Long-Distance and Rural Travel Transferable Parameters for Statewide Travel Forecasting Models (2012)

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

B-1 This appendix presents a summary of Canadian, European, and other data sources that document long-distance travel behavior. There are five primary sources of Canadian personal travel data that were evaluated: the Canadian Census, provincial travel surveys, the Travel Survey of Residents of Canada, Canadian Travel Survey, and the Rural and Small Town Canada Analysis Bulletin. This appendix also discusses Canadian tourism-related surveys and European studies on long-distance travel. For the purposes of this report, the research team analyzed readily available documentation on rural and long-distance travel from these sources, in order to both document what data might be available in the development of rural and long-distance parameters as well as whether the data were suitable for such an application. It should be noted that some of the Canadian data appears to be well suited to support the development of long-distance travel parameters. However, most are only available for a fee. The appendix concludes with a summary of studies regarding long-distance travel that were conducted outside of North America. B.1 Canadian Census Similar to the U.S. Census, the Canadian Census obtains details about workplace and typical mode to work, which can be used to estimate commute work flows, distance traveled, and travel modes for the commute trip. The Canadian Census is conducted every 5 years. The most recent Census was conducted in 2006, and the next one is scheduled for May 2011 (http://census2011. gc.ca/ccr_r000-eng.htm). Based on a preliminary assessment of publicly available data summa- ries, the 2006 Census does not appear to contain sufficient records with long-distance commutes to aid in the development of transferable model parameters for this project. However, the cover- age in rural areas of the country appears to provide sufficient observations of rural commuters to inform the development of rural parameters. According to publicly available data summaries from the 2006 Census, 1,376,340 of the 13,069,895 commuters (or approximately 10 percent) reported a commute distance of 30 kilo- meters (18.64 miles) or more. (For a fee, the disaggregate data is available to further delineate the commute distances and also to assess differences in urban/rural commute characteristics.) Average commute distance by age is shown in Figure B.1. As in the United States, commuters travel to work predominantly by auto, as shown in Figure B.2. However, use of transit is higher in Canada than what is observed in the United States, as indicated in Figure B.3. From available documentation the differences in travel A p p e n d i x B Travel Behavior Data from Other Countries

B-2 Long-distance and Rural Travel Transferable parameters for Statewide Travel Forecasting Models mode by urban versus rural commuters and the extent of public transit trips for long-distance travel are not clear. B.2 Provincial Travel Surveys Canadian officials conduct travel surveys to document travel behavior and to develop regional travel demand models. As is typical in the United States, these surveys document long-distance travel as it occurs naturally during the assigned travel period. Two of the better-known provincial travel surveys are the Transportation Tomorrow Survey in Toronto and the Greater Vancouver Figure B.1. Average commute distance by age—2006 Canadian Census. Source: 2006 Canadian Census. 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Total - Age groups 15 to 24 years 25 to 54 years 25 to 34 years 35 to 44 years 45 to 54 years 55 to 64 years 65 to 74 years 75 years and over Less than 5 km 5 to 9.9 km 10 to 14.9 km 15 to 19.9 km 20 to 24.9 km 25 to 29.9 km 30 km or more Source: 2006 Canadian Census. 72.3% 7.7% 11.0 % 6.4% 1.3% 0.1% 0.2% 0.9% Car, truck, van as driver Car, truck, van as passenger Public transit Walked Bicycl e Motorcycle Taxicab Other method Figure B.2. Commute by transportation mode—2006 Canadian Census.

