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.

47 This, and subsequent sections on transferable parameters and benchmark statistics, will fol- low a similar format. The first subsection will provide benchmark statistics and parameters from existing statewide models. This will be followed by a discussion of analytical approaches used to estimate transferable parameters. Next will be a presentation of long-distance transferable parameters and benchmarks. Each section will then conclude with rural travel parameters and benchmarks. The section on trip generation specifically touches on alternate trip generation approaches to statewide models, and statewide model trip purposes, as well as differences between urban, rural, and long-distance trip-making. This will be followed by a presentation of transferable trip production rates and guidance on making adjustments to these parameters. This section will also provide benchmark statistics on aggregate trip rates and percent trips by purpose. 4.1 Long-Distance and Rural Trip Generation Benchmark Statistics from Statewide Models and Other Sources This section of the trip generation chapter explores the characteristics of statewide models further to identify sources that could be used in comparing, developing, and recommending trip production rates for estimating rural and long-distance travel. Other statewide model statistics such as friction factors, mode choice coefficients, and peak-to-daily/time-of-day factors, and other model parameters, are summarized later in Sections 5.1, 6.1, and 7.1, which are devoted to other steps in the four-step modeling process. Other secondary sources of model parameters and benchmarks are provided for comparative purposes. Statewide Model Parameters and Benchmarks The final report for the NCHRP Statewide Model Validation Study (Cambridge Systematics, Inc., 2010d) included a series of tables describing model parameters and benchmark statistics from statewide models, including information on long-distance and rural trip purposes, where these were separated from typical urban model purposes. Some of this information was derived either from recent work on the NCHRP model validation report or prior work on national model research for FHWA. Establishment of trip purposes used in statewide models is important because this will largely determine the stratifications used in subsequent model statistics (i.e., these are reported by trip purpose). Some trip purposes in statewide models are duplicative, using different names but meaning the same thing. This has been fleshed out through discussions with state DOT contacts C h a p t e r 4 Trip Generation Parameters and Benchmark Statistics

48 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models and their consultants. Some models differentiate short-distance from long-distance trip purposes while others do not. Where long-distance trips are separated from routine travel, the percent of long-distance trips varies widely in statewide models from less than 1 percent (Florida, Louisiana) to greater than 4 percent (Massachusetts); this may reflect, to some extent, the close proximity of densely developed urbanized areas, resulting high levels of through-trip activity, long-distance commuting, and other unique factors that make transferability of this statistic difficult. Also, reported statistics make use of different thresholds for long-distance travel. Trip generation model statistics compiled by trip purpose include aggregate trip rates and percent trips by purpose. In many cases, states have incorporated methods for forecasting long- distance trips along with shorter regional trips in their statewide models. In most cases, statewide models incorporate truck and auto long-distance trips; however, in some cases, additional modes are incorporated such as air and intercity transit. The threshold for defining long-distance trips also varies among statewide models, with some states considering trips over 100 miles to be long distance, and others considering 50 miles or 75 minutes as long distance. Table 4.1 is a summary of the percentage of trips that are long-distance for each statewide model, along with a breakdown of each state’s definition for long-distance trips, as reported in available technical reports. Long-distance trip production rates were documented for Georgia and Wisconsin statewide models only, as depicted in Tables 4.2 and 4.3. No trip attraction rates were found for rural or long-distance travel in any of statewide model documents reviewed. These tables, as well as other statewide model statistics found in subsequent sections, depict passenger trips except where noted otherwise. The numbers found in these tables, in all cases, came directly from statewide model technical reports because the study team was not tasked with obtaining and executing these models. Bureau of Transportation Statistics In May 2006, the Bureau of Transportation Statistics (BTS) published findings from the 2001 NHTS on long-distance trip-making. A number of these statistics could be useful as transferable parameters or benchmark statistics against which to compare statewide model results. Although later tasks in this study will include data analysis of 2009 NHTS, 2001 NHTS, 1995 ATS, and other relevant state survey datasets, it was thought that information from the BTS report, America on the Go, Findings from the National Household Travel Survey (U.S. Department of Transportation, Long-Distance Threshold Long-Distance Total Percentage in Miles in Minutes Trips Trips Long Distance Arizona 50 – – – – California 100 – – – – Florida 50 – 176,587 52,281,363 0.34% Georgia – 75 418,000 31,223,000 1.34% Indiana – – 280,395 25,158,208 1.11% Louisiana 100 – 75,087 11,717,965 0.64% Massachusetts – – 957,046 22,951,483 4.17% Mississippi 100 – 212,862 7,095,161 3.00% Ohio 50 – 248,628 36,702,991 0.60% Utah – – 68,866 7,313,412 0.94% Virginia – 100 1,071,566 37,868,443 2.83% Wisconsin 50 – 42,966 71,313,993 0.06% Table 4.1. Percentages of long-distance trips in statewide models.

