 Ratio of Vehicles to Adults, Drivers, and Workers 36 Figure 3.14 Share of Zero-Vehicle Household in the United States 37 Figure 3.15 Share of Population with a Drivers License by Age Group and Gender, 2002 37 Figure 3.16 Annual Vehicle Miles of Travel, 1970 to 2002 38 Figure 3.17 Daily Time Allocation by Major Time Use Category 41 Figure 3.18 Annual Average Change in Minutes per Day per Year by Major Time Use Category and Travel Time 41 v

PAGE 8

Figure 3.19 Daily Time Allocation by Major Time Use Category for Males 43 Figure 3.20 Daily Time Allocation by Major Time Use Category for Females 43 Figure 5.1 Distribution of Travel Time Expenditures by Number of Persons 59 Figure 5.2 Distribution of Travel Time Expenditures by Population Share 59 Figure 5.3 Distribution of Daily Trip Count by Number of Person 61 Figure 5.4 Distribution of Daily Trip Count by Population Share 61 Figure 5.5 Distribution of Trip Distance in Miles 62 Figure 5.6 Distribution of Trip Duration in Minutes 62 Figure 5.7 Average Daily Travel Time per Traveler by Mode 64 Figure 5.8 Average Daily Travel Time per Person by Mode 64 Figure 5.9 Distribution of Daily Travel Time Expenditures by Private Vehicle 65 Figure 5.10 Average Daily Trip Count per Person by Mode 66 Figure 5.11 Average Trip Distance in Miles by Mode 67 Figure 5.12 Distribution of Person Miles of Travel by Share of Travelers 68 Figure 5.13 Distribution of Vehicle Miles of Travel by Share of Travelers 68 Figure 5.14 Average Number of Trips per Person by Purpose 71 Figure 5.15 Average Daily Travel Time per Traveler by Purpose 72 Figure 5.16 Average Daily Travel Time per Person by Purpose 72 Figure 5.17 Share of Daily Travel Time by Purpose in 1983 73 Figure 5.18 Share of Daily Travel Time by Purpose in 2001 73 Figure 5.19 Estimated Activity Durations per Person in 1983 75 Figure 5.20 Estimated Activity Durations per Person in 2001 75 Figure 5.21 Average Daily Travel Time per Person by Travel Day 77 vi

PAGE 9

Figure 5.22 Share of Non-Travelers by Travel Day 77 Figure 5.23 Growth in Average Daily Travel Time per Person by Trip Purpose and Travel Day from 1983 to 2001 78 Figure 5.24 Average Daily Travel Time per Traveler by Gender 80 Figure 5.25 Average Daily Travel Time per Person by Gender 80 Figure 5.26 Distribution of Daily Travel Expenditures by Share of Males 81 Figure 5.27 Distribution of Daily Travel Expenditures by Share of Females 81 Figure 5.28 Average Daily Travel Time per Male by Purpose 84 Figure 5.29 Average Daily Travel Time per Female by Purpose 84 Figure 5.30 Average Daily Travel Time per Person by Age Group 86 Figure 5.31 Share of Non-Travelers by Age Group 86 Figure 5.32 Average Daily Travel Time per Male by Age Group 88 Figure 5.33 Share of Male Non-Travelers by Age Group 88 Figure 5.34 Average Daily Travel Time per Female by Age Group 90 Figure 5.35 Share of Female Non-Travelers by Age Group 90 Figure 5.36 Increase of Average Travel Time per Person for Males and Females from 1983 to 2001 91 Figure 5.37 Difference between Average Travel Time for Males and Females by Age Group 91 Figure 5.38 Average Daily Travel Time per Person by Worker Status 93 Figure 5.39 Share of Adult Non-Travelers by Worker Status and Gender 93 Figure 5.40 Average Daily Travel Time per Person by Life Cycle 96 Figure 5.41 Share of Non-Travelers by Life Cycle 96 Figure 5.42 Average Daily Travel Time per Person by Household Income 98 Figure 5.43 Share of Non-Travelers by Household Income 98 vii

PAGE 10

Figure 5.44 Average Daily Travel Time per Person by Number of Vehicles 101 Figure 5.45 Share of Non-Travelers by Number of Vehicles 101 Figure 5.46 Average Daily Travel Time per Person by Driver Status 104 Figure 5.47 Share of Non-Travelers by Driver Status 104 Figure 5.48 Average Daily Travel Time per Person by Education Level 106 Figure 5.49 Share of Non-Travelers by Education Level 106 Figure 5.50 Average Daily Travel Time per Person by Race 108 Figure 5.51 Share of Non-Travelers by Race 108 Figure 5.52 Average Daily Travel Time per Person by Home Ownership 109 Figure 5.53 Average Daily Travel Time per Person by MSA Size 111 Figure 5.54 Share of Non-Travelers by MSA Size 111 viii

PAGE 11

A COMPARATIVE ANALYSIS OF TRAVEL TIME EXPENDITURES IN THE UNITED STATES Lavenia Anne Toole-Holt ABSTRACT Literature on transportation planning and modeling is replete with the concept of a travel time budget. According to this concept, average daily travel times tend to be relatively constant. However, evidence from the 1983 Nationwide Personal Transportation Survey and the 2001 National Household Travel Survey suggest that the average daily travel time has increased by 1.9 minutes per year. Understanding travel time expenditures is important for forecasting travel demand, especially future vehicle miles of travel. Historically, travel demand models considered vehicle availability and income as limiting factors for travel, but going forward time may be the constraint. As individuals spend more time devoted to travel, less time will be available for other activities. Therefore, future travel demand is dependent on peoples willingness to spend time traveling. Growth of travel demand has been per capita based not just population based. This has been enabled by several cultural trends, including fewer children to care for; specialization of activities; multitasking during travel, for example, cell phone use can add value to travel time; seeking socialization away from home; and increases in real income enabling more activity participation. ix

PAGE 12

This study will report the increase in average daily travel time expenditures and analyze the increase by various demographic segments of the population. Travel time expenditures are also related to activity participation, the characteristics of the area, and many other interrelated factors at the person level. Aggregate values will be used to investigate the general relationships between daily travel time expenditures and socio-demographic characteristics. Careful consideration of the implications of the increase in travel time, as well as the changes in society that have contributed to these changes will be explored. The increase in travel time expenditures is likely to play a significant role in future travel demand growth in the United States and will impact the performance of the transportation system going forward. If travel time expenditures continue to grow, the hope for slowing VMT growth may not materialize. Understanding the mechanics of why people are traveling more will aid planners and modelers in estimating future travel demand. x

PAGE 13

CHAPTER 1 INTRODUCTION 1.1 Background Over the past several decades, significant changes in travel behavior in the United States have occurred. Evidence from the 1983 Nationwide Personal Transportation Survey and the 2001 National Household Travel Survey suggest that the average daily travel time per person has increased by 1.9 minutes per year. Figure 1.1 presents the trend in average daily travel time expenditures from the national travel survey series. Changes in society, technology, income, attitudes, and socio-demographics may have contributed to the travel time growth. Figure 1.1 Average Daily Travel Time Expenditure per Person 82.366.247.458.201020304050607080901983198519871989199119931995199719992001Minutes per Person per Day 1

PAGE 14

The increase of daily travel time expenditures is contrary to the concept of a travel time budget. According to this concept, average daily travel times tend to be relatively constant. One group of researchers, including Zahavi, argued that average daily travel time expenditures are stable at about 1.1 hours per day per traveler. However, most researchers have concluded that travel time expenditures are not constant, except perhaps at an extremely aggregate level. The concept of a travel time budget was developed in an attempt to improve travel demand models by finding regularities that could be attributed to household characteristics, area characteristics, or the transportation system (Mokhtarian and Chen, 1999). According to Pas, since the 1970s, the concept of time has moved from relative obscurity to center stage in travel demand analysis and modeling (1997). Although some researchers have attempted to validate the concept of travel time budgets at a highly aggregated level, other researchers consider travel time expenditures at disaggregate levels in an attempt to better understand travel demand. 1.2 Motivations The observation that average travel time expenditures have steadily increased based on data from the national travel survey series motivated a more detailed analysis of the trend. The intent of this analysis is to determine the relevance of socio-demographic characteristics in the growth of average daily travel time to either verify or refute the hypothesis that these characteristics are key factors in driving daily travel time expenditures at the person level. Understanding travel time expenditures is important for forecasting travel demand, especially future vehicle miles of travel. Historically, travel demand models 2

PAGE 15

considered vehicle availability and income as limiting factors for travel, but going forward time may be the more critical constraint. As individuals spend more time devoted to travel, less time will be available for other activities. Therefore, future travel demand is dependent on peoples willingness to spend time traveling. One obviously wonders about the extent to which travel time budgets will continue to increase. Changes in culture, work force participation, family composition and responsibilities, and household activities all underlay the extent to which individuals may be willing or able to divert more time to travel. More fast food, fewer children to parent, stay-press fabrics with no ironing, microwaves for fast cooking, smaller households resulting in a need to travel for socialization, women in the work force who knows what other changes may result in a desire or need to spend more time traveling. Unless individuals quit budgeting more time to travel, the increase in travel time expenditures is likely to play a significant role in future travel demand growth in the United States. Developing a richer understanding of what factors are driving travel time commitments will be a prerequisite to having more confidence in estimating future travel demand. 1.3 Objective and Scope of the Study The primary objective of this thesis is to determine what factors may have contributed to the growth of average daily travel time expenditures. Travel time expenditures are related to socio-demographic characteristics, activity participation, the characteristics of the area, and many other interrelated factors at the person level. Detailed analysis of daily travel time expenditures by various socio-demographic and area characteristics will be completed using data from the 1983 Nationwide Personal 3

PAGE 16

Transportation Survey and the 2001 National Household Travel Survey. Aggregate values will be used to investigate the general relationships between daily travel time expenditures and socio-demographic characteristics. This research aims to determine if various socio-demographic characteristics are related to the growth in average daily travel time expenditures or if other factors are involved. The contribution of changing distribution for several characteristics to total travel time will be quantified. Distributions of travel time by mode and trip purpose will be included. Changes in technology and society that could have contributed to the increase of travel times will also be discussed as will the implications of continued growth. 1.4 Outline of Thesis The remainder of this thesis is organized as follows. The next chapter provides a review of literature relevant to the study of travel time expenditures, including discussions of travel time budgets, previous studies relating socio-demographics to travel time, activity time allocation, and international trends. This chapter is followed by a discussion of trends in the United States that have influenced travel behavior and particularly time allocation. Chapter 4 describes the data sets used for the analysis of travel time expenditures. This chapter is followed by a detailed analysis of person level characteristics from the 1983 Nationwide Personal Transportation Survey and the 2001 National Household Travel Survey. Chapter 6 discusses quantifying the relevance of socio-demographic characteristics to overall travel time growth. Finally, conclusions and implications of the research findings are discussed in Chapter 7 of this thesis. 4

PAGE 17

CHAPTER 2 LITERATURE REVIEW Understanding the complexities of personal time allocation is critical to understanding travel demand. For more than forty years, researchers have grappled to understand the demand for travel by considering time (Pas, 1997). This chapter will briefly explore literature related to travel time, including travel time budgets, relationships with socio-demographic characteristics, area characteristics, activity time allocation, and international trends. The topics discussed as part of this literature review provide the context for the study of the growth of travel time expenditures in this thesis. 2.1 Travel Time Budgets Literature on transportation planning and modeling is replete with the concept of a travel time budget. According to this concept, average daily travel times tend to be relatively constant. The word budget implies that the allocation of time, money, or generalized resources to travel would not be influenced by policy, trends, or costs (Goodwin, 1981). Despite the resilience of this concept in the literature, many researchers have not observed constant time expenditures for travel. Peters, Wilde, Clement, and Peeters drew from various fields including home economics, psychology, biology, sociology, and other fields to explain a constant travel time budget based on three categories (2001; cited in Wee, Rietveld, and Muers, 2002). First, reductionistic approaches assume that most human behavior is related to genetic 5

PAGE 18

history. With regard to travel times, these approaches argue that humans have a need for a minimum amount of exercise and that a complex system of hormones must balance the costs of travel, such as stress and energy use, with the benefits and the biological clock (Peters et al., 2001; cited by Wee et al., 2002). The second set of approaches is reconstructive which describe behavior in terms of mathematics. According to this theory, human behavior can be explained by utility where individuals act rationally and find an optimal balance between time for activities and related travel (Bhat and Koppelman, 1999 and Peters et al., 2001; cited by Wee et al., 2002). Travel time budgets can also be explained using history, culture, socio-psychology, social or geographic perspectives. These approaches are referred to as contextualizing. According to these explanations, travel time can not be explained by individual behavior, but instead by the context within which the individual acts (Peters et al., 2001; cited by Wee et al., 2002). While the reductionistic approach provides little opportunity to explain increases in travel time expenditures, contextualizing can be used to explain increases based on changes within the society (Wee et al., 2002). Chapter 3 will attempt to contextualize the growth of travel time expenditures in the United States by exploring various societal trends. The concept of a travel time budget implies that changes in technology, economic growth, or transportation policy do not have an impact on travel time expenditures (Wee et al., 2002). In particular, as speed increases, because of faster modes or additional capacity, travel distances also increase so that travel time remains relatively stable (Zahavi and Ryan, 1980; Hupkes, 1982; cited in Mokhtarian and Chen, 2002 and Wee et al., 2002). Furthermore, trips that may be avoided because of telecommunications, such 6

PAGE 19

as online shopping or telecommuting, or reducing travel time by having closer destinations will not redistribute the time saved to other activities, but instead dedicate the time to additional travel (Mokhtarian and Chen, 1999). This concept connects travel time budgets to the debate over induced demand (Noland and Lem, 2002, cited in Mokhtarian and Chen, 2002). Some argue that, from an energy and air quality perspective, it is ineffective to expand capacity and improve speeds, because people will be prone to travel more (Mokhtarian and Chen, 2002). The theory that people dedicate the same generalized expenditures, time and money, for travel was first discussed by Tanner in the early 1960s (Tanner, 1961; cited in Schafer, 2000). Zahavi also concluded that urban travelers, who make at least one motorized trip per day, spend about 1.1 hours per day traveling and 10 to 15 percent of household income on travel if the household owns a vehicle (1981). Using these constraints, Zahavi developed an urban travel demand model to simulate various characteristics of the United States transportation system. The characteristics included travel distances, modal splits, and trip speeds (1981; cited in Schafer, 2000). The specifications of the Unified Mechanism of Travel (UMOT) model, developed by Zahavi, include a predictable functional form of expenditures, spatial and temporal stability of time and money expenditures, and temporal stability of travel budgets within various population groups (Zahavi, 1979). The UMOT process was based on the assumption that travel time expenditures showed regularities according to household characteristics, the transportation system, and urban form (Zahavi, 1979). Zahavi and Ryan note that a potential problem with the assumptions used to model forecasts is that as conditions change there are no available data on reasonable travel behavior for the new conditions 7

PAGE 20

(1980). For example, changes in tastes, life style, and technology produce shifts in travel characteristics (Zahavi and Ryan, 1980). While several researchers in the early 1980s claimed stability of daily travel time expenditures over time, many others have since noted instability. A study of travel time expenditures in the San Francisco Bay Area found an increase per traveler from 1965 to 1981, but a decreased from 1981 to 1990 (Purvis, 1994). Another study of travel in Washington, D.C. also found an increase of daily travel time expenditures from 1968 to 1988 (Levinson and Kumar, 1996). Additionally, this study found that during the time period average trip times declined or remained stable in Washington, D.C. indicating that trip frequency increased (Levinson and Kumar, 1994). Although some researchers have attempted to validate the concept of travel time budgets at a highly aggregated level, other researchers consider travel time expenditures at disaggregate levels in an attempt to better understand travel demand. This thesis will not attempt to validate the concept of a travel time budget, but instead explore the relationships between travel time growth and socio-demographic and area characteristics. The following section discusses research relating travel behavior to many socio-demographic characteristics. 2.2 Socio-demographics Characteristics Researchers have taken different approaches to relate socio-demographic and household characteristics to daily travel expenditures. These characteristics are used as surrogates in an attempt to capture underlying differences in the population (Principio and Pas, 1997). Socio-demographic variables that have been found significant in past studies include age, gender, auto ownership, employment status, household size, income 8

PAGE 21

and life style group (Mokhtarian and Chen, 2002). The following sections will discuss the findings by various researchers in each of these areas. In 1999, Patricia Mokhtarian and Cynthia Chen at the Institute of Transportation Studies, University of California, Davis completed a thorough review of literature relating to travel time. Many of the studies and findings discussed in the remainder of this section are reviewed in depth in their report, A Review and Discussion of the Literature on Travel Time and Money Expenditures. It is important to consider differences in methodology and scope when comparing results of various surveys. Modes included in the research, survey period, survey type, analysis unit such as per traveler or per person, and trip types included can all play a significant role in the outcome of a study (Mokhtarian and Chen, 1999). 2.2.1 Age Many studies have found the influence of age to have a significant relationship with daily travel time expenditures (Mokhtarian and Chen, 1999). Generally persons in middle age groups travel more than older persons or young children. Several studies support this finding. Adults between 18 and 50 years of age were found to travel significantly more than persons over 50 years of age (Kitamura et al., 1992; cited by Mokhtarian and Chen, 1999). Another study found that persons between 21 and 64 years of age spent more time traveling than the average (Prendergast and Williams, 1981; cited by Mokhtarian and Chen, 1999). Gunn concluded that persons between 17 and 24 years of age traveled the most, while persons less than 16 and over 60 years of age travel significantly less than the average (1981). 9

PAGE 22

2.2.2 Gender Some studies have found gender to be a significant factor related to daily travel time expenditures, while others have found the relationship to be insignificant. Many researchers have found that on average women spend less time traveling than men (Prendergast and Williams, 1981; Gunn, 1981; Kitamura et al., 1992; Levinson and Kumar, 1995; Robinson, 1999; cited by Mokhtarian and Chen, 1999). Other researchers have found the opposite relationship (Lu and Pas, 1999) or the relationship to be insignificant (Roth and Zahavi, 1981; Zahavi and Talvitie, 1980; cited by Mokhtarian and Chen, 1999). 2.2.3 Employment Status Supernak found that unemployed persons on average spend less time traveling than employed persons (1982). His study concentrated on Baltimore, Maryland and the Twin Cities of Minneapolis-St. Paul, Minnesota and included all travel modes. In 1977, average travel time for employed persons in Baltimore was 65 minutes, while unemployed persons traveled 35 minutes per day (Supernak, 1982). This result, that employed individuals spend more time traveling that unemployed persons, has also been supported by many other studies (Mokhtarian and Chen, 1999). With advancements in technology, more persons are able to work from home. Individuals that work from home make more trips than non-workers and also spend more time at home (Levinson and Kumar, 1996). Workers that must commute to and from work tend to spend an additional 30 minutes per day traveling than at home workers (Levinson and Kumar, 1996). 10