Travel Behavior data from Other Countries B-3 Trip Diary Survey. Based on a preliminary review of the results of these surveys, neither appears to have sufficient observations of rural households or long-distance trips to support the develop- ment of transferable model parameters. Transportation Tomorrow Survey The Transportation Tomorrow Survey (TTS) is an ongoing data collection program in the Toronto metropolitan region (http://www.dmg.utoronto.ca/transportationtomorrowsurvey/ index.html). It is sponsored by the Transportation Information Steering Committee, which includes 21 local, regional, provincial, and transit agencies in the Greater Toronto area. The sur- vey itself is conducted by the Data Management Group, which is associated with the Department of Civil Engineering at the University of Toronto. The survey was first conducted in 1986, with adjustments in methods and administration over time. The most recent survey, in 2006, documented demographic and travel behavior details for 149,000 households in the greater Toronto area. Travel behavior was reported for a 24-hour weekday for all household members ages 11 and older. The number of data elements obtained in this survey is less than what is typically obtained in U.S. surveys, focusing only on key attributes needed for modeling (http://www.dmg.utoronto.ca/pdf/tts/2006/regional_travel_summaries/ TTS_report4_full.pdf), as follows: • Household Characteristics—Location of residence, dwelling type, household size, and household vehicles; • Person Characteristics—Age, gender, employment and student status, possession of a driver’s license and transit pass, location of work and school, availability of parking at work, occupa- tion, and whether the worker worked at home on the travel day; • Travel Information—Nature of trip (start time, trip purpose, origin and destination) and means of travel (travel mode and detailed transit routes); and • Trip—defined as a one-way movement between two locations for a single purpose. Within the TTS sampling area, four sampling regions had fewer than 100 persons per square kilometer (ppsk): Peterborough (35 ppsk), Dufferin (37 ppsk), Wellington (75 ppsk), and Simcoe Sources: American Community Survey, 2006; and Statistics Canada, Census of Population, 2006. Figure B.3. Proportion of workers using public transportation for work—U.S. versus Canada for select metropolitan areas.

B-4 Long-distance and Rural Travel Transferable parameters for Statewide Travel Forecasting Models (87 ppsk). Of the 149,631 households that participated in the 2006 TTS, 12,586 reside in these four sampling areas (http://www.dmg.utoronto.ca/pdf/tts/2006/expansion2006.pdf). Depend- ing on how “rural” is defined for this study, the TTS may be useful in the development of rural travel parameters. Long-distance travel was not explicitly requested as part of the TTS. However, reported average trip lengths by mode suggest that some long-distance trips were captured: auto drivers at 5.6 kilometers, auto passengers at 4.1 kilometers, transit users at 6.3 kilometers, and train riders at 30.2 kilometers. Although median, minimum, and maximum values are not reported, the TTS does not appear to include sufficient observations of long-distance trips to be useful in this effort. Greater Vancouver Trip Diary Survey In 2008, the British Columbia Ministry of Transport and the Greater Vancouver Transporta- tion Authority funded the Regional Trip Diary Survey (http://info.nathanp.org/Reports, scroll to Transportation Section for 2008 TransLink Trip Diary Survey Report.pdf). The survey was conducted by the Mustel Group and Halcrow. Prior surveys were most recently conducted in this region in 1994, 1999, and 2004. The 2008 survey was designed to document 24-hour weekday travel behavior characteristics of regional residents. A total of 17,603 households agreed to participate in the effort, reporting a total of 92,187 trips. The study was conducted in the fall of 2008, using a combination of online and mail-out/telephone retrieval methods to obtain the following: • Household Characteristics—Household size, household vehicles, and household location; • Person Characteristics—Age, gender, driver’s license status, public transit usage and fare pay- ment, employment status and related employment information, student status and related school information; and parking at work or school; and • Travel Information—Number of trips, trip purpose, travel mode, travel party size if traveled by auto, time of day of travel, origins and destinations of travel, and land use at destination. The final report documents an overall 24-hour average trip length of 9.3 kilometers (or 5.78 miles). Average trip lengths by travel mode and trip purpose are shown in Figures B.4 and B.5 based on kilometers. As noted in both, the short trip lengths suggest that these data would not be useful in constructing the necessary long-distance parameters. In addition, Source: TransLink’s 2008 Regional Trip Diary Survey – Final Report, Exhibit 3.11a. 10.7 7.6 12 .0 5.3 2. 0 9. 9 9. 3 0. 0 2. 0 4. 0 6. 0 8. 0 10.0 12.0 14.0 Auto-D Auto-P Transit Bike Walk Other Total 24-hour Average Trip Length (km) Figure B.4. Vancouver survey trip length by mode.