trip Generation parameters and Benchmark Statistics 49 Table 4.2. Georgia long-distance internal and external trip rates by purpose, income, area, and persons per household. Income Area Persons per Household HBW-IE (GA Int-Ext) HBW-II (GA Internal) HBO-II (GA Internal) NHB-II (GA Internal) Low Urban 1 0.008 0.001 0.036 0.005 2 0.045 0.002 0.063 0.009 3 0.025 0.003 0.083 0.020 4 0.077 0.005 0.060 0.154 Rural 1 0.045 0.045 0.016 0.010 2 0.020 0.043 0.087 0.130 3 0.091 0.003 0.045 0.040 4 0.056 0.167 0.667 0.056 Non-Low Urban 1 0.016 0.003 0.013 0.010 2 0.046 0.005 0.041 0.017 3 0.051 0.009 0.041 0.054 4 0.051 0.015 0.127 0.036 Rural 1 0.015 0.002 0.032 0.021 2 0.035 0.022 0.104 0.042 3 0.052 0.007 0.095 0.087 4 0.070 0.022 0.081 0.059 Source: Atkins, Development of Statewide Model Draft Report, prepared for Georgia Department of Transportation, April 15, 2011. 1 Household Member 2 Household Members 3 Household Members 4 Household Members 0 Autos 1 Auto 2 Autos 0 Autos 1 Auto 2 Autos 0 Autos 1 Auto 2 Autos 0 Autos 1 Auto 2 Autos Business Appleton/Oshkosh/Green Bay 0.00000 0.00367 0.01316 0.00000 0.00479 0.01997 0.00000 0.00057 0.02917 0.00000 0.00792 0.03429 Madison 0.00046 0.00545 0.02347 0.00000 0.00624 0.02737 0.00000 0.01451 0.02480 0.00000 0.06359 0.03617 All other MPOs 0.00145 0.00633 0.03949 0.00145 0.01656 0.02953 0.00145 0.00829 0.04266 0.00145 0.02101 0.04851 SEWRPC Region 0.00148 0.00405 0.00851 0.00148 0.00370 0.02202 0.00148 0.00251 0.01687 0.00148 0.00909 0.03399 Rest of Wisconsin 0.00060 0.00647 0.04464 0.00060 0.01875 0.02487 0.00060 0.00590 0.04473 0.00060 0.02524 0.05001 Personal Business Appleton/Oshkosh/Green Bay 0.00133 0.00293 0.00576 0.00133 0.01358 0.01133 0.00133 0.01370 0.02396 0.00133 0.01475 0.02079 Madison 0.00352 0.00427 0.00583 0.00352 0.01175 0.01246 0.00352 0.01092 0.01461 0.00352 0.00836 0.02078 All other MPOs 0.00107 0.00436 0.01110 0.00107 0.01415 0.02099 0.00107 0.01373 0.03127 0.00107 0.03690 0.02986 SEWRPC Region 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 0.00895 Rest of Wisconsin 0.00077 0.00479 0.01266 0.00077 0.01523 0.02234 0.00077 0.01025 0.03376 0.00077 0.04535 0.03144 Pleasure Appleton/Oshkosh/Green Bay 0.00752 0.01399 0.02085 0.00752 0.05138 0.06096 0.00752 0.05508 0.07750 0.00752 0.05330 0.12458 Madison 0.01538 0.01773 0.01712 0.01538 0.04875 0.06335 0.01538 0.04614 0.08682 0.01538 0.07146 0.12193 All other MPOs 0.00684 0.01829 0.02851 0.00684 0.04550 0.07099 0.00684 0.04713 0.08072 0.00684 0.06755 0.09718 SEWRPC Region 0.00717 0.01443 0.01740 0.00717 0.03076 0.05512 0.00717 0.03350 0.06772 0.00717 0.02557 0.07727 Rest of Wisconsin 0.00440 0.01575 0.26130 0.00440 0.03908 0.05959 0.00440 0.05192 0.08034 0.00440 0.08730 0.10344 Source: Cambridge Systematics, Inc. and HNTB, Wisconsin Statewide Model –Passenger and Freight Models, prepared for Wisconsin Department of Transportation, September 2006. Table 4.3. Wisconsin daily long-distance trip rates by purpose, household size, and number of autos.

50 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models Research and Innovative Technology Administration, Bureau of Transportation Statistics, 2006) fit into the context of this discussion. This 2006 BTS analysis of long-distance trips identified characteristics such as percent of trips by mode and purpose, as depicted in Table 4.4. According to BTS’ analysis, over 50 per- cent of long-distance trips would be considered for the purposes of pleasure, with another 16 percent of trips occurring for business purposes. Travel modes are fairly consistent for most long-distance purposes with the exception of business trips, which are far more likely to use air travel than other long-distance trip purposes. This shows the importance of modeling long- distance business trips separately from other LD trip types when modeling multiple transpor- tation modes. Oak Ridge National Laboratories A 2006 report, Trends in New York State Long-Distance Travel (Oak Ridge National Labora- tory, 2006), produced by staff from the Oak Ridge National Laboratory, provides a number of statistics on long-distance travel patterns based on analyses of 1995 ATS and 2001 NHTS data for residents of, and visitors to, the State of New York. Statistics provided in this report include growth in long-distance trips, the number of person trips, trips per person, miles per person, and miles per trip tabulated by means of transportation, trip purpose, income, age, and gen- der. These statistics are not necessarily transferable to other states but could be useful in benchmark checking against comparable statistics calculated from other surveys or for other states. A 2009 NHTS Update to this report is under way and select chapters are available for downloading (https://www.dot.ny.gov/divisions/policy-and-strategy/darb/dai-unit/ttss/nhts/ 2009-comparision-report). Statewide Travel Surveys by State DOTs The Ohio statewide model considers long-distance trips to be 50 miles or greater, excluding work tours (Ohio Department of Transportation, Report-Ohio-LongDistanceTravelModule- Extracted.pdf Section 4.7, LDT). Background information on the Ohio Long-Distance Travel Survey was provided in Section 2.5 of this report. Attention was focused on summary statis- tics that already were reported in available survey documentation. Section 2.5 provides graphs depicting the frequency of, and travel modes used in, long-distance trip-making based on the Ohio surveys. In addition to concerns over resource sufficiency to analyze additional state data- sets using SAS as part of this research effort, there is the issue of whether or not a supplemental Percent Trips by Mode LD Purpose Percent by Purpose Personal Vehicle Air Bus Train Other Pleasure 55.5% 90.4% 6.7% 2.2% 0.5% 0.2% Business 15.9% 79.3% 17.8% 0.8% 1.6% 0.5% Commuting 12.6% 96.4% 1.5% 0.5% 1.7% 0.0% Personal Business 12.6% 89.3% 4.7% 5.6% 0.3% 0.1% Other 3.4% 96.6% 1.9% 0.5% 0.0% 1.0% Total 100.0% 89.5% 7.4% 2.1% 0.8% 0.2% Source: BTS. Table 4.4. 2001 long-distance trips by purpose and mode.

trip Generation parameters and Benchmark Statistics 51 analysis of state DOT datasets would provide the same results as previously reported, due to weighting/expansion and tools used in the analysis. The Ohio statewide travel survey documentation provided to the study team did include person trip rates per household (7.78) and person (4.94) for rural versus urban settings (7.56–8.76 per HH and 4.83–5.49 per person, depending on the specific urban area). The survey documentation, however, does not include long-distance trip rates, trips by pur- pose, trip distribution factors, average trip lengths, mode splits, or auto occupancy rates. Beyond sociodemographic characteristics of survey respondents and the graphs and charts previously depicted in earlier chapters of this Guidebook, information was also provided on the number of stops (60.9 percent made stops) for long-distance trips and the percent of nonhome-based long-distance trips (53 percent). Section 2.6 of this report provides background information on the Michigan surveys, includ- ing annual long-distance household trip rates (7.34 in 2004 and 6.25 in 2009), trip purposes, travel modes, and long-distance trip distribution by state. The difference between Michigan long-distance trip rates and those based on ATS (10.15) and 2001 NHTS (12.32) excluding 50–100 mile trips, indicate potential issues of transferability as long-distance trips were defined as 100 miles or greater in the Michigan surveys. Considerable information was provided in the Michigan documentation about trip characteristics that are more relevant to the discussion of rural trip rates in Chapter 3 of this report. The Michigan statewide household survey documentation provided household person trip rates for different urban and rural stratifications. Table 4.5 depicts person trip rates per house- hold for the first and second Michigan Travel Counts Surveys (2004 and 2009, respectively). Nonurbanized and rural household and person trip rates are depicted in bold underlined text, Sample Areas Households Weighted Persons Weighted HH Trip Rates Person Trip Rates Unweighted Weighted Unweighted Weighted Estimated Number of HHs, Persons, and Trips by MTC I (2004) SEMCOG 1,846,277 4,638,216 9.09 9.14 3.62 3.64 Small Cities (0-50k pop) 129,369 296,162 9.74 8.82 3.89 3.85 Upper Peninsula Rural 87,115 209,919 8.35 8.40 3.49 3.49 Northern Lower Peninsula Rural 206,210 501,075 8.08 7.96 3.30 3.27 Southern Lower Peninsula Rural 394,588 1,044,969 8.98 9.41 3.53 3.55 TMAs 579,415 1,465,017 9.68 9.53 3.78 3.77 Small Urban Modeled Areas 545,557 1,360,511 9.29 9.39 3.75 3.77 State Total 3,788,531 9,515,870 9.05 9.17 3.63 3.65 Estimated Number of HHs, Persons, and Trips by MTC II (2009) SEMCOG 2,071,786 4,820,277 7.73 8.60 3.66 3.70 Small Cities (0-50k pop) 147,121 315,640 9.10 8.14 3.96 3.80 Upper Peninsula Rural 90,553 212,970 7.29 7.64 3.28 3.25 Northern Lower Peninsula Rural 218,238 520,125 7.32 7.93 3.30 3.33 Southern Lower Peninsula Rural 412,944 1,078,905 8.07 9.15 3.43 3.50 TMAs 622,928 1,519,419 8.34 8.83 3.65 3.62 Small Urban Modeled Areas 593,556 1,399,086 7.87 8.72 3.68 3.70 State Total 4,157,125 9,866,421 7.97 8.63 3.57 3.64 Source: Michigan Travel Counts Surveys. Table 4.5. Michigan TCS rural versus urban household trip rates.