PAGE 23

It is important to note the interactions between gender and employment status. One study found the combination of employment status and gender significantly increased the range of average travel time expenditures, with employed males having the highest average and retired females traveling, on average, a third of the average for employed males (Prendergast and Williams, 1981; cited by Mokhtarian and Chen, 1999). 2.2.4 Household Size The relationship between household size and daily travel time expenditures is not agreed upon by researchers. Some studies have found a positive relationship, while other studies, some by the same researchers, found a negative relationship (Purvis, 1994; Roth and Zahavi, 1981; cited by Mokhtarian and Chen, 1999). 2.2.5 Income The relationship between income and travel time expenditures is also not agreed upon by researchers. Some studies have found that as income increases travel time expenditures increase (Prendergast and Williams, 1981; Tanner, 1981; Zahavi and Talvitie, 1990; Roth and Zahavi, 1981; Lu and Pas, 1999; cited by Mokhtarian and Chen, 1999). Zahavi and his colleagues also found a negative relationship and independent relationships when studying travel times in different cities (Mokhtarian and Chen, 1999). The inclusion of different travel modes may influence the results of daily travel time expenditures by income (Mokhtarian and Chen, 1999). 2.2.6 Private Vehicle Ownership The influence of the availability of a private vehicle is also not agreed upon by researchers. Similar to income, different travel modes used for analysis may influence 11

PAGE 24

the results (Mokhtarian and Chen, 1999). Also private vehicle ownership may decrease travel time expenditures by other modes, including public transportation, bike, and walking (Golob, 1990; cited by Mokhtarian and Chen, 1999). Many studies have found a positive relationship between private vehicle ownership and travel time (van der Horn, 1979; Prendergast and Williams, 1981; Roth and Zahavi, 1981; Purvis, 1994; Lu and Pas, 1999; cited by Mokhtarian and Chen, 1999), while other studies found a negative relationship (Zahavi and Talvitie, 1980; Roth and Zahavi, 1981; Robinson et al., 1972; cited by Mokhtarian and Chen, 1999). Vehicle availability has also been found insignificant (Downes and Morrell, 1981; Purvis, 1994; cited by Mokhtarian and Chen, 1999). 2.2.7 Life Style Group The life style concept was used by Salomon to describe travel behavior, but he excluded time-use data from his analysis (1983; cited in Principio and Pas, 1997). Salomon primarily used demographic characteristics to determine life style groups. Principio and Pas, however, grouped persons into seven life style groups based upon cluster analysis of time-use patterns (1997; citied in Mokhtarian and Chen, 1999). Their study used data from the 1994/1995 Triangle Travel Behavior Survey, which was conducted in Northern California (Principio and Pas, 1997). Workaholics spend an average of 85 percent of time working or devoted to work-related activities. This group also spends the least amount of time on recreation, maintenance, and social activities. Workaholics tend to make less trips and coordinate their trips more efficiently than the average of the sample. Workaholics and the second life style group, Active Workers have similar socio-demographic 12

PAGE 25

characteristics, but different time use patterns. Active Workers spend an average of 63 percent of time working, but also participate in other activities. Active Workers had the highest total travel time expenditures, because of the high number of trips and tours. Socializers, however, made the fewest number of trips and were not efficient at trip linking. This group spent little time working or attending school, but on average devoted 59 percent of their time socializing. Leisure Enthusiasts tend to spend a great deal of time on recreational activities, while Domestic Caretakers spent most time maintaining their household. Both of these groups made few trips. Leisure Enthusiasts, which are primarily retired individuals, spent the least amount of time traveling. Diverse Participants spent time on a variety of activities. The final group, Scholars spent the majority of time on school and school-related activities. Scholars on average only spent 13.7 percent of their time working. Principio and Pas concluded that travel behavior patterns varied significantly across the seven groups (1997). This variation may not have been identified if socio-demographic characteristics alone were used because many of the groups had similar characteristics (Principio and Pas, 1997). 2.3 Area Characteristics The relationships between area characteristics, land use, the transportation system, and travel time expenditures have also been discussed by many researchers. The manner in which areas are divided varies between studies. Gordon et al. argued that the reason people living in large cities had higher travel time expenditures than those in smaller cities was the spatial structure and not population density (1989). According to this argument, the relationship between population density 13

PAGE 26

and travel time expenditure is unclear if spatial structure is ignored. The authors stated, In a monocentric city, high densities imply shorter trips, and low densities mean longer trips. In a polycentric city, low densities could mean either shorter or longer trips depending upon whether workers choose homes around employment subcenters or whether cross-commuting across metropolitan areas is common (Gordon et al., 1989). The impact of spatial structure may be why other studies have concluded that higher densities do not imply higher travel times. For example, one study considered travel times in the inner, middle, and outer area of Reading, Britain and found that area type made no difference in daily travel time per person (Downes and Morrell, 1981). Also Supernak found that while urban travel times in Baltimore, Maryland were higher than suburban travel times, the opposite was true in the Twin Cities, Minnesota (1982). Another study that considered the effect of commute time on employment in New York found that central city and urban residents reported significantly different commute times in the 1995 Nationwide Personal Transportation Survey (NPTS) than their suburban counterparts (Macek et al., 2001). Private vehicle commute times for both urban and suburban residents averaged about 24 minutes. Transit commute times, however, ranged from about 40 minutes for urban transit commuters to about 60 minutes for suburban transit commuters (Macek et al., 2001). Several studies have found that commute times are higher in large cities. Van der Hoorn concluded that travel time per person per week was the highest for dense urban areas for all purposes except school trips (1979). Therefore, total travel time expenditures were the greatest for dense urban areas. This result was also found by Landrock who studied travel times in British and London metropolitan areas (1981). 14

PAGE 27

Gordon also found that commute times were higher in large cities in the United States (1991). Although Gordon argued that the relationship between population density and travel time is ambiguous if spatial structure is not considered, several researchers have considered the attributes of an area, such as population density and size, for classification purposes. The study by Landrock mentioned above also considered the effects of population size and population density on daily travel time expenditures per person in Britain (1981). For population size, he found that all areas except London fell between 56 minutes and 60 minutes for all persons and between 72 and 76 minutes for travelers. Average daily travel times in London were longer than in other areas with an average of 68 minutes for all persons and 88 minutes for travelers only (Landrock, 1981; cited by Mokhtarian and Chen, 1999). The higher travel times in London resulted mainly from large amounts of time spent on work, shopping, and social activities (Landrock, 1981; cited by Mokhtarian and Chen, 1999). Regarding population density, Landrock found that persons living in low densities had a lower daily travel time than those living in higher densities (1981). According to Mokhtarian and Chen, the effect of population density on travel time expenditures seems to be significant and non-linear, while the interactive effect of population size and density seems to be insignificant except for persons living in low density large populations, who tended to have higher travel times compared to other areas (1999). Many other measures are related to daily travel time expenditures including commute times, vehicle miles of travel (VMT), mode share, and distance traveled. Researchers have studied how various spatial designs affect these measurements 15

PAGE 28

(Cervero, 1996; Erwing et al., 1994; Frank and Pivo, 1994; Handy, 1996). Many of these studies have considered smart growth or neo-traditional neighborhoods. These neighborhoods are designed to have mixed land uses for residential, commercial, and recreational activities. Streets are inter-connected and designed to encourage walking and bicycling (Mokhtarian and Chen, 1999). Results that show fewer automobile trips in neo-traditional neighborhoods imply lower total daily travel times by automobile, but when transit, walk, and other non-motorized modes are included, total travel time may be greater than traditional suburban neighborhoods. One study found that residents of mixed land use or neo-traditional neighborhoods in Seattle traveled 28 percent fewer miles than residents in adjacent areas and up to 120 percent fewer miles than residents in suburban areas (McCormack et al. 2001). This trend occurred regardless of socio-demographic characteristics. When total travel time expenditures were considered, differences between various area types were not apparent. According to this study, daily travel time was about 90 minutes per person regardless of location or socio-demographic characteristics (McCormack et al. 2001). A synthesis of existing literature on land use-travel relationships has been completed by Ewing and Cervero (2001). Studies focus on a variety of measures, including trip frequencies, trip lengths, mode choice or modal split, person miles traveled (PMT), vehicle miles traveled (VMT), and vehicle hours traveled (VHT). Each study varies in methodology, survey area, and results. Some of the results indicated that transit trip rates increased with densities. VMT is lower in households in higher densities. Higher densities induce more walk and bike trips, and VHT is lower in areas with high household density and high employment density (Ewing and Cervero, 2001). 16

PAGE 29

One study has linked travel behavior and activity time allocation to roadway capacity. This study found that increases in highway capacity cause significant, but small changes in travel behavior (Levinson and Kanchi, 2002). Vehicle miles of travel increased, but the time spent traveling remained fairly stable from 1990 to 1995 (Levinson and Kanchi, 2002). For the purpose of this thesis, limited attention will be given to area characteristics, because of limitations of the dataset used for analysis. 2.4 Activity Time Allocation While travel time expenditures seem to be related to socio-demographic characteristics and land use, the inclusion of activity participation can provide additional insight into travel behavior (Lu and Pas, 1999). The relationship between travel and activities was first discussed by Evans in 1972 (Chen and Mokhtarian, 2002). He noted that travel time is not independent of the time spent participating in an activity, because a minimum amount of travel may be required to access an activity (Evans, 1972; cited by Chen and Mokhtarian, 2002). The allocation of time has been considered by several disciplines. Sociologist have examined changes in time at work and leisure, planners have considered time by activity and location, and economists have proposed theories that households combine time and market goods to produce commodities (Levinson and Kumar, 1996). Levinson and Kumar state that continuous tradeoffs among activities and between household members enable adaptation to changes in technology and socioeconomic characteristics (1996). 17

PAGE 30

Activities are grouped into various categories, including mandatory, maintenance, subsistence, and discretionary (Chen and Mokhtarian, 2002). Mandatory activities include work or other paid activities. Maintenance and subsistence activities include medical activities, eating, and sleeping. Discretionary activities are all social or recreational activities (Chen and Mokhtarian, 2002). Many studies have considered activity duration and travel time (Chen and Mokhtarian, 2002). Generally, activity duration and travel time are positively related or as activity duration increases the time spent on travel may also increase (Hamed and Mannering, 1993; Kitamura et al., 1997; cited in Chen and Mokhtarian, 2002). As travel time increases, the total time available for participation in other activities decreases (Mokhtarian and Chen, 1999). The frequency of participation in an activity also has a positive effect on travel time expenditures (Chen and Mokhtarian, 2002). In 1992, Kitamura et al. studied the relationship between work and travel time (cited by Mokhtarian and Chen, 1999). As work duration increases, time used to travel for non-work travel decreases. If commute time was reduced by 10 minutes, in-home activity duration would increase 7.1 minutes, out-of-home activities would increase by 1.9 minutes, and travel time would only increase by 0.4 minutes (Kitamura et al., 1997; cited in Mokhtarian and Chen, 2002). Golob and McNally found that persons traveled 2.8 minutes for one hour of work, 7.8 minutes for one hour of maintenance activity, and men traveled 5.5 minutes for one hour of discretionary activity while women traveled 8.5 minutes (1997; cited by Mokhtarian and Chen, 1999). Lu and Pas considered the interactions of in-home and out-of-home activity durations (1999). They found that as time spent at work or on out-of-home subsistence 18

PAGE 31

activities decreases in-home activity duration increases. Also out-of-home activities can substitute time spent on the same type of in-home activities (Lu and Pas, 1999). Activity duration is an integral component of activity-based travel demand models with a temporal aspect (Pas, 2002). Understanding activity duration and start time can provide insight into an individuals travel behavior. Because of the importance of activity participation in determining most travel demand, it has been integrated into many transportation planning models. In recent years, many researchers have developed a variety of models of time use (Pas, 2002). Activity-based models include structural equations models, utility maximizing models, and hazard-based models. Some models focus on time devoted to activities over a day or week, but others concentrate on activity duration (Pas, 2002). Activity based models must account for changes in household relationships and person constraints. Modeling constraints in time, space, and society has received attention in activity based models. Pendyala has used regression models to model time use and travel behavior in time and space (2003). The activity-based approach is founded on the premise that travel demand is a result of an individuals need or desire to access various locations to participate in various activities (Pas, 2002). Hence, travel is often considered to be a derived demand. Work by Mokhtarian, Salomon, and Redmond does discuss that some travel demand is not derived and has an intrinsic positive utility and is valued for its own sake (2001). Although this thesis focuses on total travel time expenditures and does not develop an activity-based model of transportation demand, it is important to mention this application of time-use studies. 19

PAGE 32

2.5 International Trends in Travel Time Comparisons of travel time expenditures among various countries can expose general trends and possibly provide a better understanding of contributing factors to travel demand (Schafer, 2000). As mentioned above, survey methodology and scope vary between surveys, significantly impacting the comparability of data. In 1982, Supernak compared average daily travel times for several cities around the world. He states that his findings do not support the concept of travel time budgets, because of the wide range of travel times even at a very aggregate level (Supernak, 1982). For example, the average travel time was 46 minutes in Britain, 73 minutes in Washington, DC, 125 minutes in Belgium, 79 minutes in Singapore, 94 minutes in Colombia, 173 minutes in Peru, and 186 minutes in Nigeria (Supernak, 1982). While Supernak argued that the variation of travel times in various cities did not support stability, Robinson, Converse, and Szalai concluded that average travel time expenditures per person in twelve countries were relatively stable, even though travel times varied from 39 minutes per person in Germany to 90 minutes in Lima-Callao, Peru (1972). In 2000, Schafer studied travel time expenditures for 10 countries. He concluded that travel time budgets exhibited rough regularities across space and time for the countries included in his report. Most of the analysis was completed using national travel surveys (Schafer, 2000). The average travel time per capita per day was found to be 1.09 hours or 65.4 minutes with a 9.6 minute standard deviation (Schafer, 2000). Schafer notes that although he observed overall stability, daily travel time expenditures varied between individual countries (2000). Of the countries studied, Switzerland exhibited the 20

PAGE 33

highest average daily travel time expenditures. Schafer also mentions that the Netherlands and the United States have had increasing trip rates and a gradual increase in daily travel times per person (2000). Researchers in the Netherlands have searched for explanations for increases in travel for the Dutch population (Wee et al., 2002). Based on one survey, travel time increased 4.9 minutes or about 8 percent between 1979 and 1998, while another time use survey indicated a 15 minute per day or 26 percent increase between 1975 and 2000. According to Wee, Rietveld, and Meurs, differences in time period and survey methodology are likely to explain variation of the increases (2002). Possible reasons for the increase in travel time expenditures for the Dutch population include increases in the utility of travel (Wee et al., 2002). The utility of travel may have increased as a result of changes in spatial trends, for example increases in the scale of services that are available, changes in job locations, and increases in the size of cities (Wee et al., 2002). Specialization of the labor force and jobs, which require training and education, may also impact the utility of travel. For example, a person may now have to search a larger area to find possible positions (Wee et al., 2002). A change in the preference for housing has also increased the search area for a home, which can increase the travel time required for work trips and visiting friends or relatives (Wee et al., 2002). Also individuals tend to participate in more leisure activities because of increases in income and may even travel for the fun of it (Mokhtarian and Salomon, 1999; cited in Wee et al., 2002). Changes in the economy, including the shift to a service orientated society also play a role in changing the utility of travel (Wee et al., 2002). 21

PAGE 34

Wee, Rietveld, and Meurs also report that changes in the cost of travel are related to the increase in travel time expenditures. The share of car kilometers, similar to vehicle miles of travel (VMT), has increased resulting in a decrease in the average cost per road length and increases in safety rate measures (Wee et al., 2002). Although improvements to the roadways have not been as significant as in the past, car ownership and use has increased which result in increased congestion. Wee, Rietveld, and Meurs state that because the travel times increase slowly and in a rather smooth way people might get used to the increase and accept them instead of changing their destination (2002). Increase in travel time for some origins and destinations may seem to conflict with the general trend for increasing speed. However, despite increases in congestion the average speed on highways or motorways is much higher than other road classes and the share of travel on high speed facilities has increased (Wee et al., 2002). The researchers also add that the decrease in persons using bicycles, increases in the level of comfort of cars, improved roadway safety, and the increased ability to combine other activities with travel has decreased the disutility of travel (Wee et al., 2002). Many of these explanations for the increase of travel time expenditures in the Netherlands are applicable to the travel demand growth in the United States. The following chapter will explore changes in the United States that influence travel behavior trends and the allocation time. 22

PAGE 35

CHAPTER 3 SETTING THE STAGE Over the last several decades, many significant social changes have occurred in the United States that may influence travel demand and daily travel time expenditures. According to Robinson, the way Americans viewed the world changed following World War II: Baby Boomers were the Americans most affected by the period of optimism that followed World War II. That period saw the gradual development of the idea that progress and the increases in the standard of living were infinite. As the United States rose to the position of the worlds greatest economic power after the war, Americans began to believe in infinitely expanding economic opportunity. Consumption of goods came to be thought of in open-ended terms that is, with no ceiling on what could be purchased and owned. Mobility was infinitely upward (Robinson, 1999, pp. 44). This chapter will explore many of the trends that may influence travel patterns, including the aging population, decreases in household size, increases in labor force participation for women, increases in household income, and increases of vehicle availability to contextualize the growth of travel time expenditures in the United States. Changes in daily activity time allocation will also be discussed. 23

PAGE 36

3.1 Aging Population The median age of the United States population has increased steadily as the Baby Boom generation ages. The Baby Boom generation is defined as persons born in the United States between 1946 and 1964. According to the U.S. Census Bureau, the median age increased from 28.1 in 1970 to 35.3 in the year 2000. This trend is shown in Figure 3.1. Figure 3.2 shows the age distribution of the United States population in 1970 and 2000. The projected age distribution for 2020 is also presented in this figure. In 2000, almost 30 percent of the population was between 36 and 54 years of age, with an additional 12 percent above retirement age. Going forward, higher shares of the population will be in the older age groups, which historically have had lower travel levels because of the reduction of work related responsibilities. Travel time expenditures by age group will be discussed in the following chapter. As the Baby Boom generation approaches retirement, significant changes in travel behavior are likely to occur. In the year 2011, the first of the Baby Boom generation will turn 65 years old. Between 2010 and 2030, the population over 65 years of age will increase at a rate four times that of the overall population. Robinson suggests that as Americans get older, the pace of life is likely to decrease Life may get slower, even for young people, because the desires of the older, slower-paced people are likely to be considered first, as the society reshapes itself to meet their needs (1999). However, since this group is use to higher levels of mobility than the elderly populations of the past, they are likely to continue traveling more than historically was the case. The aging population raises many unanswered questions relating to mobility and travel behavior in the future. 24