Travel Behavior data from Other Countries B-5 details regarding the population of the study area suggest that the study area is predominantly urban and would not have sufficient rural cases to support the rural analysis. B.3 Travel Survey of Residents of Canada The Travel Survey of Residents of Canada (TSRC) is a supplemental survey to the Canadian Labour Force Survey, which is a monthly survey of approximately 54,000 households regarding employment levels. (Unless otherwise noted, the details in this section come from http://www. statcan.gc.ca/cgi-bin/imdb/p2SV.pl?Function=getSurvey&SDDS=3810&lang=en&db=imdb& adm=8&dis=2) The TSRC was initiated in 2005, and by 2009, there were about 7,000 responses per month with each month documenting about 3,500 trips. The study is sponsored by Statistics Canada, the Canadian Tourism Commission, provincial governments, and federal organizations. It is designed to measure the size and status of Canada’s tourism industry at the national level. As such, it measures the volume, characteristics, and economic impact of domestic travel. A precursor to the TSRC is the Canadian Travel Survey, which is summarized in the next section of this appendix. “The objectives of the [TSRC] survey are to provide information about the volume of trips and expenditures for Canadian residents by trip origin, destination, duration, type of accom- modation used, trip reason, mode of travel, etc.; to provide information on travel incidence and to provide the sociodemographic profile of travelers and nontravelers.” Data obtained through the TSRC that are potentially relevant to the development of transferable parameters for long-distance travel include • Total volume of same day and overnight trips taken by the residents of Canada with destina- tions in Canada; • Main purpose of the trip/key activities on trip; • Modes of transportation (main/other) used on the trip; • Use of travel packages and associated spending and use of motor coach/other guided tours; • Demographics of adults that took or did not take trips; and • Travel party composition. Source: TransLink’s 2008 Regional Trip Diary Survey – Final Report, Exhibit 3.11a. Note: PS = Post-secondary school. 14 .1 13 .9 11.2 4. 7 4.6 7.4 6. 9 9. 3 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 24-hour Average Trip Length (km) To Work/ PS From Work/PS During Work To Grade/ Other School From Grade/ Other School Recreation/ Dining/ Shopping Personal Business/ Other Total Figure B.5. Vancouver survey trip length by purpose.

B-6 Long-distance and Rural Travel Transferable parameters for Statewide Travel Forecasting Models The long-distance trips are obtained for the month prior to the TSRC interview. Figure B.6 shows the question series used to obtain the number of long-distance trips. The TSRC is administered predominantly by CATI. A small portion of interviews are con- ducted using CAPI. The sample consists of a cross-sectional subset of those who completed the Canadian Labour Force Survey. The interviews are largely conducted with the sampled house- hold member; proxy reporting is permitted only under very strict conditions. In Canada, the Labour Force Survey is mandatory, but the TSRC is voluntary. According to Statistics Canada, the Labour Force Survey has a 90 percent participation rate, while the TSRC has a “slightly lower response rate of 80 percent with a travel incidence rate of about 30 percent” (http://www.statcan. gc.ca/imdb-bmdi/document/3810_D3_T9_V1-eng.pdf). The high response rates, combined with an apparent low level of proxy reporting suggests that the travel documented therein is fairly representative of the Canadian long-distance travel market. Data summaries are available, as well as the dataset itself, for varying costs. B.4 Canadian Travel Survey The Canadian Travel Survey (CTS) is the precursor to the Travel Survey of Residents of Can- ada (TSRC). This survey was conducted as a supplemental survey to the Canadian Labour Force Survey. It was initiated in 1979, became an ongoing survey in 1980, and was last administered in 2005 as a quarterly survey (http://www.statcan.gc.ca/imdb-bmdi/document/3810_D2_T9_ V1-eng.pdf). The study was sponsored by Statistics Canada, the Canadian Tourism Commission, provincial governments, and federal organizations. Like the TSRC, it was designed to measure OD_R01 Please think about all of the out-of-town trips you took that ended last month, that is in (reference month), (reference year). The trips could have been for visiting friends or relatives, for pleasure, vacation or holiday, for personal or business reasons. OD_Q01 Did you take any out-of-town trips of one or more nights away from home that ended last month, that is in (reference month), (reference year)? 1 Yes 2 No .................................................................................................... (Go to OD_Q03) DK, RF ................................................................................................ (Go to OD_Q03) Coverage: All respondents. OD_Q02 How many? (Overnight trips) ____(2 spaces) [Min: 1 Max: 31] DK, RF Coverage: Respondents who took at least one overnight trip. OD_Q03 Did you take any same day out-of-town trips of at least 40 kilometers one-way last month? INTERVIEWER: If the respondent is not sure, explain that we mean trips to any destinations that are located 40 kilometers or more from their place of residence and in which they left and returned home on the same day. Forty kilometers is equivalent to about 25 miles. We are interested in one-way distance only. 1 Yes ................................................................................................... (Go to OD_Q04) 2 No DK, RF ................................................................................................ (Go to OD_C05) Coverage: All respondents. OD_Q04 How many? (same-day trips) ____(2 spaces) [Min: 1 Max: 40] DK, RF Source: http://www.statcan.gc.ca/imdb-bmdi/instrument/3810_Q3_V7-eng.pdf Figure B.6. TSRC questions to document long-distance travel.