52 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models separately for small cities, upper peninsula, northern lower peninsula, and southern lower peninsula. With the exception of the more heavily populated southern lower peninsula, non- urbanized households exhibit lower weighted trip rates than urban households. This finding is somewhat contrary to the 2009 NHTS analysis in Appendix G, unless most of the Michigan urbanized household surveys were conducted in suburban settings, since suburban house- holds showed higher trip rates than rural households in the 2009 NHTS. It is also worth noting that the differences found among nonurbanized areas of Michigan might indicate limitations to transferability. Recent and Ongoing GPS Surveys in the United States Table 4.6 depicts the split between weekend and weekday travel from recent GPS surveys described earlier in Section 2.3 of this Guidebook. These splits, after survey expansion, are simi- lar in each survey. In all cases, as expected, the percent of weekend trips is highest for the long- distance trips (most 30 percent or higher) when compared against urban and rural travel. Table 4.7 depicts four different trip/demographic measurements for each of the four surveys and overall, for the same three different geographic definitions found in the prior two tables. Daily trip production rates for long-distance trips are low, as expected. Rural trip production rates are substantially higher than urban rates for these four surveys, although the rates are comparable to analysis of 2009 NHTS, as described later in this chapter. Long-Distance Rural Urban Weekday Weekend Weekday Weekend Weekday Weekend Overall 68% 32% 77% 23% 79% 21% Atlanta 66% 34% 75% 25% 79% 21% Denver 72% 28% 78% 22% 80% 20% Massachusetts 68% 32% 86% 14% 88% 12% Chicago 64% 36% 73% 27% 75% 25% Source: Geostats based on recent GPS-based travel surveys. Table 4.6. Travel day statistics from recent GPS-based surveys. Number of Trips Trip Production Rate Average Household Size Average Number of Vehicles/HH L on g D is ta n ce R u ra l U rb an A ll T ri p s L on g D is ta n ce R u ra l U rb an L on g D is ta n ce R u ra l U rb an L on g D is ta n ce R u ra l U rb an Overall 1,253 31,689 103,832 6.20 0.04 9.39 5.23 2.85 2.86 2.75 2.23 2.18 2.05 Atlanta 580 16,932 48,098 5.92 0.03 8.24 5.03 2.89 2.85 2.77 Denver 395 9,836 31,377 6.11 0.04 9.42 4.85 2.77 2.81 2.68 2.22 2.16 2.07 Massachusetts 176 3,349 10,325 5.90 0.04 13.56 5.04 2.92 2.96 2.82 2.27 2.21 2.02 Chicago 102 1,572 14,032 7.89 0.05 8.86 7.05 Source: Geostats based on recent GPS-based travel surveys. Table 4.7. Trips and households from recent GPS-based surveys.

trip Generation parameters and Benchmark Statistics 53 2010 Travel Survey of Residents of Canada As described in Appendix B, the Travel Survey of Residents of Canada (TSRC) is designed to measure the size and status of Canada’s tourism industry at the national level. Through direct contact with Canadian officials, the research team was able to obtain a spreadsheet data analysis of the 2010 TSRC. Without direct access to the data, which would have required additional budget for purchasing data, this study was limited to information provided in this spreadsheet. Table 4.8 depicts the percent of long-distance trips by purpose, with single-day travel sepa- rated from overnight travel. Although the trip purposes used in the TSRC are different from those found in the ATS, business-related trips are considerably less in the TSRC, at slightly more than 5 percent, versus the ATS at 22 percent. These statistics are from fully weighted survey data; however, without additional analysis, it is unclear whether the lower percent is a function of sampling or that long-distance business travel is considerably less common than in the United States. 4.2 Analytical Approach to Estimating Long-Distance and Rural Trip Generation Parameters and Benchmarks One key to implementing the analytical plan and developing transferable parameters was to obtain access to all datasets from the American Travel Survey (ATS) and identify trip purposes, average trip lengths, vehicle occupancies, and other statistics typified by long-distance travelers. The 1995 ATS datasets are dated; however, these data are the only long-distance data that provide statistically sound estimates of long-distance travel in and between the states. Although the 2001 National Household Travel Survey (NHTS) had a long-distance compo- nent, this survey did not have sufficient samples to calculate estimates of long-distance travel for most states (New York and Wisconsin were exceptions to this, because of the large Add-On in the former and stratified sampling of the latter, although neither Add-On was included in the official 2001 NHTS long-distance file). The approach to using NHTS 2001 data was based on discussions with FHWA NHTS support staff, both past and present, as well as members of the research team with extensive experience using different versions of the NHTS. All of these dis- cussions pointed to concerns over the use of NHTS 2001 for long-distance trips and at least some Trip Duration: Total – Domestic Travel (Age 18+): Person Trips with the Destination in Canada *** Row Percents *** Main Trip Purpose: Standard person trip stub variables Total Pleasure, Vacation, Holiday Visiting Friends or Relatives Business and All Conferences or Conventions Shopping and Other >>> Final Data <<< Total Long-Distance Trips: 100.00% 36.99% 46.72% 5.40% 10.90% Single-Day Long-Distance Trips 100.00% 34.43% 45.59% 5.14% 14.84% Overnight Long-Distance Trips 100.00% 40.79% 48.41% 5.77% 5.03% Source: Travel Survey of Residents of Canada. Table 4.8. Canadian residents’ long-distance trips by percent purpose.