PAGE 37

Figure 3.1 Median Age of the United States Population, 1970 to 2000 30.0 32.9 35.3 28.10102030401970198019902000Median Age (Years) Source: U.S. Census Bureau (November 2002) Demographic Trends in the 20th Century. Figure 3.2 Age Distribution of United States Population 0%2%4%6%8%10%12%Under 5 5-9 10-1415-1920-2425-2930-3435-3940-4444-4950-5455-5960-6465-6970-7475-7980-8485+Share of Population (%) 1970 2000 2020 Source: U.S. Census Bureau (November 2002) Demographic Trends in the 20th Century. 25

PAGE 38

3.7 Household Size The make up of American households has changed dramatically since the early part of the century. According to a report by the U.S. Census Bureau, contributing factors include increased mobility, affordable housing, women having fewer children, and an increase in diversity of the U.S. population. As Figure 3.3 shows, the average household size has been decreasing over time. Since 1900, household size has decreased by an average of two persons per household. The rate of change in household size is slowing. The change between 1990 and 2000 is the lowest decade change of the century. Figure 3.4 shows that the distribution of households by size has also changed considerably. The growth of the number of households between 1990 and 2000 was mainly due to an increase in the number of one or two person households. One factor contributing to the decrease in household size is families having fewer children. As shown in Figure 3.5, the average number of children under 18 years of age per household has decreased since the 1970s, but has remained relatively stable during the last decade. The presence of children in a household implies a significant time commitment and can have a considerable impact on travel behavior. Similar to other patterns, many families, especially single parents and dual-career families, now look outside the home for services, including day care (Levinson and Kumar, 1996). Marital status can also be related to household size. The share of married persons 15 years old and over has decreased since the 1960s, while the share of divorced persons has increased. Figure 3.6 presents this trend. Robinson notes that role factors, which include parenthood, marital status, and work hours, are the most important factors affecting the amount of free time that people have available (1999). 26

PAGE 39

Figure 3.3 Average Household Size, 1930 to 2000 4.01 3.38 3.29 3.11 2.75 2.59 2.63 3.6801234519301940195019601970198019902000Household Size Source: U.S. Census Bureau (November 2002) Demographic Trends in the 20th Century. Figure 3.4 Distributions of Households by Size, 1940 to 2000 7.79.313.317.622.724.625.8 24.8 28.1 28.0 29.6 31.3 32.0 32.622.422.818.917.217.417.416.518.118.417.215.415.415.114.227.021.422.520.213.211.010.80%20%40%60%80%100%1940195019601970198019902000Share of Households 1-person 2-person 3-person 4-person 5-or-more Source: U.S. Census Bureau (November 2002) Demographic Trends in the 20th Century. 27

PAGE 40

Figure 3.5 Average Number of Own Children under 18 per Family 0.00.30.60.91.21.519701972197419761978198019821984198619881990199219941996199820002002Average Number of Children Per Family Source: U.S. Census Bureau (June 2003) Table FM-3. Figure 3.6 Share of Population by Marital Status, 1950 to 2000 67.569.366.863.260.757.9 26.4 25.3 28.1 29.6 29.9 31.36.88.30%20%40%60%80%100%195019601970198019902000Share of Persons Over 15 Married Never Married Widowed Divorced Source: U.S. Census Bureau (June 2003) Table MS-1. 28

PAGE 41

3.3 Labor Force Participation The labor force participation rates for men and women are displayed in Figure 3.7. The participation rate for women has increased since World War II, while the rate for males has decreased. Over the past decade rates for both men and women have remained relatively constant which may suggest that the labor force participation rates have reached equilibrium. For employed persons, the commute trip is often the most substantial and regular travel time expenditures. Relationships between worker status, gender, and daily travel time expenditures are discussed in Chapter 5 of this thesis. The increase in the number of workers has resulted in increases in per capita income and increased mobility (Levinson and Kumar, 1996). Increased work responsibilities, especially for women, have resulted in dramatic time allocation shifts. These changes will be discusses in the last section of this chapter. Figure 3.7 Civilian Labor Force Participation Rate, 1948 to 2002 0204060801001948195419601966197219781984199019962002Rate Population Male Female Source: US Department of Labor, Bureau of Labor Statistics, Current Population Survey. 29

PAGE 42

3.3.1 Labor Force Specialization As mentioned in Chapter 2, the labor force has become increasingly specialized as the market in the United States has changed. One example of this is the medical profession. Historically, patients may have visited only their family doctor for treatment of a variety of conditions. In todays society, however, patients will be frequently referred to specialist for treatment. Many other fields have also become more specialized, including technology industries and service industries. Increased specialization may result in more trips or longer trip distances as individuals must travel to various locations to access services or activities. Spatial trends, which have resulted from suburbanization and sprawl, directly impact the location of services and job location. Individuals may have to travel farther to access a work location that matches their skill set or specific retail store. The size of services and retail centers has also increased and require larger service areas for support. School, churches, and many retail centers are examples of the increase in scale. Historically, a small local school or church may have met the needs of a community, but today mega-churches and schools with several thousand students are increasingly common. Large retail chains attract shoppers from a large area with their large selection and low prices. Specialized services often replace activities that previously occurred at home. Levinson and Kumar state to get the most out of every day, individuals try to substitute money for time (1996). This substitution is evident by the increase in the number of meals eaten away from home each year. The share of spending dedicated to food eaten away from home is used as a surrogate measure to the number of meals in Figure 3.8. The share of food expenditures spent on food away from home increased from 26 percent 30

PAGE 43

in 1970 to 39 percent in 1996 (Lin et al., 1999) The affordability and convenience of many restaurants and reduced household size make eating out a very competitive alternative to cooking at home for many indivi duals. According to several studies, nonwork trips rates have increased to over half of the total number of daily trips by adults (Gordon, Kumar, and Richardson, 1988; cited in Levinson and Kumar, 1996). This increase in trip rate is a significant reason for the increase in total daily travel time expenditure. Trips rates and trip pur pose will be discussed in Chapter 5. Figure 3.8 Share of Total Food Expenditures Spent on Meals Away from Home 39.9% 42.2% 41.9% 41.4% 42.0% 42.3% 38.8% 0% 10% 20% 30% 40% 50% 1996199719981999200020012002Share of Total Food Expenditures Source: US Department of Labor, Bure au of Labor Statistics, Consumer Expenditure Sur vey, 1996-2002. Specialized fields often require advanced training. The share of the population with higher educational training has increased significantly over the last several decades. Figure 3.9 provides that share of the population by educat ion level. Generally, as education increases personal income increases. The following section will discuss trends in income. 31

PAGE 44

Figure 3.9 Educational Attainment of the Population 15 Years and Older 60.150.834.628.321.5 24.6 26.6 35.9 28.7 31.05.919.718.76.95.78.411.614.85.25.75.97.19.57.15.70%20%40%60%80%100%19601970198019902000Share of Persons Over 15 Less than High School High School Graduate Some College or Training Associate Degree Bachelor Degree Graduate or Professional Degree Source: U.S. Census Bureau. 32 Specialization has not only evolved in the education system, but in many other activities as well. As Robinson puts it, There is always one more thing to be done to enhance our resume Even the play-time activities of children have become open-ended. Throwing a Frisbee has become so specialized that some people make it a living (1999). Social networks in the Untied States have changed dramatically. Instead of interacting with persons in your neighborhood, individuals now have networks based on personal interest or professional associations. In todays society, many children must be chauffeured to interact with friends often at organized activities, such as sport teams, instead of riding a bicycle to play with friends down the street in an unorganized baseball game. Many factors have influenced this change in society, including safety concerns for children and changes in household demographic, such as smaller household size. These changes have had a significant effect on the transportation system, especially since most trips occur by private vehicle, and travel time expenditures.

PAGE 45

3.4 Income Increased specialization and higher education of the U.S. population have partially contributed to increases in annual income per capita. Figure 3.10 provides the median income of the U.S. population from 1980 to 2002. Real income has slightly increased since 1980 with a slight decline in the early 1990s. Growth in real income is one relevant factor relating to increases in travel demand. Income not only enables individuals to afford travel, but also generates some demand to participate in activities, such as entertainment or shopping (Polzin et al., 2003). Continued growth of real incomes may continue to drive the desire to participate in additional away-from-home activities that require travel. In addition to real incomes, the cost of travel, especially by private vehicle, determines the affordability of travel (Polzin et al., 2003). The cost of vehicle travel consists of high fixed costs such as the cost of a vehicle and insurance and relatively low marginal costs. For example, the Internal Revenue Service assumes full costs per VMT of approximately 36 cents and marginal costs of 12 cents. The overall cost per vehicle mile of travel is shown in addition to the income per capita trend in Figure 3.11. This cost data is developed by combining data from the national travel surveys (NPTS 1983, 1990, 1995, and NHTS 2001) with information from the Bureau of Labor Statistics on household spending. From this figure, increases of income are apparent, while the cost of travel has remained relatively stable over the past decade. Income is important to travel demand to the extent that it enables vehicle ownership. The following section will discuss vehicle availability trends in the United States. 33

PAGE 46

Figure 3.10 Median Income in the United States, 1980 to 2002 05,00010,00015,00020,00025,00030,00035,00040,00045,000198019821984198619881990199219941996199820002002Median Income (2002 CPI adjusted dollars) Source: U.S. Census Bureau, Income in the United States: 2002. Figure 3.11 Income and Travel Cost per Vehicle Mile of Travel $0.59 $0.36 $0.31 $0.34 $15,288 $14,983 $13,468 $17,270$0.00$0.10$0.20$0.30$0.40$0.50$0.60$0.70$0.80$0.90$1.001984198619881990199219941996199820002002POV Expenditure/VMT (cents per mile, 2001 $) $0$2,000$4,000$6,000$8,000$10,000$12,000$14,000$16,000$18,000$20,000Income per Person (2001 dollars) POV Expenditure per VMT (cents per mile) Income per Person (2001 Dollars) Source: Bureau of Labor Force Statistics and NPTS/NHTS, cited in Polzin et al., 2003. 34

PAGE 47

3.5 Vehicle Availability and Use The growth in personal income and the relatively low cost of vehicle travel has partially resulted in an increase of vehicle ownership. The number of registered vehicles in the United States has more than doubled since 1970. Figure 3.12 provides the trend of motor vehicle registration in the United States from 1970 to 2002. The trend of the ratio of vehicles to workers, drivers, and person over 16 is provided in Figure 3.13. In 2001, the ratio of vehicles to workers and to drivers was over one while the ratio to adults is approaching one. This would indicate that each adult, driver and worker has at least one vehicle available to him or her. However, the distribution of vehicles across the population is not even, with some households having no available vehicle. Figure 3.14 shows the trend in zero car household shares. Some share of these are zero-car households by choice or due to medical/physical and not financial reasons. While the share of zero-car households has declined, the number of such households has only declined modestly from 11.4 million in 1969 to 10.9 million in 2001. The possession of a drivers license also contributes to the mobility rates of the population. In the past, it was less common for women to have a drivers license. However, in todays society this is a distant memory as women hold half of all licenses in the United States. The share of the population in 2002 with a drivers license by age group and gender is shown in Figure 3.15. The percent of drivers is the same for both sexes until about the age of 45, when the share of women drivers begins to decrease. As the population ages, it is likely that the share of women drivers per age group will become similar to men for all age groups, because the younger female population will be accustomed to driving and having licenses. 35

PAGE 48

Figure 3.12 Registered Motor Vehicles, 1970 to 2002 05010015020025019701972197419761978198019821984198619881990199219941996199820002002Registered Motor Vehicles (Millions) Source: Federal Highway Administration, Highway Statistics Series. Figure 3.13 Ratios of Vehicles to Adults, Drivers, and Workers 0.00.51.01.519691971197319751977197919811983198519871989199119931995199719992001Ratio Ratio of Vehicles to Persons Over 16 Ratio of Vehicles to Drivers Ratio of Vehicles to Workers Source: Nationwide Personal Transportation Surveys and 2001 National Household Travel Survey 36

PAGE 49

Figure 3.14 Share of Zero-Vehicle Households in the United States 0%5%10%15%20%25%196019651970197519801985199019952000Zero-Vehicle Households (%) Census NPTS/NHTS Source: Nationwide Personal Transportation Surveys, 2001 National Household Travel Survey, and U.S. Census Bureau. Figure 3.15 Share of Population with a Drivers License by Age Group and Gender, 2002 0%20%40%60%80%100%19 and Under20-2425-2930-3435-3940-4445-4950-5455-5960-6465-6970-7475-7980-8485 and OverAge GroupShare of Population Males Females Source: Federal Highway Administration, Highway Statistics Series 2002. 37

PAGE 50

Historically, trip generation models have considered auto ownership to be a significant factor in determining trip production. These models are used to predict trip frequencies according to trip purpose. Other variables commonly used in trip generation modeling include household size, number of workers, household income, and licensed drivers (DKS Associates, 1994). Land use variables, such as population density, are also sometimes included in the models. The two most common approaches of trip generation are cross-classification and regression models. Cross-classification models group persons according to socio-demographic characteristics. Trip rates are calculated for each characteristic or variable based upon observed data (DKS Associates, 1994). Regression models, especially linear regressions, are also used for trip generation analysis, but have difficulties accounting for variables that are not linear (DKS Associates, 1994). For example, trip frequency is not linearly related to vehicle ownership, but increases dramatically with the first vehicle and to a lesser extent with each additional vehicle available (DKS Associates, 1994). As discussed above, the ratio of vehicles to workers, adults, and drivers is an important factor in trip generation and travel time expenditures at the person level. The frequency of trip making is a key variable in determining daily travel time expenditures. Trip frequencies from the 1983 and 2001 datasets will be explored as part of the descriptive analysis. Vehicle availability and licensure trends have coincided with significant increases in vehicle miles of travel (VMT) in the United States. Between 1970 and 2002, annual VMT increased 152 percent according to the Federal Highway Administration. Figure 3.16 provides the VMT travel during this period. The rise in vehicle travel has 38

PAGE 51

corresponded with decreases in travel share by other modes, including walking and transit use. The majority of person trips and daily travel time occur by private vehicle. The share of travel time by mode will be further discussed in Chapter 5. Increases in vehicle travel have resulted in higher levels of congestion. According to the 2002 Urban Mobility Report, the share of daily travel that is congested has increased from 16 percent in 1982 to 34 percent in 2001. In recent years, extensive attention has been given to congestion issues, especially efforts to reduce congestion though traffic and incident management practices. If congestion continues to worsen, it may begin to impact travel behavior. Certain trips may be avoided, destinations altered, or starting times shifted as congestion becomes a mounting concern for the traveling population. Figure 3.16 Annual Vehicle Miles of Travel, 1970 to 2002 0500,0001,000,0001,500,0002,000,0002,500,0003,000,00019701972197419761978198019821984198619881990199219941996199820002002VMT (Millions) Source: Federal Highway Administration, Highway Statistics Series. 39

PAGE 52

3.6 Time Allocation Shifts Many of the social changes discussed above have resulted in significant shifts in time allocation patterns in the United States. Researchers have grouped time uses into four major categories for analysis purposes. These include paid work, household/family care, personal time, and free time (Robinson, 1999). Work time includes paid work completed at home as well as non-work activities at the work location (Robinson, 1999). Household/family time includes housework, child care and shopping (Robinson, 1999). Personal time consists eating, sleeping, grooming, and all other activities required to function effectively in society, while free time consists of leisure activities, such as reading, television, and hobbies, and semi-leisure activities, such as adult education, religion, and other organizational activity (Robinson, 1999). According to Robinson, travel time can be seen either as a separate maintenance activity connecting the four other types of time or as a necessary adjunct to each type (1999). Figure 3.17 portrays trends in average daily time expenditures from 1965 through 2000 based on data from the Americans Use of Time Project for persons 18 to 64 years of age. Time required for travel is incorporated within the activity group for which the travel was required. Time devoted to personal care has decreased an average of 68 minutes per day since 1985; while free time has increased an average of 23 minutes per person per day. Work and commute time has also increased almost 43 minutes per person per day. Figure 3.18 displays the average annual change in minutes per year for each of the major time use categories from 1985 to 2001 and travel time from the travel survey series. In light of these shifts in time use, the increase in daily travel time expenditures does not seem unreasonable. 40

PAGE 53

Figure 3.17 Daily Time Allocation by Major Time Use Category 199 205 189 332 339 187 234 293 283 250 255 634 596 617 652 645 350 362 298010020030040050060070019651970197519801985199019952000Minutes per Day Family Care Work and Commute Personal Care Free Time Source: Americans Use of Time Project, cited in Robinson, 1999 and 1998-2001 Time Diary Studies. Figure 3.18 Annual Average Change in Minutes per Day per Year by Major Time Use Category and Travel Time -1.1 2.7 -3.1 1.9 1.4-4-3-2-10123Minutes per Person per Year Family CareTravel TimeFree TimePersonal CareWork andCommute Source: Americans Use of Time Project, cited in Robinson, 1999; 1998-2001 Time Diary Studies; 1983 NPTS and 2001 NHTS 41

PAGE 54

The daily time allocation trends vary according to gender. Figure 3.19 and Figure 3.20 provide the trend of time allocation by the four major time use categories for males and females from 1965 to 2001. Both genders have had decreases in the amount of time dedicated to personal care, while free time allocation has increased. As more women have entered the work force, time spent at work and commuting has increased for women. Shifts in the proportion of women working explains much of the increase in working hours, rather than an increase in the number of hours spent at work (Robinson, 1999). Time spent on family care activities have also decreased for women, while the time spent on family care has slightly increased for men. The difference between the male and female trends of family care time indicates that overall less time is being spent on housework. A combination of factors have likely contributed to the decline including smaller household size and shifts to purchasing products and services rather than pursuing them at home. According to Robinson, time use patterns are becoming more androgynous (1999). He states, Greater androgyny of time use has occurred not only because females have changed their time-use patterns in ways that more closely resemble male patterns, but also because males have changed their time use in ways that more closely resemble traditional female patterns (Robinson, 1999). Continued review of time use patterns in the future may provide additional insight into the tradeoffs between travel and participation in other activities. The following subsections will briefly discuss methods individuals use to spend their time more efficiently and ways that changes in technology have impacted time allocation and travel behavior. 42

PAGE 55

Figure 3.19 Daily Time Allocation by Major Time Use Category for Males 010020030040050060070019651970197519801985199019952000Minutes per Day Family Care Work and Commute Personal Care Free Time Source: Americans Use of Time Project, cited in Robinson, 1999 and 1998-2001 Time Diary Studies. Figure 3.20 Daily Time Allocation by Major Time Use Category for Females 010020030040050060070019651970197519801985199019952000Minutes per Day Family Care Work and Commute Personal Care Free Time Source: Americans Use of Time Project, cited in Robinson, 1999 and 1998-2001 Time Diary Studies. 43