Travel Behavior data from Other Countries B-7 the size and status of Canada’s tourism industry and measured the volume, characteristics, and economic impact of domestic travel. (Unless otherwise noted, the summary for this section comes from http://www.statcan.gc.ca/imdb-bmdi/document/3810_D1_T7_V1-eng.pdf) The CTS obtained long-distance travel details for all-day trips of 80 kilometers (approximately 50 miles) or longer (one-way only), except for Ontario residents, who reported on details for all-day trips of 40 kilometers or longer. Trips involving at least an overnight stay (or longer) were documented regardless of length. To be recorded, the trip must have taken place in the referenced month. Trips excluded: “commutes between home and work or school; one-way moves to a new residence; trips made by members of the operating crew of a bus, plane, truck, etc.; ambulance rides to hospitals or clinics; trips originating outside Canada; and trips lasting more than 1 year.” Table B.1 shows the main differences in recorded trips in the CTS and its replacement, the TSRC. The following are statistics from the 2004 CTS (the most recent summary document located) (http://www.statcan.gc.ca/pub/87-212-x/87-212-x2004000-eng.pdf, Tables 2.2, 2.3, 2.5, 2.7). These statistics reflect travel from throughout the year, including summer months, and are avail- able in a PUMS-like format for use in estimating long-distance parameters. • Canadian residents reported making 216.9 million person-trips annually (as defined above), of which 175.1 million were to destinations within Canada, 36 million were to the United States, and the remaining person-trips were to locations outside of North America. • Of the 175.1 million trips to destinations within Canada, 86.4 million were same-day trips and 88.7 million were overnight trips. • Of the 175.1 million trips to destinations within Canada, 153.5 million were made to a destina- tion within the same province. Fifty-three percent of these intraprovincial trips were same-day trips, while 47 percent were overnight trips. • Of the 175.1 million trips to destinations within Canada, 21.6 million were made to a destina- tion outside the home province. As to be expected, the majority of these interprovincial trips were overnight trips (80 percent). • In terms of trip purpose, of the 175.1 million person-trips to destinations within Canada, 39 percent were for pleasure, 36 percent were to visit friends or relatives, 14 percent were for personal or nondisclosed reasons, and the remaining 11 percent of trips were for business or to attend conventions. • In terms of travel mode, 160.8 million of the 175.1 million person-trips were made by car. Seven million person-trips were made by plane, 4.6 million by bus, 1.3 million by train, and 0.5 mil- lion by boat. The remaining 0.9 million person-trips were either an “other” mode or not stated. • Note: although trip distance is not recorded, the CTS documents the reported origins and destinations of travel at the city level. Thus, center city to center city distances could be esti- mated from the data. In addition, detailed tables provide person-trip numbers by various demographic and socioeconomic groupings. Criteria CTS TSRC Target Population Age 15+ Age 18+ Overnight Trips Captured (Regardless of Distance) All Only those “out of town” Same-Day Trips Captured – Distance Ontario residents: 40 kilometers or more All others: 80 kilometers or more All: Only those 40 kilometers or more from home to a destination “out of town” Table B.1. Differences between CTS and TSRC.