54 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models of these concerns are documented elsewhere in this report. All of the NHTS 2001 long-distance data, including state Add-On samples, were made available for use by the research team as well. These two long-distance datasets can be used together, yet separately, since the 2001 ques- tionnaire relied heavily on the 1995 ATS as a template. Definitional categories for mode and purpose are comparable. The study team also obtained readily available state DOT survey data and documentation from statewide household travel surveys for Michigan and Ohio. The research team also coordinated with Canadian officials to identify available long-distance travel parameters readily available from their recent household travel surveys. Finally, recent travel surveys using Global Positioning Systems (GPS) were mined for parameters on long-distance travel as well as rural parameters. Transferable rural travel parameters largely focused on the 2009 NHTS and its state Add-On surveys. Analysis of variance (ANOVA) and other statistical tests were run on 2009 NHTS data in an attempt to identify which available attributes best explain differences in rural trip-making and whether certain parameters should be stratified for different conditions such as urban clusters and proximity to urbanized area boundaries. Existing statewide models also played a significant role in this analytical plan, in terms of quantifying reasonableness ranges against which to compare resulting ATS/NHTS survey-based model parameters. Also, documented model parameters were identified for potential transfer- ability to other statewide models, based on the characteristics of the state where the data were collected versus the state to which a parameter might be proposed for transferability. Inter- regional or intercity travel components are included in some statewide models to capture both intrastate and interstate trips. The core model design feature is the recognition that interregional travel is very different from urban area travel, where different sets of explanatory variables or different sensitivities to levels of service are involved. A set of typical long-distance and rural trip purposes was established from this analysis so that model parameters could be stratified by such categories and reasonableness benchmarks could be established for percent trips by purpose. Mean trip length statistics, both in miles and min- utes, also were estimated from the survey databases for use as benchmarks in future statewide model validation efforts; however, the survey analysis for this study did not include the calculation of state-by-state trip lengths. As discussed previously, statewide models and travel surveys have used a range of thresholds to define long-distance trip-making. Most sources cited in this study used either 50, 75, or 100 miles as the minimum threshold for trips to be considered “long-distance.” In an effort to maximize the number of long-distance trip samples, this report looks at model parameters at three different long-distance trip thresholds: 50–100 miles, 100–300 miles, and more than 300 miles. By separat- ing out 50–100 mile trips from 100–300 miles, this allows for differentiation of long-distance trips by the two most common thresholds, beginning and ending at 100 miles. The rationale for using 300 miles as another cutoff point is that preliminary data analysis indicated a mode shift from personal auto to air travel at this distance. The remainder of this section of the chapter on trip generation focuses on the data sources used for parameter estimation, along with some general comparisons among sources. American Travel Survey (ATS) As stated elsewhere, the 1995 ATS is still seen as the most robust sample of long-distance travel behavior, in spite of its age. The ATS was entirely focused on long-distance travel, unlike the 2001 NHTS, which also surveyed typical daily urban and rural travel patterns. There are numerous ways to analyze the data. For this study, household frequencies and statistical means were cal- culated separately for all households and per capita as well. Trip rates and frequencies also were

trip Generation parameters and Benchmark Statistics 55 calculated separately for annual and daily conditions. The reason for this is that long-distance trips are not an “every day” occurrence for most households. Trip rates were initially calculated on an annual basis using the ATS and then divided by 365 to provide daily trip rates as an option for users of this report. Since the ATS reported annual trips, long-distance trip characteristics in this Guidebook are likewise summarized as annual trips. Another consideration was whether to include weekdays and/or weekends, of which both were calculated. According to available documentation on the ATS “each trip was classified as a weekend trip or as not a weekend trip. A weekend trip is a trip of one to five nights, includ- ing a Friday and/or Saturday night stay. Travelers who stay one or two nights away, including a Friday or Saturday night are defined as regular weekend travelers. Those who stay three to five nights away, including a Friday and/or Saturday night stay are defined as long weekend travelers” (http://www.bts.gov/publications/1995_american_travel_survey/an_overview_ of_the_survey_design_and_methodology/index.html). Based on this description it was not practical to summarize only weekday trips because some of the “weekend” trips were partially “weekday” trips. Furthermore, analysis of weekday-only trips resulted in a dramatic drop in long-distance trip rates that would be inconsistent with results from other long-distance surveys. Therefore, long-distance trip statistics found in this report include both weekday and weekend trips. Cross-classification was used to evaluate different attributes against one another, as well as to calculate trip rates based on socioeconomic characteristics. The latter used household size by income, which is consistent with one cross-classification scheme used and documented in the previously referenced NCHRP Report 716 on urban transferable parameters. It is important to note that the ATS did not include trips of 50–100 miles in length, and so there are no 50- to 100-mile trips in the ATS statistical tables in this chapter. Statistics reported in the next section on parameters from the 2001 NHTS do include 50- to 100-mile trips, consistent with the lower long-distance trip threshold used in that survey. Statistics for 50- to 100-mile trips are presented only for analysis of the 2001 NHTS survey. Table 4.9 is an assessment of household trip rates by trip purpose and the relevant trip dis- tance categories noted earlier. As shown in this table, a typical household generates 10.15 trips of over 100 miles, or 0.0278 daily long-distance trips (annual trip rate divided by 365 days). Trips were grouped into three purposes: business, pleasure, and personal business. Pleasure trips had the highest average trip rate for all three distance categories while the majority of trips were 100–300 miles for all trip purposes. Table 4.10 compares trip rates by household income level and mileage range. A review of this table shows that the highest-income group has the greatest long-distance trip rate for both mile- age categories. Trip rates show that the propensity of making long-distance trips, and the length of those trips, has a lot to do with household income. Purpose -300 miles > 300 miles Total 01 Business 1.37 0.91 2.28 02 Pleasure 4.08 2.13 6.21 03 Personal Business 1.19 0.47 1.66 Total 6.64 3.51 10.15 Source: 1995 ATS. Table 4.9. ATS annual trip rates by distance/purpose, round-trip.