PAGE 56

3.6.1 Time-Deepening For many Americans, time has become the most precious commodity and the ultimate scarcity (Robinson, 1999). As discussed previously in this chapter, for many people money has become a substitute for time as people purchase services that traditionally were accomplished at home. In an attempt do more with a limited amount of time, people also use time-deepening techniques. These techniques include speeding up activities, substituting leisure activities that require more time for one that requires less, multitasking or doing more than one activity at a time, and planning leisure activities with minimal time tolerances (Robinson, 1999). These techniques are commonly used by most Americans for a variety of activities, such as watching television while eating dinner or purchasing fast food instead of preparing it at home. Time-deepening can also be employed during travel. For example, increases in cell phone use in vehicles can add value to travel time, although this issue raises many safety concerns. As Americas spend more time in their cars, they incorporate other activities into travel. One columnist notes, They put on makeup. They shave with electric razors. They brush their teeth and rinse into little Fred Flintstone spit cups, which I have seen with my own eyes. They eat elaborate meals: salads with creamy dressings, Chinese food and fried rice, burritos with cheese dripping from them, which they balance precariously in their laps or on the dashboard. They type on their laptops in bumper-to-bumper traffic and check their Blackberrys for messages. They chat on their cell-phones endlessly (Cowherd, 2004). While these activities are not ideal matches with driving from a safety perspective, people continue to include these and many other activities into their daily travel. 44

PAGE 57

3.6.2 Technology Technology often facilitates changes in society, including social and economic activities, and alters time use patterns. Technological advances can often be used to do things more quickly and efficiently (Robinson, 1999). Television use is one example of how technology radically changed time use patterns in the United States. Television viewing, including viewing as a secondary activity, consumes more than half of Americas free time (Robinson, 1999). The home computer and the Internet have dramatically altered time use for many Americans. Over the past decade, the use of home computers and the Internet has rapidly increased. One may argue that the increased accessibility to activity opportunities may generate trips. Robinson notes that technology users are more likely to participate in other cultural and leisure activities (1999). The long term effects on time allocation have yet to be realized for the home computer and internet. Also the impact on travel behavior is difficult to quantify. For example, the share of shopping done on-line is steadily increasing, but how this will impact trip making is not completely understood. As discussed above, cellular telephones have also changed the way Americas communicate with each other. Cellular technology allows users to generate trips on the fly and possibly change destinations during travel in ways that historically were not possible. This technology has increased the mobile lifestyle of Americas by allowing users to stay in touch while away from home. Persons who have grow accustomed to this freedom, find it hard to remember or imagine life without a cellular phone. Going forward, cellular applications may be further expanded and have additional impacts on 45

PAGE 58

travel. The wireless internet in one example of such a technology advancement that is already available. Given the trends in daily time allocation, continued growth in a daily travel time may be possible if individuals continue to make tradeoffs among activities or find new ways of multitasking. The following chapter will explore daily travel time expenditures in 1983 and 2001 by various socio-demographic and area characteristics. 46

PAGE 59

CHAPTER 4 DESCRIPTION OF DATA The intent of this chapter is to provide an overview of the data used to explore trends in travel time expenditures. The next section gives a brief description of the history of national travel surveys in the United States. This is followed by a more detailed description of the 1983 Nationwide Personal Transportation Survey (NPTS) and the 2001 National Household Travel Survey (NHTS), including an overview of the survey samples. The comparability of these surveys is also discussed. 4.1 Survey Description The two data sets used in this analysis are part of a series of national travel surveys in the United States. In 1969, the U.S. Department of Transportation conducted the first in a series of Nationwide Personal Transportation Surveys (NPTS). Additional surveys were conducted in 1977, 1983, 1990 and 1995. The purpose of these surveys was to collect up-to-date information on national travel patterns, including daily and long distance trips. In 2001, the National Household Travel Survey continued the series of household travel surveys. The NHTS was a combination of the NPTS survey and the American Travel Survey (ATS). The scope and methodology of the surveys have changed significantly over time. The 1969 survey focused primarily on auto travel, while the 1977 survey was expanded 47

PAGE 60

to all vehicle travel. More recent surveys, especially the 2001 NHTS, were designed to collect more information on non-motorized travel. These surveys are commonly used to track travel patterns and social characteristics that contribute to travel demand in the United States over time. The series provides detailed information on person travel and is the respected source for national level information on traveler demographics and travel characteristics. Through exploration of these relationships, planners and decision makers can more efficiently improve the nations transportation system. 4.1.1 1983 Nationwide Personal Transportation Survey The 1983 NPTS was the third national travel survey in the series sponsored by agencies of the U.S. Department of Transportation, including the Federal Highway Administration, National Highway Traffic Safety Administration, the Urban Mass Transportation Administration (now the Federal Transit Administration), and the Office of the Secretary of Transportation. The survey includes data on household and person level characteristics, motor vehicles, and person trips. The survey was designed to collect data on travel period and travel day trips occurring on the same day. A total of 6,438 households were interviewed between February 1983 and January 1984. Households were selected from the expired Current Population Survey (CPS). The CPS was designed to measure the rate of change for unemployment using a stratified multi-stage cluster sample. Originally 7,900 households were selected for the NPTS based on the national probability sample, but 1,000 of these were not eligible for the survey. An additional 450 households were not available to complete the survey, because 48

PAGE 61

the household members could not be reached by phone, refused to participate, or were unavailable for some other reason. Interviews were conducted by Bureau of the Census professional field staff from twelve Regional Offices located throughout the United States. Interviews were conducted at the home of the participants within four days of their designated travel day. Similar to the two proceeding surveys, travel day data was recalled from the respondents memory. Follow up interviews were conducted by phone for any household member absent at the time of the in-home visit. All persons over the age of fourteen were asked to report all trips taken during the 24 hour travel day and all trips of 75 miles or more one way taken during the 14 day travel period. A member of the household over the age of fourteen was asked to report all trips, except for bicycle and walk trips, taken by household members between 5 and 13 years of age. Information was collected on the purpose, mode, trip length, day-of-week, time-of-day, vehicle used, and vehicle occupancy of trips. This data is available in five separate files, including the household, person, vehicle, travel day trip, and travel period trip file. In this study, the household, person and travel day files are used for analysis. The following is a brief description of each file. The household file contains information on household characteristics including demographic, economic, and area variables. Each household is identified by an identification number. Each record in the data file represents a different household. There are 6,438 households in the household file. The person file includes information about the person characteristics for each person in the sample. These characteristics include age, gender, employment 49

PAGE 62

status, etc. The household identification number and person identification number uniquely identify each person. There are 17,382 person records in the person file; of these, 16,058 persons over five are included in the data file. Each record describes an individual person. The travel day trip file consists of details about the characteristics of the person trip. Each record in the file represents a trip, which includes length, mode, and purpose. There are 45,155 records in the trip file. Examination of travel behavior was completed at the person level. The household and person identification numbers are used to link the information in the data files. Household attributes were merged into the person file according to the household identification number. Trip characteristics, such as total daily travel time, were aggregated and merged into the person file using the household and person identification numbers. This combined data file was then explored to better understand relationships between socio-demographic characteristics and travel time expenditures. Person weights provided in the person file are used to obtain national estimates. Likewise, weights in the travel day file can be used to estimate the number of annual trips and miles of travel by mode at the national level. Table 4.1 provides an overview of the survey sample and weighted population for the 1983 NPTS. 50

PAGE 63

Table 4.1 Sample Size and Weighted Population for 1983 NPTS Sample Size Weighted Population Total Households 6,438 85,371,411 Total Interviewed Persons 17,382 229,702,620 Total Interviewed Persons 5 and older 16,058 213,190,888 Number of Adults (18+) 12,393 168,098,008 Average Household Size 2.7 2.7 People (5+) Who Did Not Travel 4,474 58,812,307 Share of Non-Travelers 27.9% 27.6% Number of Licensed Drivers 10,534 143,280,048 Number of Workers 7,576 103,290,374 Number of Vehicles 10,847 143,806,639 Total Person Trips 45,155 227,589,256,446 4.1.2 2001 National Household Travel Survey The 2001 NHTS continued the NPTS series. For the first time, it combined data collection for the NPTS and the American Travel Survey (ATS), previously collected in 1977 and 1995. The Bureau of Transportation Statistics (BTS) and the Federal Highway Administration (FHWA) shared the lead role coordinating the survey. The National Highway Traffic Safety Administration (NHTSA) also sponsored the 2001 survey. The survey includes data on household and person level characteristics, motor vehicles, drivers, and person trips. The survey was designed to collect data on travel period and travel day trips. Data for the travel day are collected for all modes, purposes, lengths, and area types. Additional data on one-way trips of 50 miles or more during a four-week travel period were also collected. The January 2004 version of the 2001 NHTS was considered for this analysis. This dataset includes nine-add-on areas not released in the January 2003 version. A total of 69,817 households are included in the survey. The original national sample consisted of 26,038 useable households. Households were interviewed from March 2001 through 51

PAGE 64

May 2002. The sample was selected through Random Digit Dialing (RDD). Phone numbers were checked to determine that the households were eligible for the survey. To be eligible, households were civilian, non-institutionalized population. Persons living in dormitories, nursing homes, other medical institutions, prisons, and military bases were excluded from the sample. Westat conducted the national household interviews and two add-ons, while Morpace interviewed the seven additional add-ons. All interviews were conducted by phone using Computer-Assisted Telephone Interviewing (CATI) technology. Household interviews were completed by persons at least 18 years old. A travel diary package was mailed prior to the travel day. Information on trips occurring on the travel day and long distance trips during the 28 day travel period was collected. This data includes trip purpose, travel mode, travel time, day of week, and vehicle occupancy. The 2001 NHTS data is available in three separate files, including the household, person, and daily trip file. In this study, the household, person, and travel day files are used for analysis. The following is a brief description of each file. The household file includes information on the relationship members, income, housing characteristics, area type, and other demographic variables. Each household is identified by an identification number. Each record in the data file represents a different household. There are 69,817 households in the household file. The person file includes information about the person characteristics for each person in the sample. These characteristics include age, gender, employment status, etc. The household identification number and person identification number 52

PAGE 65

uniquely identify each person. There are 160,758 person records in the person file, of these 148,616 persons over five are included in the data file. Each record describes an individual person. The travel day trip file consists of details about the characteristics of the person trip. Each record in the file represents a trip, which includes length, mode, and purpose. For the first time, trip information was collected for household members under the age of five. There are 642,292 records in the trip file, of these 603,698 trips were made by persons over the age of five. Analysis of travel time expenditures was completed at the person level. The household and person identification numbers are used to link the information in the data files. The person file includes attributes from the household interview, such as income and housing characteristics. Trip characteristics, such as total daily travel time, were aggregated and merged into the person file using the household and person identification numbers. Like the 1983 data file, the combined file was then explored to better understand relationships between socio-demographic characteristics and travel time expenditures. Person weights provided in the person file are used to obtain national estimates. Likewise, weights in the travel day file can be used to estimate the number of annual trips and miles of travel by mode at the national level. Table 4.2 provides an overview of the survey sample and weighted population for the 2001 NHTS. 53

PAGE 66

Table 4.2 Sample Size and Weighted Population for 2001 NHTS Sample Size Weighted Population Total Households 69,817 107,365,346 Total Interviewed Persons 160,758 277,203,235 Total Interviewed Persons 5 and older 148,616 253,140,309 Number of Adults (18+) 120,332 200,917,024 Average Household Size 2.5 2.6 People (5+) Who Did Not Travel 17,486 29,306,757 Share of Non-Travelers 11.8% 11.6% Number of Licensed Drivers 116,345 190,424,751 Number of Workers 85,350 145,272,118 Number of Vehicles 139,382 202,586,200 Total Person Trips (Persons 5+) 603,698 378,536,040,509 4.2 Survey Comparability Over time the national travel surveys have changed considerably. It is important to recognize these changes and the characteristics particular to each survey, so that a better understanding of the data and comparability can be obtained. Key issues of comparability, as discussed in the 2001 NHTS Users Guide, are highlighted below. The 1983 survey is often considered an anomaly in the NPTS series, which creates issues when comparing the data with other surveys. The relatively small national sample size and economic conditions during the sample period are believed to result in trip and travel data being too low. Interviews in 1983 were completed with in-home interviews, while the 2001 interviews were accomplished over the telephone. The small sample size in 1983 resulted in Census field interviewers completing only two to three interviews per month. As a result, surveyors may not have been able to maintain proficiency with the complex NPTS questionnaire. Another important difference between the 1983 NPTS and 2001 NHTS is that the 2001 NHTS collected data for household members from birth to four years of age. All 54

PAGE 67

prior surveys had limited data collection to household members five years and older and assumed that all travel for younger individuals occurred with other household members. As a result of the limitation of the 1983 data, analysis of the 2001 data are limited to persons over five years of age. The 2001 survey also focused more on capturing walk and bike trips. This increased focus may cause walk and bike trip rates to be higher than earlier surveys in the series. The impact of non-motorized travel on the overall travel time expenditure trends will be explored in the following chapter. The definition of a travel day trip in 2001 was also slightly modified. As in previous surveys trips were defined as any time a person went from one address to another. In 2001, however, respondents were asked to exclude stops made to change transportation mode. Respondents were asked additional questions on public transportation use. The intent of this change was to improve the reporting of trips using public transportation, but it may have reduced the reporting of trips where the mode change was not to public transportation. The state of the U.S. economy, the price of oil, and other events during the survey period significantly effect travel. During the 2001 survey period, two major events occurred, which undoubtedly impacted travel behavior in the United States. The first was the September 11 th terrorist attack on the World Trade Centers in New York and the Pentagon near Washington, D.C. Security of air travel in the United States was immediately increased and permanent changes in the system occurred. The attack and this security change disrupted travel and altered the amount and modes of travel during the NHTS survey period. The other event that may have affected the NHTS survey was 55

PAGE 68

the threat of anthrax being sent by U.S. mail. The extent this health concern impacted travel is not known, but it may have affected the NHTS response rates, since there was a mail component to the survey. Many of these differences are difficult to account for when comparing travel data from the 1983 and 2001 surveys. When possible, appropriate action has been taken to increase the comparability between the surveys. For example, travel data for persons under five years of age has been ignored for 2001 since this information was not collected in 1983. 56

PAGE 69

CHAPTER 5 DESCRIPTIVE ANALYSIS 5.1 Daily Travel Time Expenditure Analysis This chapter provides a detailed description of travel time expenditures for persons five years of age and older using data from the 1983 NPTS and 2001 NHTS, discussed in the previous chapter. Distributions of travel time expenditures and average daily travel time per person and traveler are explored by various socio-demographic classes and a few area characteristics. Person level analysis was completed by merging trip data into the person file. Weights from the person file for both surveys were used to obtain national population estimates. Table 5.1 provides a summary of the average travel time expenditures, sample size, and weighted population for persons and travelers. The average travel time per person and traveler may be slightly underreported because 1.0 percent of trips in 1983 and 0.1 percent of trips in 2001 had missing durations. Table 5.1 Summary of Travel Time Expenditures 1983 2001 Persons 5 and older 213,190,888 (N=16,058) 253,140,309 (N=148,616) Travelers 5 and older 154,378,581 (N=11,584) 223,833,552 (N=131,130) Average Travel Time per Person (minutes) 47.4 82.3 Standard Deviation 63.7 84.9 Average Travel Time per Traveler (minutes) 65.7 93.2 57

PAGE 70

The share of the population within four travel groups was also considered for many of the socio-demographic classes. The travel groups are defined as persons that did not travel on the travel day and travelers in the lower 10 percent, middle 80 percent, or upper 10 percent according to travel time. In 1983, the travel times linked with the limits for the lower and upper 10 percent of travelers are 13 minutes and 130 minutes. In 2001, the lower 10 percent traveled up to 20 minutes, while the upper 10 percent traveled more than 180 minutes. These travel groupings will be used to determine which characteristics are typically associated with high or low levels of travel. Table 5.2 presents the sample size, weighted population, and share of the population that are included in each travel group for 1983 and 2001. Figure 5.1 and Figure 5.2 display the distribution of daily travel time expenditures by the number of persons and share of the population. The share of non-travelers decreased 58 percent between 1983 and 2001. Discussions of the population distributions among non-travelers classes are included for several of the socio-demographics characteristics. Table 5.2 Sample Size and Weighted Population by Travel Group 1983 2001 Did Not Travel 59,432,113 (N=4,520) 27.9% 29,586,370 (N=17,629) 11.7% Lower 10% of Travelers by Daily Travel Time 15,566,245 (N=1,165) 7.3% 22,427,109 (N=14,032) 8.9% Middle 80% of Travelers by Daily Travel Time 122,017,682 (N=9,181) 57.2% 177,598,299 (N=104,726) 70.2% Upper 10% of Travelers by Daily Travel Time 16,174,848 (N=1,192) 7.6% 23,528,532 (N=12,229) 9.3% 58

PAGE 71

Figure 5.1 Distribution of Travel Time Expenditure by Number of Persons 01020304050600306090120150180210240270300330360Persons (millions)Daily Travel Time Expenditure (minutes) 1983 2001 Over 36 0 Figure 5.2 Distribution of Travel Time Expenditure by Population Share 0%5%10%15%20%25%30%0306090120150180210240270300330360Daily Travel Time Expenditure (minutes)Share of Population 1983 2001 Over 360 59

PAGE 72

The number of trips per person per day also increased between 1983 and 2001. The average number of trips per person in 1983 was 2.8 per day, while in 2001 the average was 4.1 trips per day. The average number of trips per traveler did not increase to the same extent as the per person average. In 1983, the average number of trips per traveler was 3.9 per day. The average in 2001 was 4.6 trips per traveler per day. Figure 5.3 and Figure 5.4 provide the distribution of daily trips by the number of persons and share of the population based on the NPTS and NHTS data. It is interesting to note the peaks in two and four trips per day, which possibly indicate two round trips, and the small share of one way trips. The average speed of travel can be determined by dividing the trip distance by the duration. Both the average trip distance and duration have increased since 1983. The average trip distance increased 16.4 percent from 8.7 miles per trip to 10.1 miles per trip. The average duration of a trip increased from 16.8 minutes to 19.3 minutes or 14.9 percent. Figures 5.5 and 5.6 provide the distributions for trip length and duration for 1983 and 2001. The share of trips longer than 10 miles and 15 minutes increased in 2001. Many survey participants round trip durations to fifteen minute intervals. This is evident by the peaks in trip duration at 30, 45, and 60 minutes. The average speed during this period only increased 1.3 percent, from 31.1 miles per hour in 1983 to 31.5 miles per hour in 2001. Total daily travel time expenditures have increased as a result of the increase in trip frequency and duration. The following subsections will explore the increases in daily travel time expenditures by travel mode and trip purpose. Many socio-demographic characteristics will also be discussed in subsequent subsections. 60