B-8 Long-distance and Rural Travel Transferable parameters for Statewide Travel Forecasting Models B.5 Rural and Small Town Canada Analysis Bulletin In 2002, Statistics Canada issued a Rural and Small Town Canada Analysis Bulletin. (Unless otherwise noted, the facts for this section come from http://www.statcan.gc.ca/pub/21-006-x/ 21-006-x2005005-eng.pdf). This document summarized details from the 2002 Canadian Travel Survey and its companion, the 2002 International Travel Survey (the CTS as described above focused on domestic travel, while the ITS documented statistics on travel to and from Canada). As with the other Canadian data sources, this bulletin was produced for the purpose of evaluating the tourism sector. The research was sponsored by the Agricultural Division of Statistics Canada, and “rural” regions were those with less than 150 persons per square kilometer. The purpose of the bulletin was to examine the number and characteristics of travelers to rural Canada in 2002. Although the analysis is strictly focused on leisure travel, the relevant statistics can be useful to benchmark against parameters developed from other data sources. Relevant statistics include • In 2002, there were 21 million leisure tourist visits to Canadian destinations. Of these, 83 per- cent were by Canadian travelers, 12 percent by U.S. residents, and the remainder from elsewhere in the world. • About half of all Canadian tourist visits were to predominantly rural regions. • About 39 percent of the 25.6 million visits by U.S. residents were to predominantly rural regions. • Almost half (41 percent) of the U.S. visitors were age 55 or older. Although not sufficient to develop parameters for this effort, the available statistics from the ITS can help to benchmark the parameters that are developed using other sources. Pertinent variables from the ITS that can help to inform both rural and long-distance analyses include (http://www.statcan.gc.ca/cgi-biN/Af-fdr.cgi?l=eng&loc=http://www.statcan.gc.ca/dli-ild/meta/ its-evi/2002/its2002cdn-gid.doc): • Household—Area of residence (province or tourism region); • Traveler—Age and sex (a combined variable); and • Travel Details—Canadian custom port of exit, date of exit, travel party size, main reason for trip, activities of travel party, mode of transportation for exit and entry, mode of transporta- tion while used, total nights in place visited. B.6 Canadian Tourism Surveys As noted above with the TSRC and the CTS, the Canadian government invests significant resources into understanding the tourist market in Canada. This includes the 2006 Travel Activi- ties and Motivations Survey, which documents all places traveled to in the past 2 years, activities undertaken, and general tourist details, as well as regional tourism surveys such as the 2006 Visitor Exit Survey of the Northwest Territories. The surveys are informative in terms of the residential location of visitors, which can be used to inform the development of attraction rates for specific tourist locations. The data from these surveys cannot, however, be used to estimate long-distance trip rates or other statistics of interest to this research. Northwest Territories Visitor Surveys The government of the Northwest Territories conducts visitor exit surveys on a regular basis to document baseline information on visitor numbers, spending patterns, and visitation char-