56 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models Table 4.11 provides information from the 1995 ATS on weekday versus weekend long-distance travel based on three trip purposes. Not surprisingly, business and personal business trips are less likely to occur on weekends when compared to pleasure trips, of which over 50 percent involve weekend travel. As noted earlier, some portion of the “weekend” trips in fact take place on weekdays, and so the term “not weekend” was chosen instead of “weekday.” Simply put, the “not weekend” trips are those that take place entirely on weekdays, with no portion of the long- distance trip taking place on a weekend. Table 4.12 presents the number of long-distance intermediate stops by trip/tour purpose. More than 90 percent of long-distance trips did not include any stops. The percent of intermedi- ate stops is highest for business trips and lowest for personal business trips. The trip, including all intermediate stops, could be analogous to the concept of a trip tour. 2001 National Household Travel Survey The 2001 NHTS database also was used to develop long-distance model parameters and benchmarks similar to those produced using the 1995 ATS. A primary reason for developing these statistics using the 2001 NHTS was to overcome concerns about the age of the 1995 ATS data. Use of the 2001 NHTS was considered acceptable for long-distance analysis because this survey included a targeted sample of long-distance trips, unlike the more recent 2009 NHTS database. Unfortunately, there are several shortcomings with the 2001 long-distance survey com- ponent, including the following: • Much lower response rate using a telephone survey approach for the 2001 NHTS versus the panel survey approach used in the 1995 ATS; • Shorter recall period of the 2001 NHTS also resulted in a much smaller sample size of long- distance trip-makers (45,000 in 2001 NHTS versus 550,000 in 1995 ATS); Income -300 miles > 300 mi Total 01 $0-$24,999 2.97 1.30 4.27 02 $25,000-$99,999 8.48 4.34 12.82 03 $ 100,000+ 13.78 12.49 26.26 Total 6.64 3.51 10.15 Source: 1995 ATS. Table 4.10. ATS annual trip rates by distance/income, round-trip. Purpose Not Weekend Weekend Total Percent Weekend 01 Business 85,261 33,910 119,171 28.45% 02 Pleasure 156,188 159,595 315,783 50.54% 03 Personal Business/Other 67,097 33,469 100,566 33.28% Total 308,546 226,974 535,520 Trip % 58% 42% 100% Source: 1995 ATS. Table 4.11. ATS annual frequency by purpose/weekend trip, round-trip.

trip Generation parameters and Benchmark Statistics 57 • The impacts of 9/11 on travel resulted in a much lower share of air travel in the 2001 NHTS when compared against the 1995 ATS; and • Thresholds used to define long-distance trips differ between the two surveys with 1995 ATS defined as 100 miles or greater and 2001 NHTS as more than 50 miles. Chapter 2 of this Guidebook described the latter difference as a potential strong point of NHTS 2001 as a way of obtaining information on these mid-range 50–100 mile trips. Analysis of 2001 NHTS long-distance trip data showed a sizeable sample of 50–100 mile trips (21,500 out of 45,000 trips, or nearly 48 percent of the 2001 NHTS long-distance sample), such that exclusion of these trips for NHTS 2001 statistics would be problematic. Hence, 50–100 mile trips were included for NHTS 2001 analyses. Table 4.13 depicts annual long-distance household trip rates from the 2001 NHTS, including 50–100 mile trips. The overall trip rate (excluding the 50–100 mile trips) is 12.32, about 21 percent higher than the 1995 ATS trip rate of 10.15. The patterns within the cross-classification table are relatively similar between the two surveys (i.e., which cells have higher or lower rates than others). It is worth noting that the percentages of trips by purpose are somewhat similar between the two long-distance surveys, as follows: • Business—28.38 percent for NHTS 2001 versus 22.25 percent for ATS; • Pleasure—54.84 percent for NHTS 2001 versus 58.97 percent for ATS; and • Personal Business—16.78 percent for NHTS 2001 versus 18.78 percent for ATS. Rural Typologies Identification of rural travel parameters took a different focus than long-distance travel param- eters. First, rural trip-making data are well represented in the recent 2009 NHTS. Therefore, the Purpose 2 3 4 Total Percent Stop/ Purpose 01 Business 106,212 2,059 7,515 3,385 119,171 10.87% 02 Pleasure 293,727 8,243 8,412 5,401 315,783 6.98% 03 Personal Business 94,888 2,470 1,998 1,210 100,566 5.65% Total 494,827 12,772 17,925 9,996 535,520 7.60% Percent Stops by Number 92.40% 2.38% 3.35% 1.87% 100.00% Source: 1995 ATS. Table 4.12. Annual frequency by stops from destination/purpose, round-trip. Purpose 50-100 Miles 100-300 Miles > 300 Miles Total 01 Business 4.04 1.85 0.97 6.85 02 Pleasure 5.71 5.08 2.39 13.17 03 Personal Business/Other 1.78 1.52 0.52 3.83 Total 11.53 8.45 3.87 23.85 100+ Mile Trip Rate 12.32 Source: 2001 NHTS. Table 4.13. 2001 NHTS annual trip rates by distance/purpose.

58 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models research team was able to focus primarily on this one survey database, unlike the multiple and considerably older survey databases used to identify long-distance travel parameters. Second, the points of reference are quite different for rural trips. Long-distance travel characteristics were generally summarized by different trip length categories, whereas rural travel parameters required establishing typologies for classification and comparison against comparable statistics on travel in urbanized areas. Finally, the temporal issues for rural travel are not as complex as long-distance trips. For example, the database does not deal with international travel or multiple stops, and the greater share of travel is on the weekdays, with a much smaller share of weekend travel than with long-distance trips. The first step in the assessment of rural travel parameters was the identification of rural typolo- gies and an exploration of how these different typologies can be used to describe the trip-making of rural households. This also includes the need to define what is and is not rural travel and how typical rural travel behavior differs from that in more urban settings. These efforts started with a focus on attributes contained within the NHTS 2009 “DOT version” of the database, including the Claritas attributes described earlier in this Guidebook. The following similar, yet not identical, attributes from the 2009 NHTS DOT version were used to identify potential rural typologies: • URBAN—Identifies whether or not the home address is located in an urban area, typically defined as a concentrated area with a population of 50,000 or greater; • URBRUR—Identifies whether or not the home address is located in a rural area; • URBANSIZE—Population size of the urban area in which the home address is located; • HBHUR—Urban/Rural Indicator, appended to the NHTS by Claritas (http://nhts.ornl. gov/2009/pub/UsersGuideClaritas.pdf)—this classification reflects the population density of a grid square into which the household’s block falls; • HBRESDN—The number of housing units per square mile by block group; and • HBPOPDN—The population per square mile by block group. Additionally, the rural typologies recommended as part of NCHRP Project 25-36, “Impacts of Land Use Strategies on Travel Behavior in Small Communities and Rural Areas” and described earlier also were considered in this effort. The four typologies recommended by NCHRP Proj- ect 25-36 were as follows, along with the study definitions of each, as quantified by “commuting zones” developed by the USDA’s Economic Research Service: • Population Density—Computed as number of people divided by unit area of developed or developable land; • Road Density—Calculated as road length in miles per square mile of developed or develop- able land; • Land-Use Mixture—A proxy of land-use mixture measuring how residents, jobs, and other activities are distributed in relation to each other; and • Variation in Population Density—Variation in population density distinguished where most residents are located in a relatively small set of concentrated areas at relatively high densities from locations where residents are spread more evenly. This project did not pursue full consideration of commuting zones, which are defined in NCHRP Project 25-36 as “multicounty regions that convey the typical pattern of commuting trips in a spatially defined labor market: a much higher proportion of commuting trips have origins and destinations that are both inside the zone than those trips for which one end is outside” (Department of City and Regional Planning Center for Urban and Regional Studies, University of North Carolina at Chapel Hill, 2011). In place of data on commuting zones, the analysis presented here uses readily available data to simulate some of these typologies. Population Density already was an attribute included in