PAGE 73

Figure 5.3 Distribution of Daily Trip Count by Number of Persons 0102030405060024681012141618202224Persons (millions)Number of Trips 1983 2001 Over 25 Figure 5.4 Distribution of Daily Trip Count by Population Share 0%5%10%15%20%25%30%024681012141618202224Number of TripsShare of Population 1983 2001 Over 25 61

PAGE 74

Figure 5.5 Distribution of Trip Distance in Miles 0%10%20%30%40%50%60%70%Under 5101520253035404550Over 50Trip Distance (miles)Share of Trips 1983 2001 Figure 5.6 Distribution of Trip Duration in Minutes 0%5%10%15%20%25%30%35%Under 5101520253035404550556065707580859095100105110115120Over 120Trip Duration (minutes)Share of Trips 1983 2001 62

PAGE 75

5.1.1 Travel Mode The mode used for travel is related to many of the trip characteristics including speed, distance, and duration. This subsection will discuss travel time expenditures by mode. Table 5.3 provides a summary of the average minutes of daily travel spent traveling by private vehicle, public transportation, bike, walk, and other. The total minutes per person and traveler are less than the overall averages reported in Table 5.1, because the trips with a missing mode have been excluded from the analysis. In 1983, 3.3 percent of trips were missing the mode of travel, while only 0.1 percent of trips were missing the mode in 2001. Figure 5.7 and Figure 5.8 display the average minutes of travel per day for persons and travelers. The modes are grouped into five categories: private vehicles, public transportation, bike, walk and other. Private vehicles include automobiles, vans, pickup, other trucks, RV or motor homes and motorcycles. Public transportation includes motorbuses, commuter trains, streetcar or trolleys, elevated rail and subway systems. Trips by airplane, school bus, taxi, Amtrak, and moped are included in the other category. Table 5.3 Summary of Travel Time Expenditures by Mode 1983 2001 Mode Average Minutes per Traveler Average Minutes per Person Share of Trips by Mode Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Trips by Mode Average Time Per Trip with Reported Duration Private Vehicle 53.1 38.3 84.5 16.5 78.1 69.0 87.3 18.6 Public Transportation 3.2 2.3 2.4 36.4 3.5 3.1 1.3 46.4 Bike 0.4 0.3 0.8 14.8 0.9 0.8 0.8 20.7 Walk 3.3 2.4 8.6 9.9 5.9 5.3 8.0 15.1 Other 3.4 2.4 3.6 26.3 4.5 4.0 2.6 38.9 Total 63.4 45.7 100.0 16.8 92.9 82.2 100.0 19.2 63

PAGE 76

Figure 5.7 Average Daily Travel Time per Traveler by Mode 01020304050607080Private VehiclePublicTransportationBikeWalkOtherTrip ModeMinutes 1983 2001 Figure 5.8 Average Daily Travel Time per Person by Mode 01020304050607080Private VehiclePublicTransportationBikeWalkOtherTrip ModeMinutes 1983 2001 64

PAGE 77

Private vehicles account for the highest share of trips and the largest increase of daily travel time expenditures per person and traveler from 1983 to 2001. The daily travel time distribution by population share for private vehicle is displayed in Figure 5.9. From this figure it is evident that higher shares of the population are spending more time per day traveling by private vehicle. In 1983, only 62.6 percent of the population or 86.8 percent of travelers made at least one trip by private vehicle. This share increased to 80.0 percent of the population and 90.6 percent of travelers in 2001. Figure 5.9 Distribution of Daily Travel Time Expenditure by Private Vehicle 0%5%10%15%20%25%30%35%40%0153045607590105120135150165180Over 180Daily Travel Time Expenditure (minutes)Share of Population 1983 2001 The number of trips by each mode plays a role in determining overall travel time expenditures. The average number of trips by mode per person is shown in Figure 5.10. The number of private vehicle trips per person increased 53 percent from 2.3 trips per person per day in 1983 to 3.5 trips per person per day in 2001. This increase in trip making by privately owned vehicles helps to explain why overall travel time expenditures 65

PAGE 78

have increased. The average time per vehicle trip also increased from 16.5 minutes to 18.6 minutes. The number of trips per person by the walk mode also increased from 0.24 trips per person per day to 0.35 trips per person per day. The increase in walk trips in 2001 maybe attributed to the additional focus on capturing more walk and bike trips people made, which could have previously been not reported. The impact of walk trips on overall travel time was considered to determine if additional corrections for the methodology change in 2001 were required. The shape of the daily travel time distribution without the walk mode was similar to the distribution with the walk mode. The removal of the walk mode increased the share of the population that did not travel by 3.6 percent in 1983 and 3.4 percent in 2001. The inclusion of the walk mode increases average daily travel time expenditures by 2.4 minutes per person in 1983 and 5.3 minutes per person in 2001. For the purposes of this report, the walk mode is included in the analysis. Figure 5.10 Average Daily Trip Count per Person by Mode 01234Private VehiclePublicTransportationBikeWalkOtherTrip ModeNumber of Trips 1983 2001 66

PAGE 79

Trip distance, in miles, is also a characteristic of the travel mode. The average trip distance by mode is shown in Figure 5.11. The other category has the longest trip distances, because air and Amtrak are included in this category. The average trip distance for private vehicles has increased, while the average distance for public transportation has decreased. Distributions for daily person miles of travel (PMT) and vehicle miles of travel (VMT) per traveler are presented in Figures 5.12 and 5.13. From these figures, it is apparent that VMT drives most PMT. The occupancy of a private vehicle determines the actual amount of vehicle miles that will be taken on the roadway. In 1983, approximately 47 percent of vehicle trips occurred with only one occupant. The drive-alone share of vehicle trips increased to 49 percent in 2001. The average vehicle occupancy decreased slightly from 1.7 in 1983 to 1.6 in 2001. Figure 5.11 Average Trip Distance in Miles by Mode 010203040Private VehiclePublicTransportationBikeWalkOtherTrip ModeMiles 1983 2001 67

PAGE 80

Figure 5.12 Distribution of Person Miles of Travel by Share of Travelers 0%5%10%15%20%25%Under 5101520253035404550556065707580859095100Over 100Person Miles of TravelShare of Travelers 1983 2001 Figure 5.13 Distribution of Vehicle Miles of Travel by Share of Travelers 0%5%10%15%20%25%Under 5101520253035404550556065707580859095100Over 100Vehicle Miles of TravelShare of Travelers 1983 2001 68

PAGE 81

5.1.2 Trip Purpose As discussed in Chapter 2 of this report, activity participation typically drives travel demand, except for trips made solely for the purpose of traveling. The survey data for 1983 includes 11 reasons for travel day trips. For trips that have more than one trip purpose, the main reason is chosen if the destination of both trips is the same. The 2001 survey data provides less aggregate trip purpose data, but for the purpose of comparison the categories from the 1983 NPTS are used. The trip purposes can be grouped into four major purposes, including earning a living, family and personal business, school or church, and social or recreational. The earning a living category includes trips to or from work and work-related business. Family and personal business trips are for shopping, visits to the doctor or dentist, and other related categories. These trips also include taking passengers to a destination. Vacation, visits to family and friends, pleasure diving, and other trips are grouped into the social or recreational trip purposes. The average number of trips per person has increased for the four major trip purpose categories. The majority of the increase in the average number of trips per day can be attributed to increases in the number of family and personal business trips and social or recreational trips. Family and personal business trips increased by 0.8 trips per person per day; of these, 0.3 trips per person were for shopping. Figure 5.14 shows the average number of trips by purpose category. The purpose of a trip can be related to the trip characteristics, including travel time. For workers, the commute trip is often the most regular daily travel time expenditure. Total travel time expenditures for workers are covered later in this chapter. 69

PAGE 82

While the work trip is a significant portion of total travel for persons with a job, discretionary trips including the family and personal business, shopping, and social recreation trips have shown larger increases in the average minutes per day. During this time period, work trips increased 43.5 percent (4.7 minutes per person per day), shopping trips increased 93.4 percent (6.1 minutes per person per day), other family and personal business trips increased 120.8 percent (8.0 minutes per person per day), and other social recreation trips increased 98.0 percent (7.7 minutes per person per day). Table 5.4 provides a summary of travel time expenditures by purpose. Figure 5.15 and Figure 5.16 present the average time spent in travel for each purpose. The share of daily travel time spent on each trip purpose is presented for 1983 and 2001 in Figures 5.17 and 5.18. Travel time expenditures by purpose, also vary by gender. This will be discussed later in this chapter. As mentioned above, the 2001 NHTS provides more detailed trip purposes than the 1983 dataset. Table 5.5 provides a summary of the average daily travel time expenditure by purpose category in 2001. The share of trips by trip purpose is also included in this table. Average daily activity duration can also be estimated from the survey data. However, the only activity durations that can be estimated from the data are those that occurred between the start of the first trip and end of the last trip. Activities that did not occur during this time and the activities of non-travelers are included into the home/other time expenditure category. As the number of travelers has increased and persons are traveling more, the average time per day spent at home has decreased. Figure 5.19 and Figure 5.20 display the share of time per person for each activity for 1983 and 2001. 70

PAGE 83

0.00.51.01.52.0hool or ChurchSocial or RecreationalseAverage Number of Trips Work or RelatedFamily and PersonalBusinessScTrip Purpo 1983 2001 Table 5.4 Summary of Travel Time Expenditures by Purpose Figure 5.14 Average Number of Trips per Person by Purpose 1983 2 001 Purpose Average Minutes per Traveler Average Minutes per Person Share of Trips by Purpose Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Trips by Purpose Average Time Per Trip with Reported Duration To or From Work 14.8 10 .7 20 .2 18.2 17.4 15.4 16.1 23.8 Work Related 2.1 1. 5 2. 5 22.7 4.8 4.2 2.8 33.8 Shopping 9. 1 6.6 18.7 12.6 14.4 12.7 19.3 15.8 Other Family or Personal Business 9.1 6. 6 14.7 14.2 16.5 14.6 19.6 16.2 School/Church 7. 3 5.3 12.0 15.4 8.0 7.0 10.1 17.6 Doctor/Dentist 1.0 0. 7 1. 5 20.7 2.3 2.1 2.6 23.3 Vacation 1.8 1. 3 0. 4 158.0 1.7 1.5 0.8 60.4 Visit friends or relatives 8. 8 6.3 11.5 19.9 8.8 7.8 8.7 24.9 Other Social Recreational 10 .9 7.9 15.4 17.2 17.6 15.6 18.5 20.8 Other 0.8 0. 4 3. 0 19.5 1.6 1.4 1.4 34.0 Total 65 .7 47.4 100.0 16.8 93.2 82.3 100.0 20.1 71

PAGE 84

05101520To or FromWorkWork RelatedShoppingOther Family orPersonalBusinessSchool/ChurchDoctor/DentistVacationVisit friends orrelativesOther SocialRecreationalTrip PurposeMinutes 1983 2001 05101520To or FromWorkWork RelatedShoppingOther Family orPersonalBusinessSchool/ChurchDoctor/DentistVacationVisit friends orrelativesOther SocialRecreationalTrip PurposeMinutes 1983 2001 Figure 5.15 Average Daily Travel Time per Traveler by Purpose 72 Figure 5.16 Average Daily Travel Time per Person by Purpose

PAGE 85

Figure 5.17 Share of Daily Travel Time by Purpose in 1983 1983Avg 67.16 min per traveler48.63 min per person To or From Work21%Work Related3%Visit friends or relatives13%Other Family or Personal Business14%Other Social Recreational16%Shopping14%Vacation3%Doctor/Dentist2%School/Church11%Other 3% Figure 5.18 Share of Daily Travel Time by Purpose in 2001 2001Avg 93.08 min per traveler82.30 min per person Work Related5%Shopping16%School/Church9%Doctor/Dentist3%Vacation2%Visit friends or relatives9%Other 1%Other Social Recreational19%Other Family or Personal Business18%To or From Work18% 73

PAGE 86

Table 5.5 Summary of Travel Time Expenditures by Purpose in 2001 Trip Purpose Share of Trips Person Miles of Travel Minutes per Person Return home 34.1 13.1 26.5 Go to work 7.8 3.9 7.2 Buy goods: groceries/clothing/hardware store 11.4 2.9 6.7 Visit friends/relatives 4.6 3.2 4.9 Get/eat meal 4.8 1.4 3.0 Go to gym/exercise/play sports 3.1 0.8 3.0 Other work related 1.9 1.8 2.4 Go to school as student 2.9 0.7 2.1 Shopping/errands 2.9 0.9 1.9 Drop someone off 3.0 1.0 1.9 Go out/Hang out 1.7 1.0 1.8 Pick up someone 2.7 0.8 1.6 Buy services: video rental, dry cleaner, post office, car service, bank 3.3 0.7 1.6 Family personal business/obligations 1.5 0.8 1.4 Medical/dental services 1.4 0.6 1.3 Social/recreational 1.1 0.6 1.2 Rest or relaxation/vacation 0.4 1.1 1.2 Buy gas 1.5 0.7 1.0 Other 0.8 1.2 1.0 Return to work 1.8 0.5 1.0 Go to religious activity 1.5 0.4 0.9 Social event 0.5 0.3 0.5 Visit public place 0.4 0.2 0.5 School/religious activity 0.6 0.2 0.4 Meals 0.6 0.2 0.4 Attend business meeting or trip 0.2 0.5 0.4 Attend meeting 0.5 0.2 0.4 Take someone and wait 0.5 0.2 0.4 Pet care: walk the dog/visit vet 0.4 0.1 0.3 Coffee/ice cream/snacks 0.6 0.1 0.3 Attend wedding/funeral 0.2 0.2 0.2 Use personal services: grooming/haircut/nails 0.4 0.1 0.2 Use professional service: attorney/accountant 0.2 0.1 0.2 Day care 0.2 0.0 0.1 Go to library: school related 0.1 0.0 0.1 Transport someone 0.1 0.0 0.1 74

PAGE 87

Figure 5.19 Estimated Activity Durations per Person in 1983 1983 Persons Work/Work Related(1.8 hrs)7%Home/Other(19.7 hrs)82%Travel(0.8 hrs)3%Social/Recreational(0.5 hrs)2%Other Family/Personal Business (0.4 hrs)2%Shopping (0.3 hrs) 1%School or Church (0.8 hrs) 3% Figure 5.20 Estimated Activity Durations per Person in 2001 2001 Persons School or Church (1.0 hrs) 4%Shopping (0.6 hrs) 3%Other Family/Personal Business (0.8 hrs)3%Social/Recreational(1.3 hrs)5%Travel(1.3 hrs)5%Home/Other(17.1 hrs)71%Work/Work Related(2.4 hrs)10% 75

PAGE 88

5.1.3 Average Daily Travel Time by Travel Day Travel time expenditures vary with the day of the week. Table 5.6 provides a summary of travel time expenditures by the day of the week. Higher travel time expenditures for persons and travelers tend to occur on Friday and Saturday. Average travel time expenditures have increased for each day of the week. The largest increases per person occurred on Friday, which increased 41.7 minutes, and Wednesday, with an increase of 40.2 minutes per person. Figure 5.21 displays travel time expenditures per persons for 1983 and 2001. Figure 5.22 presents the share of non-traveler. Changes in daily activity patterns could explain the increases in the share of population which travels and daily travel time expenditures. Table 5.7 provides the average travel time per person by week day and trip purpose. Figure 5.23 shows the difference between 2001 and 1983 average daily travel time by general trip purpose. As discussed in the previous subsection, the majority of growth resulted from family or personal business and other social or recreational travel increases. Table 5.6 Summary of Travel Time Expenditures by Travel Day 1983 2001 Travel Day Average Minutes per Traveler Average Minutes per Person Share of Population Traveling Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Traveling Average Time Per Trip with Reported Duration Sunday 67.9 44.8 66.4 18.6 88.6 74.0 83.6 20.4 Monday 65.5 47.7 73.3 16.5 88.0 78.5 89.3 18.7 Tuesday 63.5 47.8 75.5 17.2 89.2 79.2 88.9 18.5 Wednesday 60.1 43.7 72.8 15.3 93.1 83.9 90.2 18.9 Thursday 64.5 49.0 76.3 16.6 90.5 81.7 90.4 18.7 Friday 67.6 50.7 75.0 16.2 101.8 92.4 90.8 19.5 Saturday 71.2 48.1 67.8 17.9 100.9 86.6 85.9 20.4 76

PAGE 89

Figure 5.21 Average Daily Travel Time per Person by Travel Day 020406080100SundayMondayTuesdayWednesdayThursdayFridaySaturdayTravel DayMinutes 1983 2001 Figure 5.22 Share of Non-Travelers by Travel Day 0%10%20%30%40%50%SundayMondayTuesdayWednesdayThursdayFridaySaturdayTravel DayShare of Non-Travelers 1983 2001 77

PAGE 90

Table 5.7 Average Travel Time Expenditures per Person by Travel Day and Trip Purpose 1983 Purpose Travel Day Work or Related Family or Personal Business School or Church Social or Recreational Other Sunday 3.1 10.0 6.1 25.0 0.6 Monday 12.8 15.1 5.0 13.7 1.2 Tuesday 16.2 13.6 6.6 10.5 0.9 Wednesday 15.7 12.0 5.3 10.1 0.6 Thursday 16.6 14.0 5.9 11.1 1.4 Friday 13.0 15.6 5.1 15.8 1.2 Saturday 4.9 15.8 2.0 24.3 1.1 2001 Purpose Travel Day Work or Related Family or Personal Business School or Church Social or Recreational Other Sunday 4.4 23.6 8.3 33.4 1.8 Monday 22.2 27.1 6.4 17.2 1.6 Tuesday 23.3 26.4 7.2 16.8 1.2 Wednesday 23.4 27.7 8.0 18.5 1.6 Thursday 23.0 27.0 6.8 18.9 1.4 Friday 21.1 31.8 6.0 26.5 1.6 Saturday 7.1 33.0 2.0 39.4 1.8 Figure 5.23 Growth in Average Daily Travel Time per Person by Trip Purpose and Travel Day from 1983 to 2001 0510152025303540SundayMondayTuesdayWednesdayThursdayFridaySaturdayTravel DayMinutes Work or Related Family or Personal Business School or Church Social or Recreational 78