Travel Behavior data from Other Countries B-9 acteristics (http://www.iti.gov.nt.ca/publications/2007/tourismparks/2006%20Exit%20Survey %20-%20Outdoor%20Adventure.pdf). The objectives of these surveys include the following: • Obtain sufficient data to develop an estimate of total visitation to the Northwest Territories; • Obtain sufficient data to develop estimates of total visitation and spending for six visitor segments; • Obtain sufficient data to produce profiles for each visitor segment; and • Obtain data on interest and participation in Aboriginal Tourism products. A variety of survey instruments and methods are used to gather the data, including self- administered surveys, trip diaries, and tally count sheets. In the 2006 Visitor’s Exit Survey, 10,674 surveys were completed. Data elements obtained in this survey that might be relevant to this study included the following: • Primary destination; • Mode of travel to the Northwest Territories and while in the Northwest Territories; • Travel party size and composition (family, friends, etc.); • Age and gender of those in the immediate travel party; • Number of nights spent in the Northwest Territories, total, and by accommodation type; • Home location; • Trip purpose and activities undertaken while in the region; and • Demographics (educational attainment, occupation, household income). Through these surveys, Northwest Territories visitation for 2006/2007 was estimated at 63,461 total visitors, of which 38,819 were for leisure and 24,642 traveled for business. Details regarding home origin and distances traveled for all visitors were not published online. B.7 International Studies on Long-Distance Travel During the course of conducting literature scans for sources of long-distance travel data, several related studies conducted outside of North America were identified. Available studies are sum- marized in this section. The research team did not use data from outside North America, given differences in transportation options, development patterns, and other factors. However, it is important to note availability for future reference, as needed. European Studies Transportation researchers in Germany and France undertook a study to evaluate long- distance travel data from several European sources in order to develop harmonized results (Kuhnimhof et al. 2009). Their study considered data from the following eight different surveys: 1. Denmark National Travel Survey (2006)—Focus of this survey was on daily travel. There was not an explicit long-distance survey component. 2. French National Travel Survey (1993/1994)—The main survey was focused on daily travel. In addition, respondents were asked about long-distance travel (i.e., trips greater than 80 kilome- ters in “crow-fly” distance) during a 12-week period. 3. German National Travel Survey INVERMO (1999–2002)—Long-distance survey capturing trips more than 100 kilometers in length (network distance) over an 8-week period. 4. German National Mobility Panel (2006)—Panel survey of daily travel (no explicit long- distance survey component). 5. Swiss Microcensus (2005)—Daily travel survey that also obtained data on excursions more than 3 hours in length and/or overnight excursions over a 2- to 8-week period.

B-10 Long-distance and Rural Travel Transferable parameters for Statewide Travel Forecasting Models 6. Swedish National Travel Survey RES (2006)—This was also a daily travel survey that asked about trips more than 100 kilometers (network distance) within a 4-week period and those over 300 kilometers in network distance within an 8-week period. 7. United Kingdom National Travel Survey (2005)—Daily travel survey that obtained details about long-distance trips of 50 miles or more over a 4-week period. 8. Design and Application of a Travel Survey for European Long-Distance Trips Based on an International Network of Experience (DATELINE)—A 2001/2002 survey of long-distance travel across 16 European counties. Long-distance trips were defined as those of more than 100 kilometers in length (crow-fly distance) for a 3-month period, with extended questions about holiday travel over a 12-month period. Other Studies The following articles investigated some aspect of long-distance travel that may inform the interpretation of other data sources, using data from outside North America: • International Mobility Biographies—A Swedish (Frandberg, 2006) researcher surveyed 162 students about their international travel (i.e., all international trips made during their childhood and adolescence). Her theory was that the level of international travel was influ- enced by the educational level of the parents, their economic resources, and whether they lived in an urban or rural area. The 162 students reported almost 3,300 international trips, averaging 21 international trips per student. Almost half (44 percent) of the students reported regular travel abroad. The paper did not report on factors that influenced travel (i.e., did not directly address the research question). Instead, this paper offers interesting insights into retrospectively collecting long-distance travel, what results are more reliable, and why. • Destination Loyalty—This study (McKercher and Guillet, 2011) investigated the research question of whether destination loyalty exists either at the tourist or market level. The research evaluated year-by-year and repeat-visitation intentions of Hong Kong residents to 11 popu- lar destinations. The findings suggest, “low individual repeat visitation intention, but overall market stability” in terms of travel to specific destinations, with similar profiles identified for those who had traveled to a specific destination and those who intended to travel to that same destination. This information may help to inform distributions of leisure travel patterns in the United States.