trip Generation parameters and Benchmark Statistics 59 the 2009 NHTS dataset so it was easily addressed. Road Density was calculated using the 2005 National Highway Planning Network and geographic information systems (GIS) tools, based on a simple formula of Road Length/Census Tract Area. The resulting Road Density was a con- tinuous variable, so a regression analysis was conducted and then the variable was re-coded as a categorical variable. There was no practical way to simulate land-use mixture or the variation in population density using the data readily available for this project. One additional typology analyzed was “urban proximity” because the NCHRP Project 8-84 research team thought that the proximity to urban areas could impact the number and purpose of trips. Latitude/longitude address information was not stored for each household in the 2009 NHTS DOT database, which is necessary for accurate depiction in GIS. The database did have Census tract and block group information, and this information was appended to an NHTS Census tract/block group shapefile. Once the 2009 NHTS DOT database was joined to the NHTS CT/BG shapefile by a block group ID number, the households were spatially referenced to the block group. Figure 4.1 depicts a map of concentric rings formed during the proximity analysis, zoomed into north Florida/south Georgia, as an example. In cases where a block group was in proximity to multiple urban areas, distance to the closest urban area was applied. Unfortunately, Proximity to Urban Area did not show any clear trip rate trend, and so the analysis focused on the other measures. As noted previously, while it would have been ideal to use a national land coverage database to identify subcategories of rural areas such as exurban, agricultural, and recreational, the research Figure 4.1. Example map depicting proximity to urban area.

60 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models team had concerns over how to define and classify rural areas. It was thought that proximity to urbanized area, residential density, and roadway density allowed for a more objective classification of rural households. In order to narrow the number of rural typologies used in recommending transferable rural parameters, analysis of variance (ANOVA) and t-tests were conducted on each of the typolo- gies discussed below. The trending of trip rates up or down in relation to different settings for each attribute also was reviewed to further ascertain the explanatory power of each typology variable. In some cases, the number of categories was narrowed to assess the viability of each. Appendix G includes a separate page for each typology variable, along with t-test values for each category, analytical and trend observations, and revisions to the number of categories for each attribute. Four typologies were subsequently recommended for the purposes of calculating transferable rural trip production rates. These four typology variables were as follows: 1. HBHUR—Urban/Rural Indicator reflecting population density of a grid square; 2. URBAN—Whether or not the home address is located in an urban area; 3. URBRUR—Whether or not the home address is located in a rural area; and 4. HBRESDN—Number of housing units per square mile by block group. Table 4.14 depicts how these typology attributes could interact and be cross-classified into three dimensions for trip generation, focusing on the first three attributes to deal with different geog- raphies and the latter, housing units per square mile (HBRESDN), being applied against all rural and urban categories. The number of 2009 NHTS household samples by row and column also is provided in parentheses. Cells with “N/A” represent combinations of three attributes that should not exist. For example, rural areas should not also be classified into suburban, second city, or urban. This approach is further refined into a set of trip production rates cross-classified by socio- economic household characteristics, and specified by trip purpose, as described in a later section of this chapter. 2009 National Household Travel Survey With its large sample size of rural households, the 2009 NHTS was the principal source of rural travel model parameters described in this study. Based on the rural typology analysis described in the previous section, four trip production cross-classification matrix sets were prepared. As with long-distance rates, socioeconomic data used in the cross-classification scheme are income (three categories) by household size (five categories). Each set includes separate trip rates for home-based work, home-based nonwork, and nonhome-based purposes, as well as separate rates for each substrata included with the typology. The substrata for each cross-classification set are depicted below (the underlined strata reflect the rural components/subsets of each set): • HBHUR—Town and Country, Suburban Areas, Secondary City, Urban All; • URBAN—Not in Urbanized Area, Urbanized Area; • URBRUR—In Rural Area, In Urban Area; and • HBRESDN—Low Density (0–999 units/square mile), Medium Density (1,000–9,999 units/ square mile), High Density (10,000+ units/square mile). Although the definition of what is predominantly rural changes from one attribute/set to another, if the underlined categories above are compared, there is little difference in the trip rates. For example, the total number of person trips per household for all trip purposes is 9.72 for URBRUR, 9.83 for URBAN, and 9.56 for HBHUR. A more significant difference is shown using HBRESDN with a total trip rate of 11.76, although it is not clear how much of this lower housing unit density category consists of rural households. Appendix H depicts trip rates for

trip Generation parameters and Benchmark Statistics 61 URBRUR URBAN HBRESDN Housing units per sq mile – Block group 0-999 (75,937) 1,000-9,999 (52,450) 10,000-999,999 (2,120) Household in urban/rural area Home address in urbanized area Rural (38,014) Not in an urban area (38,014) (In an urban cluster – placeholder) N/A N/A N/A Urban (92,493) In an urban area (79,569) In an area surrounded by urban areas (51)a In an urban cluster (12,873) Household in urban/rural area Urban/Rural indicator – Block group Rural (38,014) Town and Country (38,014) Suburban (N/A) N/A N/A N/A Second City (N/A) N/A N/A N/A Urban (N/A) N/A N/A N/A Urban (92,493) Town and Country (24,227) N/A N/A N/A Suburban (30,491) Second City (23,550) Urban (14,225) Size of urban area in which home address is located Urban/Rural indicator – Block group Not in an urbanized area (50,938) Town and Country (50,938) Suburban (N/A) N/A N/A N/A Second City (N/A) N/A N/A N/A Urban (N/A) N/A N/A N/A All other categories combined – AKA urbanized (79,569) Town and Country (11,303) N/A N/A N/A Suburban (30,491) Second City (23,550) Urban (14,225) Note: Numbers in parentheses represent 2009 NHTS sample sizes for each category. Sample size numbers were found to be somewhat inconsistent among different urban/rural attributes and categories. “N/A” reflects an attribute combination that does not exist/would be illogical. a Probably should merge with “In an urban area” due to small sample size. Table 4.14. Recommended rural typology variables.