PAGE 91

5.1.4 Average Daily Travel Time by Gender As discussed in Chapter 2, a persons gender is related to daily travel time expenditure and travel behavior. Although some researchers have found contradictory results, data from the 1983 NPTS and 2001 NHTS suggest that men are more likely to spend more time traveling per day than women. Table 5.8 presents the average minutes of travel per day for men and women per traveler and per person. The share of the population and average trip duration is also provided in this table. Figure 5.24 and Figure 5.25 show the average daily travel time per traveler and person by gender for 1983 and 2001. In 2001, men spent an average of 87.4 minutes per day traveling, while women spent an average of 77.4 minutes per person per day traveling. Travel time distributions for males and females are presented in Figure 5.26 and Figure 5.27. The share of the population by gender for each travel group was also explored as part of the analysis. In 2001, 90.0 percent of males and 86.7 percent of females were travelers. In 1983, 73.1 percent of males and 71.3 percent of females traveled on the travel day. Slightly higher shares of females did not travel for both study years, whereas men tend to have a higher share of travelers in the upper 10 percent of daily travel times. Travel time expenditures by gender will also be considered as a part of other areas including age group and worker status. Table 5.8 Summary of Travel Time Expenditures by Gender 1983 2001 Gender Average Minutes per Traveler Average Minutes per Person Share of Population Average Time per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time per Trip with Reported Duration Male 68.4 49.9 48.2 17.6 97.2 87.4 48.8 21.4 Female 63.1 45.0 51.8 16.0 89.3 77.4 51.2 18.9 79

PAGE 92

Figure 5.24 Average Daily Travel Time per Traveler by Gender 0102030405060708090100MaleFemaleGenderMinutes 1983 2001 Figure 5.25 Average Daily Travel Time per Person by Gender 0102030405060708090100MaleFemaleGenderMinutes 1983 2001 80

PAGE 93

Figure 5.26 Distribution of Daily Travel Time Expenditure by Share of Males 0%5%10%15%20%25%30%0306090120150180210240270300330360Daily Travel Time Expenditure (minutes)Share of Males 1983 2001 Over 360 Figure 5.27 Distribution of Daily Travel Time Expenditure by Share of Females 0%5%10%15%20%25%30%0306090120150180210240270300330360Daily Travel Time Expenditure (minutes)Share of Females 1983 2001 Over 360 81

PAGE 94

On average, men and women have different daily time expenditure patterns. Therefore, it is important to consider the differences in daily travel expenditures by purpose for each gender separately. Tables 5.9 and 5.10 provide a summary of the average daily travel time for each gender. The averages for 1983 are slightly higher than the total averages reported for males and females because of different weighting mechanisms in the travel day file. For the other analysis, total travel time per person had been merged into the person file, but average travel time by purpose was calculated directly from the travel day file which had slightly different person weights for total travel. Figure 5.28 and Figure 5.29 present the average daily travel time per person for each trip purpose for males and females. These figures and Tables 5.9 and 5.10 show that men tend to travel more for work and work related activities, while women spend more time traveling for shopping activities and other non-work related purposes. In a later section, travel times by worker status for males and females will be discussed. In terms of change, both males and females have experienced increases in average daily travel time per person for all trip purposes. For males, work trips increased 5.5 minutes per person between 1983 and 2001. Work related trips also increased 4.6 minutes per person. Work and work related trips only increased 3.9 minutes and 0.9 minutes per person for females. Both men and women exhibited increases in shopping, other family and personal business, and other social or recreational activities. Average daily travel time expenditures for females increase a total of 22.1 minutes per person for these trip purposes. Similarly, average travel time expenditures for males increased 21.7 minutes per person for these trip purposes. 82

PAGE 95

83 Table 5.9 Summary of Travel Time Expenditures by Purpose for Males 1983 2 001 Purpose Average Minutes per Traveler Average Minutes per Person Share of Trips by Purpose Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Trips by Purpose Average Time Per Trip with Reported Duration To or From Work 18.8 13 .8 23.5 19.2 21 .4 19 .3 18.4 25.7 Work Related 3.2 2. 3 3.0 25.8 7. 6 6. 9 4.2 37.7 Shopping 7.8 5.7 15.8 12.4 13 .0 11 .7 18.6 16.0 Other Family or Personal Business 9.0 6.6 15.0 15.0 15 .6 14 .1 19.5 17.1 School/Church 7.1 5. 2 11.8 16.2 7. 5 6. 8 9.4 17.6 Doctor/Dentist 0.7 0. 5 0.9 21.3 1. 8 1. 6 1.8 24.4 Vacation 1.9 1. 4 0.3 176.0 1. 7 1. 5 0.7 60.9 Visit friends or relatives 8.4 6. 1 10.6 19.7 8. 5 7. 7 7.2 25.2 Other Social Recreational 11.4 8.3 16.2 17.7 18 .4 16 .5 19.1 21.2 Other 2.2 1. 6 2.9 20.3 1. 3 1. 2 1.0 33.6 Total 70.4 51 .6 100.0 17.6 97 .2 87 .4 100.0 21.4 Table 5.10 Summary of Travel Time Expenditures by Purpose for Females 1983 2 001 Purpose Average Minutes per Traveler Average Minutes per Person Share of Trips by Purpose Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Trips by Purpose Average Time Per Trip with Reported Duration To or From Work 11.1 7.9 16.3 16.8 13 .6 11 .8 14.3 21.4 Work Related 1.1 0. 8 1.6 17.2 2. 0 1. 7 1.7 24.2 Shopping 10.3 7.4 20.7 12.7 15 .8 13 .7 22.4 15.6 Other Family or Personal Business 9.3 6.6 17.3 13.5 17 .5 15 .1 22.7 15.6 School/Church 7.4 5. 3 12.6 14.7 8. 4 7. 3 9.6 17.6 Doctor/Dentist 1.3 0. 9 1.7 20.4 2. 9 2. 5 2.7 22.7 Vacation 1.8 1. 3 0.3 143.3 1. 7 1. 5 0.6 59.9 Visit friends or relatives 9.1 6. 5 11.4 20.1 9. 1 7. 9 7.5 24.6 Other Social Recreational 10.4 7.4 15.1 16.7 16 .9 14 .7 17.7 20.4 Other 2.2 1. 6 3.0 18.8 1. 3 1. 1 0.8 34.5 Total 64.1 45 .8 100.0 16.0 89 .3 77 .4 100.0 18.9

PAGE 96

05101520To or FromWorkWork RelatedShoppingOther Family orPersonalBusinessSchool/ChurchDoctor/DentistVacationVisit friends orrelativesOther SocialRecreationalTrip PurposeMinutes 1983 2001 05101520To or FromWorkWork RelatedShoppingOther Family orPersonalBusinessSchool/ChurchDoctor/DentistVacationVisit friends orrelativesOther SocialRecreationalTrip PurposeMinutes 1983 2001 84 Figure 5.28 Average Daily Travel Time per Male by Purpose Figure 5.29 Average Daily Travel Time per Female by Purpose

PAGE 97

5.1.5 Average Daily Travel Time by Age Group Many studies have found the age of an individual to be a significant factor related to daily travel time expenditures. Table 5.11 provides a summary of daily travel time expenditures by age group. This table also contains the share of the population in each group and average trip duration. As the United States population ages, higher shares of the population will be in the upper age groups, which historically have had lower travel levels. Figure 5.30 also displays the daily travel time expenditure by age group for all persons. Peak years of travel tend to occur for persons between 25 and 64 years of age, which are typical years for employment. Younger and older population groups tend to have lower average daily travel time and also include higher shares of non-traveler. Figure 5.31 provides the share of non-travelers in each age group. The share of non-travelers increases after the age of 45. The share of non-travelers has decreased for every age group. The shares of travelers for persons under 14 years of age and persons over 65 years of age increased the more than the middle aged population. Table 5.11 Summary of Travel Time Expenditures by Age Group 1983 2001 Age Group Average Minutes per Traveler Average Minutes per Person Share of Population Average Time per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time per Trip with Reported Duration 5 to 14 54.8 36.5 15.8 16.5 69.4 61.6 16.3 18.0 15 to 24 70.3 55.0 18.7 16.3 90.0 81.2 13.8 20.1 25 to 34 71.5 57.8 18.7 16.7 97.6 89.6 16.1 20.6 35 to 44 71.9 58.8 13.8 17.3 103.6 95.7 17.2 20.3 45 to 54 66.0 50.0 10.4 17.5 102.7 93.3 14.1 20.8 55 to 64 63.0 44.8 11.1 17.2 99.5 86.4 9.5 21.0 65 to 74 54.3 34.9 7.0 16.3 95.9 77.2 7.3 20.0 75 to 84 46.2 23.5 3.6 16.1 83.0 60.3 4.5 19.6 85 and older 49.2 18.7 0.8 24.2 68.7 34.2 1.1 18.4 85

PAGE 98

Figure 5.30 Average Daily Travel Time per Person by Age Group 0204060801005 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 andolderAge GroupMinutes 1983 2001 Figure 5.31 Share of Non-Travelers by Age Group 0%10%20%30%40%50%60%70%80%5 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 andolderAge GroupShare of Non-Travelers 1983 2001 86

PAGE 99

As discussed in the previous section, gender is related to daily travel time. Therefore, analysis of age groups by gender may provide insight into travel time expenditure trends. Some changes in society tend to work through various age groups as the population ages and may be more evident among younger individuals than older ones. For example, in the past lower shares of women worked and were licensed drivers. However, in todays society labor force participation and licensure rates for women are approaching those of men. These factors are important to mobility and daily travel time expenditure trends and will be further discussed later in this chapter. A summary of daily travel time expenditures for males by age group is provided in Table 5.12. Figure 5.32 exhibits the average minutes of travel per day for males by age group for persons. Average daily travel times for males is higher than the overall average for all persons. Males for all age groups are more likely to travel and have higher shares of travelers in the upper travel group. Figure 5.33 provides the share of non-travelers for males by age. Table 5.12 Summary of Travel Time Expenditures by Age Group for Males 1983 2001 Age Group Average Minutes per Traveler Average Minutes per Person Share of Male Population Average Time per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Male Population Average Time per Trip with Reported Duration 5 to 14 57.8 37.5 16.8 17.2 69.9 62.1 17.1 17.8 15 to 24 71.6 56.5 19.2 16.8 89.2 81.0 14.6 20.7 25 to 34 73.8 60.7 19.0 17.5 101.3 94.1 16.4 22.5 35 to 44 73.9 61.7 14.0 19.2 110.7 103.5 17.4 23.0 45 to 54 65.2 51.0 10.5 18.5 109.6 100.7 13.9 22.7 55 to 64 66.2 49.9 10.9 17.9 106.4 94.9 9.4 22.0 65 to 74 54.2 40.4 6.3 16.4 100.8 85.0 6.6 20.2 75 to 84 45.8 30.9 2.8 16.3 89.9 69.7 3.8 20.3 85 and older 26.2 18.1 0.5 19.7 78.4 48.1 0.8 19.2 87

PAGE 100

Figure 5.32 Average Daily Travel Time per Male by Age Group 0204060801001205 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 andolderAge GroupMinutes 1983 2001 Figure 5.33 Share of Male Non-Travelers by Age Group 0%10%20%30%40%50%60%70%80%5 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 andolderAge GroupShare of Non-Travelers 1983 2001 88

PAGE 101

Females tend to travel less on average than males. A summary of daily travel time expenditures for females by age group is available in Table 5.13. The peak in daily travel time per traveler for females 85 years of age and older is a result of the small sample size, only 103 persons, for this group in 1983. Figure 5.34 displays the average minutes of travel per day for females by age group for persons. Average daily travel times for females is lower than the overall average for all persons. Also, females account for higher shares of non-travelers, especially in older age groups. Figure 5.35 provides the share of non-travelers for females by age. Travel time has grown for both genders all age groups, excluding women over 85 years of age. Figure 5.36 shows the increase from 1983 to 2001 by gender and age group. The average for males increased more than that of females. The difference between the average daily travel times per person for men and women for 1983 and 2001 is also graphically displayed in Figure 5.37. Table 5.13 Summary of Travel Time Expenditures by Age Group for Females 1983 2001 Age Group Average Minutes per Traveler Average Minutes per Person Share of Female Population Average Time per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Female Population Average Time per Trip with Reported Duration 5 to 14 51.9 35.6 14.8 15.8 68.9 61.1 15.4 18.1 15 to 24 68.9 53.4 18.2 15.8 90.8 81.5 13.1 19.6 25 to 34 69.2 55.1 18.3 15.9 93.9 85.2 15.9 18.9 35 to 44 69.8 56.0 13.6 15.7 96.7 88.3 17.1 17.9 45 to 54 66.9 49.1 10.4 16.6 96.1 86.5 14.2 19.1 55 to 64 59.8 40.2 11.4 16.6 92.7 78.5 9.6 20.0 65 to 74 54.4 30.8 7.7 16.2 91.6 71.0 8.0 19.8 75 to 84 46.5 19.0 4.3 15.8 77.6 53.8 5.2 19.0 85 and older 72.2 18.9 1.2 26.4 61.9 27.2 1.5 17.8 89

PAGE 102

Figure 5.34 Average Daily Travel Time per Female by Age Group 0204060801001205 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 andolderAge GroupMinutes 1983 2001 Figure 5.35 Share of Female Non-Travelers by Age Group 0%10%20%30%40%50%60%70%80%5 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 andolderAge GroupShare of Non-Traveler in Group 1983 2001 90

PAGE 103

Figure 5.36 Increase of Average Travel Time per Person for Males and Females from 1983 to 2001 0 10 20 30 40 50 60 5 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 and older Age GroupIncrease of Travel Time (Minutes) Men Women Figure 5.37 Difference between Average Travel Time for Males and Females by Age Group -5 0 5 10 15 20 25 5 to 1415 to 2425 to 3435 to 4445 to 5455 to 6465 to 7475 to 8485 and older Age GroupTravel Time Difference (Minutes) 1983 2001 91

PAGE 104

5.1.6 Average Daily Travel Time by Worker Status and Gender The worker status of an individual is related to the average daily travel time expenditures. Workers typically travel more on average than non-workers. For this section, only adults, a person over 18 years of age, are included in the analysis. The share of adult women that work increased from 50.5 percent in 1983 to 61.7 percent in 2001. The share of males that work also increased during this time period from 74.2 percent to 78.0 percent. Table 5.14 provides the average daily travel time by gender and worker status for travelers and all adults. Male workers and non-workers have higher average daily travel times per person and per traveler than females. Figure 5.38 presents the average travel time per adult person for each gender and worker status. Non-workers also tend to account for higher shares of non-travelers for both genders. Figure 5.39 provides the share of each group that did not travel in 1983 and 2001. In 2001, less than 7.0 percent of workers did not travel on the travel day. While both genders and work status had decreases in the number of non-travelers, female non-workers had the largest decrease, from 40 percent non-travelers to 24 percent in 2001. Table 5.14 Summary of Travel Time Expenditures by Worker Status 1983 2001 Worker Status Average Minutes per Traveler Average Minutes per Person Share of Adult Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Adult Population Average Time Per Trip with Reported Duration Male Worker 73.6 59.8 35.1 18.0 106.5 99.6 37.5 22.6 Male Non-Worker 63.2 40.1 12.2 17.4 95.2 74.9 10.6 20.6 Female Worker 67.7 57.6 26.6 16.3 94.9 87.9 32.0 19.0 Female Non-Worker 61.0 36.6 26.0 15.8 90.8 69.4 19.9 19.0 Adult Worker 71.0 58.9 61.7 17.3 101.2 94.2 69.5 20.9 Adult Non-Worker 61.7 37.7 38.3 16.3 92.4 71.3 30.5 19.6 92

PAGE 105

Figure 5.38 Average Daily Travel Time per Person by Worker Status 020406080100120Male WorkerMale Non-WorkerFemale -WorkerFemale Non-WorkerAdult -WorkerAdult Non-WorkerMinutes 1983 2001 Figure 5.39 Share of Adult Non-Travelers by Worker Status and Gender 0%10%20%30%40%50%Male WorkerMale Non-WorkerFemale WorkerFemale Non-WorkerTotal AdultsWorker StatusShare of Non-Travelers in Group 1983 2001 93

PAGE 106

5.1.7 Average Daily Travel Time by Life Cycle The life cycle or household composition is also an indication of the amount of activities or travel persons in the household may be participating in. For example, retired persons without children will not require work trips or trips to chauffer children between activities. Table 5.15 breaks down daily travel time expenditures by life cycle group. Figure 5.40 also presents this information graphically for persons. One or two person households without children have the highest travel levels per person, whereas single adult household that are retired have the lowest on a per person basis. Many of the one adult, retired households are elderly women, which tend to have very low travel levels. The share of persons in each life cycle group has changed since 1983. As the population has aged, higher shares fall into the retired life style groups. Historically, this group had lower travel times than the average, but demonstrated the highest percent increase in travel times from 1983 to 2001. Travel times per person and traveler have increased for every life cycle grouping. In percentage terms, increases have ranged from 52 percent, increasing from 49 minutes per person to 74.7 minutes, for households with two or more adult and children under five to 129 percent, increasing from 32.9 minutes per person to 75.3 minutes, for household with two or more retired adults without children. Other households without children also experienced increases in average daily travel time expenditures. Single adult households increased 86.9 percent, from 45 minutes per person to 84.1 minutes per person. Multiple adult households without children also increased from 49.6 minutes per person to 86.0 minutes per person. The share of the population in these groups decreased during the time period. 94

PAGE 107

95 The share of non-travelers in each group has also decreased for every life cycle grouping. Figure 5.41 provides the share of nontravelers for each life cycle group in 1983 and 2001. The share of non-travelers in retired households, both single and multiple adults, without children have remain ed the highest at approximately 22 percent of the life cycle group. Table 5.15 Summary of Travel Time Expenditures by Life Cycle 1983 2 001 Life Cycle Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration one adult, no children 63.7 45 .0 8.2 17.0 91 .9 84 .1 5.3 18.5 2+ adults, no children 67.3 49 .6 24 .2 17.4 95 .2 86 .0 19 .9 19.4 one adult, youngest child 0-5 62.8 38 .5 2.1 18.0 80 .1 70 .6 1.7 21.1 2+ adults, youngest child 0-5 67.0 49 .0 20 .7 17.0 83 .4 74 .7 20 .1 17.9 one adult, youngest child 6-15 56.8 40 .4 4.5 15.0 86 .1 77 .9 3.2 18.4 2+ adults, youngest child 6-15 65.3 49 .0 23 .5 16.6 83 .8 76 .2 24 .7 19.0 one adult, youngest child 16-21 60.2 46 .8 0.9 16.6 88 .1 80 .8 1.0 18.0 2+ adults, youngest child 16-21 67.3 48 .5 8.0 17.1 90 .5 82 .4 7.2 19.2 one adult, retired, no children 58.2 32 .9 1.0 16.7 87 .1 67 .2 3.8 19.4 2+ adults, retired, no children 53.6 32 .9 6.9 15.3 92 .8 75 .3 13 .3 19.2