62 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models each category of every variable analyzed for estimating transferable trip production rates. The tables in Appendix H include both rural and urban trip rates using definitions unique to each attribute/rate set. 4.3 Long-Distance Trip Generation Model Parameters For the purposes of recommending transferable long-distance parameters, it was decided to focus on the 1995 ATS due to its larger sample size and based on the similarity of trip frequencies by purpose between the ATS and 2001 NHTS long-distance component. Although there are a multitude of ways these parameters can be summarized, this report uses the following considerations in reporting long-distance model parameters: • Include all days of the week (weekends and weekdays); • Report parameters on an annual (rather than daily) basis; • Exclude trips less than 100 miles in length; • Limit analysis to domestic travel (no international trips); and • Report at the household level rather than person level (per capita rates). The primary reason for including weekday and weekend travel is that ATS weekend travel includes weekday trips that include a weekend component. Furthermore, average annual daily traffic (AADT), by definition, includes both weekdays and weekends. Annual, rather than daily, trips are reported because this is how ATS trips were reported. Trips less than 100 miles in length were excluded because the ATS did not include these trips. International travel was excluded because these trips are not included in most statewide passenger models and these trips tend to skew trip length. Finally, household, rather than per capita, trip rates were selected because these are more commonly found in four-step travel demand models. All transferable parameters are calculated for the predominant three trip purposes (Business, Pleasure, and Personal Business) and total trips (All Purposes). The transferable parameters are described below and in subsequent chapters by model step. Trip Generation: Long-Distance Person Trip Production Rates For transferable long-distance trip production rates, it was decided to cross-classify socio- economic characteristics of each household in a comparable manner to NCHRP Report 716. The recommended cross-classification scheme for long-distance trip production rates is household income by household size. The correlation between income and long-distance trip-making is significant. For the purposes of cross-classification, household size is stratified into five categories, similar to NCHRP Report 716, whereas household income was collapsed into three categories. It also was decided to report annual trip rates since daily and monthly trip rates resulted in very low values. Table 4.15 depicts recommended long-distance person trip production rates for the three trip purposes. Since this study is primarily focused on passenger travel, the reader should refer to NCHRP Report 716 for a summary of sample truck trip generation rates derived from multiple sources. Appendix E of this report also provides metrics on rural versus urban truck travel. 4.4 Rural Trip Generation Model Parameters All rural travel parameters summarized in this section of the report were derived from sta- tistical analysis of 2009 NHTS datasets. As with long-distance parameters, the rural parameter discussion is divided into separate chapters reflecting each step in the typical model chain.

trip Generation parameters and Benchmark Statistics 63 Trip Generation: Rural Trip Production Rates From each of the four trip rate stratifications described in Section 4.2, URBAN showed the fewest trip rate anomalies (cells having higher or lower trip rates than expected compared to adjacent cells). Table 4.16 depicts unchained rural trip rates using URBAN as the 2009 NHTS attribute to differentiate between rural and urban households. Only two minor anomalies were identified in this table. HBW trip rates for highest-income four-person households and NHB 5+ person households were initially lower than found in adjacent trip rate cells (those with lower-income or household size). Rates were subsequently adjusted and smoothed for these two cells. The resulting total person trip rate per rural household using the URBAN attribute is 10.06, as opposed to an urbanized area trip rate of 9.91 as depicted in Appendix H. Urbanized area trip rates do vary by subcategory, such as secondary cities (9.50), suburban (10.34), and non suburban or second city urbanized (9.36). Some of these differences could possibly be minimized through testing of alternate socioeconomic cross-classification schemes. Trip rate comparisons found in Appendix G show a strong correlation between housing density and trip rates, with lowest hous- ing density trip rates being the highest at 9.60 and highest density housing trip rates the lowest at 7.77. In theory, opportunities for mixed-use development are more prevalent in higher density areas, thus reducing the trip rate, as opposed to lower density areas where mixed uses are less common, resulting in more trip-making to satisfy household needs. It is also possible that some of these differences might be explained by differences in household size, with testing of alterna- tive cross-classification schemes. Income by HH Size 1 2 3 4 5+ Total Business Trip Rates by Household Size/Income 01 <$25,000 0.44 0.80 0.88 0.99a 1.59 0.66 02 $25,000-$99,999 2.34 2.62 2.97 3.24 3.80 2.90 03 $100,000+ 3.70 8.00a 8.20 a 8.40 a 8.54 8.61 Total 0.99 2.35 2.70 3.24 3.51 2.28 Pleasure Trip Rates by Household Size/Income 01 <$25,000 1.70 3.49 3.50 a 5.10 5.15 2.77 02 $25,000-$99,999 3.88 6.70 8.42 9.77 11.93 7.84 03 $100,000+ 4.76 11.51 14.15 18.91 21.27 14.59 Total 2.32 6.03 7.48 9.46 10.77 6.21 Personal Business Trip Rates by Household Size/Income 01 <$25,000 0.41 1.12 1.16 1.51 2.05 0.84 02 $25,000-$99,999 1.03 1.63 2.15 2.66 3.75 2.08 03 $100,000+ 0.46 2.56 2.78 3.99 4.79 3.07 Total 0.58 1.53 1.94 2.53 3.38 1.66 Total Annual Trip Rates by Household Size/Income 01 <$25,000 2.54 5.41 5.48 7.28 8.79 4.27 02 $25,000-$99,999 7.25 10.95 13.54 15.67 19.48 12.82 03 $100,000+ 8.92 22.87 24.72 34.03 34.60 26.26 Total 3.89 9.91 12.12 15.22 17.66 10.15 a Indicates where estimated trip rates were manually adjusted and smoothed. Table 4.15. Annual long-distance person trip production rates.

64 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models Cross-classification matrices are also provided for auto availability and number of workers, con- sistent with cross-classification schemes documented in NCHRP Report 716. Rural trip production rates for auto availability by household size are found in Table 4.17, while number of workers by household size are depicted in Table 4.18. Overall, Table 4.17 (auto availability) has slightly lower total weekday trip rates than Table 4.16 (income). There are only three instances where the reverse is true, households with 5+ members for all purposes and NHB; and households with 3 members for HBW. Whereas Table 14.12 produced four trip rate anomalies, there are only two small anomalies in Table 14.13, depicted in underlined italics. HH Person Trip Rates: In Other (Rural) Areas, All Trip Purposes Trip Rates HH Size Income 1 2 3 4 5+ Total Less than $25,000 2.8 6.4 9.9 15.0 15.6 6.8 $25,000-$99,999 4.2 7.9 12.8 17.5 22.1 10.2 Above $100,000 5.1 8.8 14.0 20.1 26.2 14.0 Total 3.6 7.8 12.5 18.0 20.9 10.0 In Other (Rural) Areas HBW Trip Rates HH Size Income 1 2 3 4 5+ Total Less than $25,000 0.2 0.8 1.3 1.5 1.7 0.7 $25,000-$99,999 0.7 1.2 2.1 2.4 2.5 1.5 Above $100,000 0.9 1.6 2.3 2.4a 2.6 1.9 Total 0.5 1.2 2.0 2.2 2.3 1.4 In Other (Rural) Areas HBNW Trip Rates HH Size Income 1 2 3 4 5+ Total Less than $25,000 1.6 3.8 5.9 9.7 10.8 4.2 $25,000-$99,999 1.9 4.0 6.8 10.1 13.8 5.6 Above $100,000 2.2 4.0 7.2 11.4 15.2 7.4 Total 1.8 4.0 6.7 10.4 13.1 5.6 In Other (Rural) Areas NHB Trip Rates HH Size Income 1 2 3 4 5+ Total Less than $25,000 0.8 1.7 2.6 3.8 4.0a 1.7 $25,000-$99,999 1.5 2.6 3.9 4.9 5.7 3.1 Above $100,000 1.9 3.1 4.4 6.4 8.3 4.6 Total 1.3 2.5 3.8 5.2 5.4 3.0 Source: 2009 NHTS. a Indicates where estimated trip rates were manually adjusted and smoothed. Table 4.16. Rural person trip production rates: HH size by income, URBAN attribute identifies rural HHs.