PAGE 108

020406080100one adult, nochildren2+ adults, nochildrenone adult,youngestchild 0-52+ adults,youngestchild 0-5one adult,youngestchild 6-152+ adults,youngestchild 6-15one adult,youngestchild 16-212+ adults,youngestchild 16-21one adult,retired, nochildren2+ adults,retired, nochildrenLife CycleMinutes 1983 2001 96 0%10%20%30%40%50%one adult, nochildren2+ adults, nochildrenone adult,youngestchild 0-52+ adults,youngestchild 0-5one adult,youngestchild 6-152+ adults,youngestchild 6-15one adult,youngestchild 16-212+ adults,youngestchild 16-21one adult,retired, nochildren2+ adults,retired, nochildrenLife CycleShare of Non-Travelers 1983 2001 Figure 5.40 Average Daily Travel Time per Person by Life Cycle Figure 5.41 Share of Non-Travelers by Life Cycle

PAGE 109

5.1.8 Average Daily Travel Time by Household Income Household income is often associated with the ability to participate in activities and travel. Data from the 1983 NPTS and 2001 NHTS suggest that travel times tend to increase as household incomes increases. For the purposes of this analysis, household incomes were grouped into quartiles. Since income data from the surveys was grouped into $5,000 divisions, the actual percentage of the quartile differs from 25 percent. These percentages are presented in Table 5.16, along with average travel times by income quartile. Figure 5.42 displays the average per person for each income group. In 1983, the income values associated with the quartile breaks was $15 thousand, $25 thousand, and $35 thousand. The income values associated with the quartile breaks in 2001 were $30 thousand, $50 thousand, and $80 thousand. Quartiles were used to compare the household travel times because the growth of household incomes and the way the data was recorded made it difficult to directly compare the household incomes. The share of non-travelers also tended to decrease as household income increased. The share of non-travelers for all income groups decreased more than 50 percent during the study period. The largest decrease occurred for the highest income groups. Figure 5.43 presents these results. Table 5.16 Summary of Daily Travel Time Expenditures by Household Income 1983 2001 Household Income Quartile Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration 1 59.7 39.1 34.2 16.5 89.9 75.5 30.0 20.8 2 64.7 48.2 23.9 16.3 91.3 81.6 24.8 19.6 3 67.9 51.6 18.6 17.1 95.2 86.8 24.4 19.7 4 72.3 55.3 23.4 17.6 99.1 91.5 20.7 20.2 97

PAGE 110

Figure 5.42 Average Daily Travel Time per Person by Household Income 0204060801001234Income QuartileMinutes 1983 2001 Figure 5.43 Share of Non-Travelers by Household Income 0%10%20%30%40%1234Income QuartileShare of Non-Travelers 1983 2001 98

PAGE 111

As discussed in Chapter 2, researchers have found conflicting results with regards to the effect of household income on daily travel time expenditures. This may be a result of the effect of household income on mode choice. Average daily travel times by mode and household income per person are compared in Table 5.17. As expected, travel time expenditures per person by private vehicle increase as household income increases. Travel times by private vehicle increase approximately 30 minutes per person for all income groups between 1983 and 2001. Mode shares also vary with income. Persons in lower income households tend to make more trips by public transportation and walk mode and fewer trips by private vehicle. The average time spend traveling on public transportation for a person in the first income quartile is more than double that of the other income groups. Many of these persons are considered captive riders to transit, because they are zero-vehicle households. The share and number of zero-vehicle has been decreasing in the United States. Relationships between household income and vehicle availability will be discussed in the following section. The effects of vehicle availability on travel time expenditures will also be discussed. Table 5.17 Summary of Daily Travel Time Expenditures by Household Income and Travel Mode 1983 2001 Mode Quartile 1 Quartile 2 Quartile 3 Quartile 4 Quartile 1 Quartile 2 Quartile 3 Quartile 4 Private Vehicle 29.0 40.4 43.7 45.4 59.5 70.9 75.1 77.1 Public Transportation 3.1 2.0 1.9 1.9 5.4 2.0 1.6 2.6 Bike 0.3 0.3 0.2 0.4 0.7 0.8 0.9 0.9 Walk 3.6 2.0 1.3 1.7 6.2 4.4 4.7 5.5 Other 2.1 2.2 2.5 3.0 3.5 3.4 4.3 5.4 99

PAGE 112

5.1.9 Average Daily Travel Time by Vehicle Availability The availability of a private vehicle to an individual significantly impacts that persons ability to travel. Table 5.18 presents the average minutes of travel per person and traveler, as well as the share of the population in each vehicle category. Figure 5.44 shows the average travel time per person from Table 5.18. Zero-vehicle households tend to have the lowest average daily travel times. The share of individuals without a vehicle has decreased since 1983. Average travel time per person tends to increase as vehicle availability increases. Households without a vehicle have the highest shares of non-travelers, while households with more vehicles have a lower share of non-travelers. These households are less likely to have more drivers than vehicles and therefore are not limited by vehicle availability to travel. The share of non-travelers decreases as the number of vehicles increase. This share has significantly decreased for all person groups from 1983 to 2001. Figure 5.45 presents the relationship between non-travelers and the number of vehicles in the household. Table 5.18 Summary of Daily Travel Time Expenditures by Number of Vehicles in Household 1983 2001 Number of Vehicles in Household Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration 0 57.6 31.2 9.5 18.4 97.9 74.2 5.2 26.9 1 63.0 44.3 27.4 16.6 90.6 77.7 21.0 19.7 2 65.2 49.6 36.6 16.4 91.3 81.6 41.7 19.5 3 69.0 51.5 16.4 17.2 94.5 85.6 19.8 20.2 4 or more 74.3 56.3 10.0 17.8 99.7 90.7 12.3 20.8 100

PAGE 113

Figure 5.44 Average Daily Travel Time per Person by Number of Vehicles 02040608010001234 or morNumber of VehiclesMinutes e 1983 2001 Figure 5.45 Share of Non-Travelers by Number of Vehicles 0%10%20%30%40%50%01234 or morNumber of VehiclesShare of Non-Travelers e 1983 2001 101

PAGE 114

The daily travel time expenditures by mode are correlated with the availability of a household vehicle. Persons in households without a vehicle tend to spend more time traveling by public transportation and walk modes as compared with persons in a household with at least one vehicle. Many of these persons are considered captive transit riders because they do not have a vehicle. Table 5.19 provides a summary of average travel times by mode and vehicle availability. Table 5.19 Summary of Travel Time Expenditures by Number of Vehicles and Travel Mode 1983 2001 Mode 0 1 2 3 4+ 0 1 2 3 4+ Private Vehicle 11.2 36.3 42.6 43.5 45.4 24.8 62.0 70.9 76.2 81.7 Public Transportation 9.6 2.3 1.1 1.4 1.6 26.9 4.6 1.1 0.9 1.1 Bike 0.4 0.3 0.3 0.3 0.2 0.9 0.6 0.9 0.7 0.7 Walk 8.2 2.7 1.5 1.2 0.8 15.1 6.6 4.4 4.0 3.5 Other 1.7 2.1 2.4 3.4 2.5 6.4 3.8 4.2 3.6 3.4 In many situations, the number of vehicles in a household is related to the household income. As household income increases, vehicle ownership tends to increase. However, most household tend to have two vehicles. As shown above, both of these factors have a positive effect on total travel time, especially private vehicle travel. Table 5.20 provides the share of the population by household income and vehicle ownership. Table 5.20 Share of Population by Income Quartile and Number of Vehicles 1983 Number of Vehicles 2001 Number of Vehicles Household Income Quartile 0 1 2 3 4+ 0 1 2 3 4+ 1 8.3 14.5 8.0 2.6 0.8 3.8 11.4 9.5 3.4 1.9 2 0.7 7.6 10.3 3.5 1.8 0.6 5.4 11.2 4.8 2.9 3 0.3 3.3 8.6 4.0 2.5 0.2 2.8 11.4 6.1 3.9 4 0.3 2.1 9.7 6.3 4.9 0.2 1.2 9.7 5.7 3.9 102

PAGE 115

5.1.10 Average Daily Travel Time by Driver Status Analysis of the relationship between driver status and daily travel time expenditures will be limited to persons over 16 years of age, because for most states this is the minimum age to acquire a license. Average daily travel times are higher for individuals with a drivers license than those without. Table 5.21 provides a summary of daily travel times by driver status in 1983 and 2001. Figure 5.46 presents the average travel time per adult person by driver status. Persons who are not licensed drivers also tend to account for higher shares of non-travelers. Figure 5.47 provides the share of each group that did not travel in 1983 and 2001. In 2001, less than 10.0 percent of drivers did not travel on the travel day. Both drivers and non-drivers had increases in the share of travelers. As discussed in the previous section, vehicle availability impacts total daily travel time. Persons over 16 years of age in household with the number of vehicles equal to or greater than the number of drivers have higher average daily travel time expenditures than persons in household with a vehicle shortage or less vehicles than drivers. Table 5.21 Summary of Travel Time Expenditures by Driver Status for Persons Over 16 Years of Age 1983 2001 Driver Status Average Minutes per Traveler Average Minutes per Person Share of16+ Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of16+ Population Average Time Per Trip with Reported Duration Licensed Driver 69.2 54.5 81.4 16.7 99.2 89.8 89.5 20.2 Not a Driver 58.3 32.0 15.3 19.0 87.2 60.0 10.5 24.5 103

PAGE 116

Figure 5.46 Average Daily Travel Time per Person by Driver Status 020406080100Licensed DriverNot a DriverDriver StatusMinutes 1983 2001 Figure 5.47 Share of Non-Travelers by Driver Status 0%10%20%30%40%50%DriverNot a DriverDriver StatusShare of Non-Travelers 1983 2001 104

PAGE 117

5.1.11 Average Daily Travel Time by Education Level Data from the 1983 NPTS and 2001 NHTS suggest that education level is related to travel behavior. The proportion of persons in each educational level changed dramatically from 1983 to 2001. Table 5.22 provides the shares of the population in each education level. In 2001, higher shares of the population had educational training after high school. This table also contains the average minutes of travel per traveler and person for each group. Each group increased over 30 minutes per person per day during the study period. The average trip duration for all groups was about 20 minutes, indicating that persons with higher education levels made more trips per day on average. Figure 5.46 visually presents the same information. Average daily travel time expenditures increase as education levels increase. The share of non-travelers also decreased during the study period. Figure 5.47 shows the share of non-traveler for each education level. The share of non-travelers is the lowest for persons with a graduate degree. Table 5.22 Summary of Travel Time Expenditures by Education Level 1983 2001 Education Level Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Less than High School Graduate 56.6 36.9 38.6 16.8 88.8 71.2 14.8 21.1 High School Graduate 66.2 49.1 31.8 16.8 94.2 81.4 30.7 20.7 Some College or Training 72.3 56.1 5.1 17.0 98.7 88.8 20.9 20.0 Associate's Degree 72.0 55.9 9.5 16.6 103.7 94.8 6.5 20.6 Bachelor's Degree 76.0 59.7 9.1 17.4 104.5 96.8 16.0 20.2 Graduate School 80.0 66.9 5.9 16.9 106.2 99.7 11.2 20.4 105

PAGE 118

Figure 5.48 Average Daily Travel Time per Person by Education Level 020406080100120Less than HighSchoolGraduateHigh SchoolGraduateSome Collegeor TrainingAssociate'sDegreeBachelor'sDegreeGraduateSchoolEducation LevelMinutes 1983 2001 Figure 5.49 Share of Non-Travelers by Education Level 0%10%20%30%40%50%Less than HighSchoolGraduateHigh SchoolGraduateSome Collegeor TrainingAssociate'sDegreeBachelor'sDegreeGraduateSchoolEducationShare of Non-Travelers 1983 2001 106

PAGE 119

5.1.12 Average Daily Travel Time by Race Average daily travel time expenditures for all races, except persons in the American Indian, Aluet, or Eskimo group, are between 81 and 83 minutes per person per day in 2001. Persons in the American Indian, Aluet, or Eskimo group tended to have higher travel times, but only represented 0.5 percent of the survey population. The summary of travel time expenditures by race is presented in Table 5.22. Figure 5.48 also shows the average travel time expenditures per person by race group. All groups, except for persons in the American Indian, Aluet, or Eskimo group, exhibited a growth of daily travel time expenditures of ranging from 34 minutes per person to slightly over 50 minutes per person in the other category. In 2001, race classifications were more disaggregate than those in 1983, so individuals with combinations of 2 or more races have been included in the other category. The share of non-travelers in each group was also similar between races in 2001. Figure 5.49 presents the share of non-travelers for each group in 1983 and 2001. All groups showed declines in the share of persons that did not travel. Table 5.23 Summary of Travel Time Expenditures by Race 107 1983 2001 Race Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration White 65.9 48.4 86.4 16.7 92.4 82.2 70.9 19.4 African American, Black 63.4 40.2 11.1 18.6 95.5 82.1 12.0 22.3 Asian, Pacific Islander 57.3 39.9 1.8 15.7 92.0 81.1 2.6 21.2 American Indian, Aluet, Eskimo 105.3 78.4 0.4 27.3 107.9 91.8 0.5 22.8 Other 57.0 32.5 0.3 16.4 95.0 83.1 14.0 22.1

PAGE 120

Figure 5.50 Average Daily Travel Time per Person by Race 020406080100WhiteAfrican American,BlackAsian, PacificIslanderAmerican Indian,Aluet, EskimoOtherRaceMinutes 1983 2001 Figure 5.51 Share of Non-Travelers by Race 0%10%20%30%40%50%WhiteAfrican American,BlackAsian, PacificIslanderAmerican Indian,Aleut, EskimoOtherRaceShare of Non-Travelers 1983 2001 108

PAGE 121

5.1.13 Average Daily Travel Time by Home Ownership Home ownership is not a significant factor as it relates to daily travel time expenditures. Table 5.23 provides a summary of travel time expenditures. Average travel time for owners and renters has increased about 35 minutes during the study period. The difference between average travel times per person in 2001 was only 0.6 minutes. Figure 5.52 presents the average travel time per person in 1983 and 2001. Table 5.24 Summary of Travel Time Expenditures by Home Ownership 1983 2001 Home Ownership Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Own 65.6 47.9 68.7 16.7 92.7 81.9 73.7 19.7 Rent 65.8 46.6 31.3 17.1 94.5 83.5 26.3 21.4 Figure 5.52 Average Daily Travel Time per Person by Home Ownership 020406080100OwnRentHome OwnershipMinutes 1983 2001 109

PAGE 122

5.1.14 Average Daily Travel Time by MSA Size The household location, with regards to metropolitan statistical area (MSA) size, could be expected to influence daily travel time expenditures. Average daily travel time per person increases slightly with the population size of the MSA. Table 5.25 provides a summary of the daily travel time expenditures for each MSA Size category. Figure 5.53 also shows the relationship between MSA size and average daily travel time per person. If only workers are considered, as expected the average travel time per person increases for each MSA size category, but the general relationship between travel time expenditures and MSA size remains the same. The highest average travel time expenditure per worker occurred in the 3 million or more MSA size group with an average travel time of 99 minutes per worker in 2001. The share of non-travelers decreased for every MSA size category from 1983 to 2001. The share of non-travelers does increase a small percentage as MSA size increases. Figure 5.54 shows the relationship of MSA size to the share of persons that did not travel on the travel day. Table 5.25 Summary of Travel Time Expenditures by MSA Size 1983 2001 MSA Size Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Average Minutes per Traveler Average Minutes per Person Share of Population Average Time Per Trip with Reported Duration Less than 250,000 57.9 44.7 11.0 14.2 89.5 79.8 7.0 18.2 250K to 499,999 66.5 49.9 10.1 16.8 87.4 77.7 8.2 18.7 500K to 999,999 64.4 47.8 9.7 16.2 88.0 78.7 7.7 18.7 1M to 2,999,999 64.0 45.5 21.3 17.0 91.4 81.5 21.0 19.5 3M or more 67.7 46.0 17.9 18.7 96.7 85.2 36.4 21.4 Not in MSA 68.9 49.5 30.0 17.1 94.5 82.0 19.7 20.4 110

PAGE 123

Figure 5.53 Average Daily Travel Time per Person by MSA Size 020406080100Less than250,000250K to499,999500K to999,9991M to2,999,9993M or moreNot in MSAMSA SizeMinutes 1983 2001 Figure 5.54 Share of Non-Travelers by MSA Size 0%10%20%30%40%50%Less than250,000250K to499,999500K to999,9991M to2,999,9993M or moreNot in MSAMSA SizeShare of Non-Travelers 1983 2001 111

PAGE 124

5.2 Conclusions This chapter has described average daily travel time expenditures from the 1983 NPTS and 2001 NHTS. Travel times were compared based on mode, trip purpose, and many socio-demographic characteristics and area characteristics. Travel time expenditure per person and travel increase for every characteristic considered. Therefore, the growth can not be attributed to one particular group. Based on the travel group comparisons, the group with the lowest travel time expenditures would stereotypically be a retired female over 85 years of age that is not a driver and has a low income. The group with the highest daily travel time would be a male, worker with a high income between 35 and 44 years of age, without children and has a vehicle. The following chapter will explore opportunities to quantify the change in average daily travel time using socio-demographic characteristics. 112

PAGE 125

CHAPTER 6 DISCUSSION OF QUANTITATIVE ANALYSIS Daily travel time expenditures can be estimated as a function of trip frequency, distance, and speed. Based upon the descriptive analysis in Chapter 5, the increase of trip frequency, particularly for family and personal business trips, has been one of the main contributing factors to the increase in average travel time per person. Between 1983 and 2001, persons on average added an additional 1.3 trips per day. One could argue that trip frequency can be estimated using person characteristics, while distance and speed may be related to the area and system characteristics. However, the interactions between these variables and other factors, including human behavior, make estimating travel time a daunting task. Several avenues for attempting to explain the growth of travel times were explored for analysis. As the descriptive analysis focuses on socio-demographic characteristics, the potential power of these variables to explain travel time growth was the primary focus. 6.1 Share Analysis One consideration for the growth in travel time expenditures is how the population distributions by socio-demographic characteristics have changed. This share analysis compares what travel times would be in 1983 if average travel times per group remained the same, but the distribution of the population were as in 2001. The difference between the actual mean travel time and the estimated average in 1983 indicates the 113

PAGE 126

amount of the increase in total travel time that could potential be explained by the change in distribution. Of course, many of the socio-demographic characteristics are cross correlated, such as age, worker status, and income. Therefore, the share of the travel time increase that could be explained would likely be less than the total share of all of the components. Regardless of this the share of the increase that can be explained by socio-demographic changes is relatively small. Table 6.1 provides the share distribution analysis completed for several variables in this study. Of these variables, the education level of the population appears to have the most relevance regarding the increase in daily travel time. Table 6.1 Average Minutes of Travel per Person Based on Population Distributions Variable Actual Average Minutes per Person 1983 Estimated Average Minutes per Person in 1983 with 2001 Population Distributions Actual Average Minutes per Person 2001 Difference between Estimated and Actual Average Travel Time in 1983 Share of Travel Time Increase Explained Age Group 48.6 48.1 82.2 -0.5 -1.5% Worker Status (Adult) 50.8 52.4 87.2 1.6 4.4% Life Cycle 46.9 45.7 78.4 -1.2 -3.8% Household Income (CPI Adjusted) 47.4 48.7 83.1 1.3 3.6% Number of Vehicles in Household 47.3 48.7 82.3 1.4 43.0% Driver Status (Over 16) 49.3 52.1 86.7 2.8 7.5% Education Level 47.4 52.9 86.9 5.5 13.9% Race 47.4 45.1 82.3 -2.3 -6.6% Home Ownership 47.5 47.6 82.3 0.1 0.3% MSA Size 47.4 47.0 82.3 -0.4 -1.1% Total per Person 47.4 82.3 114

PAGE 127

6.2 Multivariate Model In an attempt to capture the interactions between variables, modeling applications were explored. A multivariate linear regression model was briefly considered. Based on preliminary attempts at developing this model, several variables were found to be significantly related to travel times including the age of an individual, classified as young or elderly, worker status, income, vehicle availability, and education level. However, the potential power of these variables to explain travel times was very weak based on the R squared value. The signs of the coefficients were as expected with young and elderly persons traveling less than middle age groups, workers traveling more than non-workers, and positive relationships between education, income, and vehicle availability and travel time. Further pursuit of this model were abandoned, because based on the descriptive analysis and share analysis the potential of the model to explain the increase seemed remote. 6.3 Travel Time Forecast Given that the socio-demographic and area characteristics do not seem to explain the increase in daily travel time, the prospect of forecasting travel times into the future based on such characteristics also was determined to be a fruitless endeavor. The future growth of daily travel time expenditures is dependant on a multitude of factors including technological and social changes in the United States, as well as peoples willingness to spend additional time traveling instead of time devoted to other activities. The future of the transportation system, including roadway conditions, congestion, and speed, and the cost of travel will also determine the amount of travel in the future. Projecting these 115

PAGE 128

conditions into the future holds great uncertainty, because no one knows for sure what changes will occur in the next few years that many impact travel behavior. Conclusions and future research opportunities regarding travel times will be discussed in the following chapter. 116

PAGE 129

CHAPTER 7 CONCLUSIONS AND FUTURE RESEARCH 7.1 Summary of Findings The increase of average daily travel time per person and per traveler is contrary to the concept of a travel time budget. This increase may indicate a positive utility of travel time or that the benefits of more and longer trips have increased, while the cost of travel has decreased. It is important to note that the share of trips that are complete entirely for the sake of travel is small and that most travel is a derived demand and sought for the utility of activity participation. The current pattern of travel time growth has not shown any signs of dampening, but at some point in the future one would expect continued growth to slow as the costs of additional travel outweigh the benefits. The majority of travel time growth is based on higher levels of travel for each individual. This has been enabled by several cultural trends, included fewer children to care for and smaller household size; specialization of activities, such as eating out versus cooking at home; increased female labor force participation rates; multitasking during travel, for example cell phone use can add value to travel time; seeking socialization away from home; and increases in real income enabling greater activity participation. Small changes in a variety of areas can add up to significant changes in overall travel time expenditures. Increased mobility of the population has also contributed to the increase of travel times. The share of non-travelers decreased from 27.9 percent in 1983 to 11.7 percent in 117

PAGE 130

2001. Travel times per traveler also increased by 42 percent from 65.7 minutes per traveler in 1983 to 93.2 minutes per traveler in 2001. The true mobility of the population can not be assessed from this survey, because of the single travel day in the survey. The results of the share analysis and multivariate model suggest that most of the growth in average daily travel time has not resulted from changes in the distribution of the population by various socio-demographic characteristics, but instead maybe a result of a combination of factors including changes in society, technology, and attitudes. Several variables were found to be significantly related to travel time including age, employment status, and education, but these variables did not explain much of the increase. Perhaps this finding, that socio-demographic characteristics are not the key factor in explaining travel time expenditures, is the most significant result of this study, because it implies that methods used to estimate travel demand may need to be reevaluated. The progression of researchers to study life style groups could capture some of the variation between individuals with similar socio-demographic characteristics, but significantly different attitudes and activity participation patterns. Different life style groups may have different factors that have contributed to the increase in overall travel time expenditures and drive travel demand. Many of the variables studied have been traditionally used in trip generation modeling including vehicle availability, household income, licensing, number of workers, and household size. The results of this study suggest that perhaps the significance of these variables in explaining travel demand has played itself out. Therefore, some of the assumptions used in trip generation modeling may no longer be valid and need to be 118

PAGE 131

reevaluated such as static trip rates. One possible improvement to the trip generation model could be to inflate the trip rates in the future given the historic trend. This method also contains uncertainty, but may offer improvements over a static trip rate. Further research into this area is needed. The majority of the growth in daily travel time expenditures is a result of increased trip rates. While trip distance has slightly increased, trip frequency has increased by 1.3 trips per person per day. Increases in family and personal business trips accounted for 0.8 trips per person per day. Changes in the economy have resulted in increases of these types of trips, as Americas purchase more goods and services that historically were completed at home. Cultural expectations have shifted as average income and auto availability in the United States have increased. Many Americans feel that there are no limits to what can be purchased. The future growth of daily travel time expenditures is dependant on a multitude of factors including technological and social changes in the United States, as well as peoples willingness to spend additional time traveling instead of time devoted to other activities. Estimations of vehicle miles of travel will be dependant on the populations eagerness to spend additional time traveling. If travel time expenditures continue to grow, the hope for slowing VMT growth may not materialize. The increase in travel time expenditures will impact the performance of the transportation system going forward. The future of the transportation system, including roadway conditions, congestion, and speed, and the cost of travel will also determine the amount of travel in the future. Projecting these conditions into the future holds great uncertainty, because no one knows for sure what changes will occur in the next few years that many impact travel behavior. 119

PAGE 132

The impact of the aging baby boom generation could be significant in shaping travel demand in the future. As will be discussed in the following section, there is still a great deal to be learned about travel time expenditures. 7.2 Future Research Continued research in the field of travel time expenditures will aid transportation planners and modelers in understanding the mechanics of factors that generate travel demand. Travel time expenditures are perhaps the most difficult aspect of travel demand to predict because so many factors are involved in determining these. Ultimately travel time expenditures are determined by the individual, who determines what trips to make and how to make them based on factors that are difficult to model. Further research into human behavior through fields such as psychology, biology, and sociology will provide insight into travel behavior that may not be capture in traditional transportation surveys. Key areas that could be studied include tolerances for congestion, impacts of changing household makeup on travel, changes in attitudes regarding travel, etc. Continued study of activity time allocation trends will provide an overview of how individual choose to spend their time. Also research on land use trends, specialization of the labor force, the condition of the roadway system network, safety and comfort of private vehicles, time deepening techniques, economic conditions, the cost of travel, and advancements in information technology will increase the knowledge of travel time expenditure trends. Transportation surveys are designed to attempt to capture certain aspects of travel behavior. However, due to the nature of these surveys, many critical factors and trends are difficult to identify. Additional insight into travel time expenditures could be gained 120

PAGE 133

through a survey that gathered travel behavior over several days or a week so that regularities in personal travel time expenditures could possibly be captured. Such a survey is difficult to conduct due to the cost and time required to obtain such information from a sample that is willing to participate. A survey of the same sample group over time could also be useful to understanding travel time expenditures as it may provide some insight into the causality of changes in travel time. Comparisons with other countries could provide perspectives on how different policies impact travel behavior and time expenditures. Another area closely tied to travel time expenditures is money expenditures for travel. While not discussed as part of this thesis, much research has been done on this issue, but additional focus on the relationships between travel time and money expenditures could be constructive. As discussed in the previous chapter, the ability to quantify the change of average travel time expenditures is critical to understanding the key factors involved in determining travel demand. While socio-demographics seem to explain only a small share of travel time expenditures, additional analysis from the national travel survey series and other surveys may result in significant findings of the factors involved. This issue should also be revisited when new survey data becomes available to determine if this trend has continued or if there are signs of moderation. Additional focus on the extent and impact of trip chaining, trip distance, trip frequency, travel mode, and speed could aid modelers as they attempt to forecast travel demand into the future. Additional research in these areas and possible many other areas will provide a better understanding of travel time expenditures and the utility of travel. 121

PAGE 134

REFERENCES Bhat, C. and F. Koppelman (1999) Activity-Based Modeling of Travel Demand. In: Handbook of Transportation Science. R.W. Hall (Ed) Kluwer, Dordrecht. pp. 35-62. Cevero, R. (1996) Mixed Land Uses and Commuting: Evidence from the American Housing Survey. Transportation Research A 30(5). pp. 361-377. Chen, C. and P. L. Mokhtarian (2002) Constrained Allocation of Time and Money between Activities and Travel: A Review of Modeling Methodologies and a New Utility Maximization Model. Institute of Transportation Studies, University of California, Davis. Cowherd, K. (2004) Hardwork Can Give Us Title for Slow Commute. In the Baltimore Sun. March 1. DKS Associates (1994) Travel Model Development and Refinement Trip Generation. Prepared for Puget Sound Regional Council. Downes, J.D. and D. Morrell (1981) Variation of Travel Budgets and Trip Rates in Reading. Transportation Research A 15. pp. 47-53. Erwing, R. and R. Cervero (2001) Travel and the Built Environment: A Synthesis. Transportation Research Record 1780. pp. 87-114. Erwing, R.; Haliyur, P. and W. Page (1994) Getting Around the Traditional City, a Suburban Planned Unit Development, and Everything Between. Transportation Research Record 1466. pp. 53-62. Evans, A.W. (1972) On the Theory of the Valuation and Allocation of Time. Scottish Journal of Political Economy. pp. 1-17. Frank, L. and G. Pivo (1994) Impacts of Mixed Use and Density on Utilization of Three Modes of Travel: Single-Occupant Vehicle, Transit and Walking. Transportation Research Record 1466. pp. 44-52. Golob, T. (1990) The Dynamics of Household Travel Time Expenditures and Car Ownership Decisions. Transportation Research A 24(6). pp. 443-463. 122

PAGE 135

Golob, T. and M. McNally (1997) A Model of Activity Participation and Travel Interactions between Household Heads. Transportation Research B 31(3). pp. 177-194. Goodwin, P. B. (1981) The Usefulness of Travel Budgets. Transportation Research A 15. pp. 97-106. Gordon, P.; Kumar, A. and H. Richardson (1988) Beyond the Journey to Work. Transportation Research A 22. pp. 419-426. Gordon, P.; Kumar, A. and H. Richardson. (1989) The Influence of Metropolitan Spatial Structure on Commuting Time. Journal of Urban Economics 26. pp. 138-151. Gordon, P.; Ricardson, H. and M. Jun (1991) The Commuting Paradox Evidence from the Top Twenty. Journal of the American Planning Association 57(4). pp. 416-420. Gunn, H. F. (1981) Travel Budgets: A Review of Evidence and Modeling Implications. Transportation Research A 15. pp. 7-23. Hamed, M. and F. Mannering (1993) Modeling Travelers Postwork Activity Involvement: Toward a New Methodology. Transportation Science 27(4). pp. 381-394. Handy, S. (1996) Methodologies for Exploring the Link between Urban Form and Travel Behavior. Transportation Research D 1(2). pp. 151-165. Hupkes, G. (1982) The Law of Constant Travel Time and Trip-Rates. Futures February. pp. 38-46. Kitamura, R.; Robinson, J.; Golob, T.; Bradley, M.; J. Leonard and T. van der Hoorn (1992) A Comparative Analysis of Time Use Data in the Netherlands and California. Research Report Number UCD-ITS-RR-92-9, Institute of Transportation Studies, University of California, Davis, June. Kitamura, R.; S. Fujii and E. Pas (1997) Time-use Data, Analysis and Modeling: Toward the Next Generation of Transportation Planning and Methodologies. Transport Policy 4(4). pp. 225-235. Landrock, J. (1981) Spatial Stability of Average Daily Travel Times and Trip Rates within Great Britain. Transportation Research A 15. pp. 55-62. Levinson, D. and S. Kanchi (2002) Road Capacity and the Allocation of Time. Journal of Transportation and Statistics 5(1). pp. 25-46. Levinson, D. and A. Kumar (1996) Activity, Travel, and the Allocation of Time. Journal of the American Planning Association Fall. pp. 458-470. 123

PAGE 136

Levinson, D. and A. Kumar (1994) The Rational Locator: Why Travel Times Have Remained Stable. Journal of the American Planning Association Summer. pp. 319-332. Lin, B.H.; E. Frazao and J. Guthrie (1999) Away-from-home Foods Increasingly Important to Quality of American Ddiets. Agricultural Information Bulletin no 749, U.S. Department of Agriculture and U.S. Department of Health and Human Services. Lu, X. and E. Pas (1999) Socio-demographics, Activity Participation and Travel Behavior. Transportation Research A 33. pp. 1-18. Macek, N.; Khattak, A. and R. Quercia. (2001) What is the Effect of Commute Time on Employment? Analysis of Spatial Patterns in New York Metropolitan Area. Transportation Research Record 1780. pp. 43-52. McCormack, E; Rutherford, G.S. and M.G. Wilkinson (2001) Travel Impacts of Mixed Land Use Neighborhoods in Seattle, Washington. Transportation Research Record 1780. pp. 25-32. Mokhtarian, P. and C. Chen (1999) A Review and Discussion of the Literature on Travel Time and Money Expenditure. Research Report prepared for the DiamlerChrysler Corporation, Institute of Transportation Studies, University of California, Davis. Mokhtarian, P. and C. Chen (2002) TTB or Not TTB, That is the Question: A Review and Analysis of the Empirical Literature on Travel Time (and Money) Budgets. Institute of Transportation Studies and Department of Civil and Environmental Engineering, University of California, Davis. Mokhtarian, P. and I. Salomon (1999) Travel for the fun of it. Access 15. pp. 20-31. Mokhtarian, P.; Salomon, I. and L. Redmond (2001) Understanding the Demand for Travel: Its Not Purely Derived. Innovation 14(4). pp. 355-380. Noland, R. B. and L. L. Lem (2002) A Review of the Evidence for Induced Travel and Changes in Transportation and Environmental Policy in the U.S. and the U.K. Transportation Research D 7. pp. 1-26. Pas, E. I. (1998) Time in Travel Choice Modeling: From Relative Obscurity to Center Stage. In: Theoretical Foundations of Travel Choice Modeling. Tommy Garling, Thomas Laitila, and Kerstin Westin (Eds.). Elsevier, New York, pp. 231-250. Pas, E. I. (2002) Time Use and Travel Demand Modeling: Recent Developments and Current Challenges. In Perpetual Motion: Travel Behavior Research Opportunities and Application Challenges. H.S. Mahmassani (Ed.) pp. 307-331. 124

PAGE 137

Pendyala, R.M. (2003) Time Use and Travel Behavior in Space and Time. In K. G. Goulias (Ed.) Transportation Systems Planning: Methods and Applications. CRC Press, Boca Raton, FL. pp. 2-1 2-37. Peters, P.; de Wilde, R.; Clement, B. and P. Peeters (2001) A Constant on the Move? Travel Time, Virtual Mobility and Constant Travel Time Budgets. Universiteit Maastricht, Faculteit der Cultuurwetenschappen. Polzin, S.; Chu, X. and L. Toole-Holt (2003) The Case for Moderate Growth in Vehicle Miles of Travel: A Critical Juncture in U.S. Travel Behavior Trends. Center for Urban Transportation Research. Prendergast, L. and R. Williams (1981) Individual Travel Budgets. Transportation Research A 15. pp. 39-46. Principio, S. and E. I. Pas (1997) The Sociodemographics and Travel Behavior of Style Groups Identified by Time Use Patterns. Paper Presented at the 76 th Annual Transportation Research Board Meeting, Washington, D.C. Purvis, C. (1994) Changes in Regional Travel Characteristics and Travel Time Expenditures in San Francisco Bay Area: 1960-1990. Transportation Research Record 1466. pp. 99-109. Robinson, J.; Converse, P.E. and A. Szalai (1972) Everyday Life in Twelve Countries. In: The Use of Time. A. Szalai (Ed.). Mouton, The Hague. Robinson, J. (1999) Time for Life. The Surprising Ways Americans Use Their Time. The Pennsylvania State University Press, University Park, Pennsylvania. Roth, G. and Y. Zahavi (1981) Travel Time Budgets in Developing Countries. Transportation Research A 15. pp. 87-95. Salomon, I. (1983) Life Styles: A Broader Perspective on Travel Behavior. In Recent Advances in Travel Demand Analysis. S. Carpenter and P. Jones (Eds.) England: Gower publishing Co. LTD. pp. 290-310. Schafer, A. (2000) Regularities in Travel Demand: An International Perspective. Journal of Transportation and Statistics. pp. 1-31. Schrank, D. and T. Lomax (2003) The 2003 Annual Urban Mobility Report. Texas Transportation Institute, The Texas A&M University System. Supernak, J. (1982) Travel Time Budget: A Critique. Transportation Research Record 879. pp. 15-25. 125

PAGE 138

126 Tanner, J.C. (1961) Factors Affecting the Amou nt of Travel. Road Research Technical Paper No. 51. London:HMSO. Tanner, J.C. (1981) Expenditure of Time and Money on Travel. Transportation Research A 15. pp. 25-36. Van der Hoorn, T. (1979) Travel Beha vior and the Total Activity Pattern. Transportation 8. pp. 309-328. Wee, B.; Rietveld, P. and H. Meurs (2002) A Constant Travel Time Budget? In Search for Explanations for an Increase in Average Travel. Research Memorandum 2002-31. Universiteit Amsterdam. Wen, C.H. and F.S. Koppelman (1999) Integr ated Model System of Stop Generation and Tour Formation for Analysis of Activity and Travel Patterns. Transportation Research Record 1676. pp. 136-144. Zahavi, Y. (1979) The UMOT Project. Report No. DOT-RSPA-DPB-20-79-3. U.S. Department of Transportation, Research and Special Programs Administration, Washington, D.C., and the Fed. Republic of Germany Min. of Transport, Bonn. Zahavi, Y. (1981) The UMOT-Urban Inte ractions. Report No. DOT-RSPA_DPB 10/7. U.S. Department of Trans portation, Washington, D.C. Zahavi, Y. and J. Ryan (1980) Stability of Travel Components over Time. Transportation Research Record 750. pp. 19-26. Zahavi, Y. and A. Talvitie (1980) Regularities in Travel Time and Money Expenditures. Transportation Research Record 750. pp. 13-19. 1983 Nationwide Personal Transportation Survey: Users Guide. Retrieved May 20, 2004, from http://www.fhwa.dot.gov/ohim/1983/1983page.htm 2001 National Household Travel Survey: User s Guide. Retrieved May 20, 2004, from http://nhts.ornl.gov/2001/ usersguide/index.shtml