trip Generation parameters and Benchmark Statistics 65 In Other (Rural) Areas, All Trip Purposes Trip Rates HH Size Autos/HH 1 2 3 4 5+ Total 0 Veh 2.3 6.2 9.1 12.0 13.5 4.9 1 Veh 3.8 6.9 11.4 14.4 15.8 6.4 2 Veh 4.6 7.9 12.4 18.2 21.4 11.8 3+ Veh 4.6 8.1 13.8 19.5 25.1 15.3 Total 3.6 7.6 12.4 17.8 20.9 9.7 In Other (Rural) Areas HBW Trip Rates HH Size Autos/HH 1 2 3 4 5+ Total 0 Veh 0.2 0.7 1.2 1.1 1.2* 0.5 1 Veh 0.5 0.8 1.2 1.5 1.4 0.8 2 Veh 0.6 1.4 2.0 2.1* 2.2 1.6 3+ Veh 1.0 1.5 2.6 3.0 3.3 2.5 Total 0.5 1.2 2.0 2.2 2.3 1.3 In Other (Rural) Areas HBNW Trip Rates HH Size Autos/HH 1 2 3 4 5+ Total 0 Veh 1.3 3.7 5.7 7.9 9.9 3.1 1 Veh 1.8 3.8 6.7 9.0 11.1 3.6 2 Veh 2.2 3.9 6.5 10.7 13.4 6.5 3+ Veh 2.3 3.9 7.1 10.8 14.6 8.3 Total 1.7 3.9 6.7 10.4 13.1 5.4 In Other (Rural) Areas NHB HH Size Autos/HH 1 2 3 4 5+ Total 0 Veh 0.7 1.7 2.2 2.9 3.0* 1.2 1 Veh 1.3 2.2 3.4 3.8 3.9* 2.0 2 Veh 1.7 2.5 3.8 5.4 5.7 3.6 3+ Veh 1.2 2.6 4.0 5.7 7.0 4.5 Total 1.2 2.4 3.7 5.2 5.4 2.9 Source: 2009 NHTS. * Indicates where estimated trip rates were manually adjusted and smoothed. Table 4.17. Rural person trip production rates: HH size by auto availability, URBAN attribute identifies rural HHs.

66 Long-Distance and rural travel transferable parameters for Statewide travel Forecasting Models In Other (Rural) Areas, All Trip Purposes Trip Rates HH Size HH Size 1 2 3 4 5+ Total 0 worker 2.796 5.863 8.456 11.721 12.494 4.952 1 worker 4.249 7.436 11.247 15.733 18.245 8.750 2 worker 0.000 9.438 14.136 19.545 23.806 14.610 3+ worker 0.000 0.000 16.316 23.564 27.678 22.544 Total 3.605 7.607 12.452 17.868 20.962 9.783 In Other (Rural) Areas HBW Trip Rates HH Size HH Size 1 2 3 4 5+ Total 0 worker 0.003 0.011 0.009 0.049 0.074 0.010 1 worker 0.966 1.148 1.298 1.460 1.500* 1.166 2 worker 0.000 2.403 2.752 2.665 2.688 2.580 3+ worker 0.000 0.000 4.993 4.808 5.414 5.063 Total 0.539 1.211 2.026 2.226 2.342 1.378 In Other (Rural) Areas HBNW Trip Rates HH Size HH Size 1 2 3 4 5+ Total 0 worker 1.931 4.088 5.603 8.807 9.688 3.483 1 worker 1.691 3.914 6.650 9.646 11.975 4.878 2 worker 0.000 3.825 7.125 11.152 15.113 7.655 3+ worker 0.000 0.000 6.637 11.773 14.417 10.963 Total 1.797 3.937 6.700 10.434 13.139 5.453 In Other (Rural) Areas NHB HH Size HH Size 1 2 3 4 5+ Total 0 worker 0.862 1.764 2.844 2.865 2.900* 1.459 1 worker 1.593 2.374 3.299 4.628 4.832 2.706 2 worker 0.000 3.211 4.259 5.728 6.005 4.376 3+ worker 0.000 0.000 4.687 6.983 7.847 6.518 Total 1.269 2.459 3.726 5.209 5.482 2.951 Source: 2009 NHTS. * Indicates where estimated trip rates were manually adjusted and smoothed. Table 4.18. Rural person trip production rates: HH size by number of workers, URBAN attribute identifies rural HHs.

trip Generation parameters and Benchmark Statistics 67 The percentage of rural trips by purpose could be a useful statistic for use in model valida- tion and reasonableness checking. Assuming the weighted number of surveys for 2009 NHTS adequately reflects the share of rural versus urban trips, the percentage of rural trips by purpose has been summarized in Table 4.19. Since the definition of rural varies somewhat from one NHTS attribute/typology to another, the number and percentage of trips by purpose was esti- mated for each of these typologies and subsequently averaged. Regardless of the attribute used to identify rural households, the results show a considerably smaller percentage of home-based work trips than commonly found in urban areas. This is not entirely surprising in a population that generally consists of a higher-than-average share of farmers, retirees, and unemployed, as well as above average household sizes (http://205.254.135.7/emeu/recs/recs2005/hc2005_tables/ hc3demographics/pdf/tablehc8.3.pdf). Typology/NHTS 2009 Attribute URBANR – Other (Not Urbanized) HBHUR – Town and Country HBRESDEN – 0-999 Units/ Square Mile URBRUR – Rural Areas Average Trip Purpose No. of Trips % No. of Trips % No. of Trips % No. of Trips % % Rural Home- Based Work 63,057 11.82 12.03 23,194 12.60 29,983 12.26 12.06 Rural Home- Based Nonwork 308,005 57.74 218,398 54.41 96,301 52.31 129,875 53.09 55.19 Rural Nonhome- Based 162,405 30.44 134,711 33.56 64,619 35.10 84,761 34.65 32.74 Non Urban Totals – All Purposes 533,467 100.00 401,388 100.00 184,114 100.00 244,619 100.00 100.00 Source: 2009 NHTS. Table 4.19. Rural trips by purpose.