Developing, Enhancing, and Sustaining Tribal Transit Services: A Guidebook (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.

30 Introduction A demographic profile of the community is important to understanding the different types of market segments a public tribal transit service will need to serve. It gives a better understand- ing of the community and helps with estimating needs and forecasting demand for passenger transportation in the community. Demographic Analysis Population Estimates Data from the most recent U.S. Census can be used as base data for population estimates. Updated information from local and tribal transportation planning offices and state agencies can be used to amend the base data to reflect the most recent and projected information. Using both types of sources will give a more accurate estimate of the total population served by an existing transit service or that will be served by a proposed transit service. Transit-Dependent Population Characteristics Transit-dependent populations are individuals for whom ride sharing, public transit, and other community transportation options are the only forms of motorized transportation available. Typically, three types of limitations prevent persons from owning or driving a private automobile: 1. Physical limitations 2. Financial limitations 3. Legal limitations Physical limitations range from permanent disabilities, such as frailty due to age, visual impairment, paralysis, or developmental disabilities, to temporary disabilities, such as acute ill- nesses and head injuries. Financial limitations relate to the inability to purchase or rent a vehicle because of cost. Legal limitations typically relate to restrictions on an individual’s maintaining a driver’s license, such as suspension or revocation due to driving under the influence (DUI), or on obtaining a license, such as age (generally under age 16). Data from the U.S. Census is easily available and contains general information about these three categories of limitation. Population numbers obtained from the census need not be pre- cise, as they will be used to develop estimates of demand. Census information is available about elderly, mobility-limited, below-poverty, and youth populations and zero-vehicle households. Elderly population: This category is defined as the population of individuals above 60 years of age. C h a p t e r 4 Transportation Needs Assessment

transportation Needs assessment 31 Mobility-limited population: This category is defined as individuals above 16 years of age with an impairment that limits their ability to leave their home without assistance. Below-poverty population: This category includes all individuals from households with income levels reported as below the poverty threshold. Youth population: This category includes individuals less than 18 years of age. Zero-vehicle households: This category includes households that do not have possession of a vehicle in working order. Assembling data for each of these population categories results in a good estimate of the dependent population living in the community. Details about how to access this information appear in the “Tools” section of this chapter. Native American Population Estimates For federally recognized tribes that have a reservation boundary and/or off-reservation trust lands, Native American population estimates can be calculated by looking up 2010 Census data and/or the 2006 to 2010 American Community Survey (ACS) 5-year estimates under the sum- mary level “tribal block groups” or “tribal census tracts.” These summary levels will have infor- mation for most tribal transit-dependent populations. The 2006 to 2010 ACS is the first data release to include estimates for tribal census tracts and tribal block groups. Boundary differences exist between a non-tribal block group and a tribal block group. A stan- dard block group does not cross boundaries of states, counties, or statistical designated bound- aries, but may cross boundaries delineated by American Indian tribal authorities. On the other hand, a tribal block group is independent of the standard block group defined within counties and may cross county and state boundaries. They do follow the same population and housing unit thresholds as standard census tracts and block groups. Tribal census tracts and tribal block groups are available only on reservations and/or off-reservation trust lands that have a popula- tion equal to or more than 2,400 and 1,200, respectively. If a tribe does not have a reservation boundary, the tribal census tract/tribal block group unit cannot be used. In such cases, information is available only at a block group or a census tract level, and specific table numbers provide transit-dependent population information for all American Indian and Alaska Natives (AIAN) living in the area. Figure 4.1 provides an example of how Native American populations can be mapped to show their densities graphically. Tourists and Visitors Tourist populations may increase the need for a transportation service. If the transit service area serves a national park, information about visitors and traffic is available at the National Park Service Public Use Statistics Office website at http://www.nature.nps.gov/stats/. Information about tourists and visitors not attached to a national park may be available through the Chamber of Commerce or through a local planning agency. Tools Several tools that can be used for estimating population and transit-dependent populations are available at • the decennial census and • the American Community Survey (ACS).

32 Developing, enhancing, and Sustaining tribal transit Services: a Guidebook Both the decennial census and the ACS data can be obtained from the main census web- site at www.census.gov. Census may periodically change its website. Usually, under “American FactFinder,” you will be able to find specific datasets such as the 2010 Census or various ACS datasets. You will then have to select the relevant summary level (tribal census tract, census tract, or whatever defines your geographical transit service area). The “tribal block groups” and the “block groups” are the smallest geographic units for which transit-dependent population data are available. You will also have the option to map each geographic area by selecting the “map” tab. Currently, various data at the census block groups can be accessed using the Summary File Excel Retrieval tool, which can be downloaded from the census website. The descriptions for each of these categories under “tribal block group” or “tribal census tract” are listed as follows: • Total Population. • Sex by Age. Select the data for both male and female categories age 60 years and older. • Persons with a Disability. Obtain information for “Population 16 to 64 Years” with a “Go- Outside Home Disability” if you are doing a public transit service or add other types of dis- ability depending on the market segment you plan to serve. • Poverty Status in the Past 12 Months by Sex by Age. Select the population with income below the poverty line. • Tenure by Vehicles Available. Select data from both owner-occupied and renter-occupied housing units with no vehicles available. Figure 4.1. Native American population in the Pala Reservation area.

transportation Needs assessment 33 For tribes without a reservation boundary, information on Native American transit- dependent populations is available by selecting the following descriptions under “census tracts” or “census block groups.” As mentioned, you will have to select certain pieces of information from these tables based on the information needed. • Sex by Age (AIAN alone). Select the data for both male and female categories for age 60 years and older. • Persons with a Disability (AIAN alone). Obtain information for “Population 16 to 64 Years” with a “Go-Outside Home Disability” if you are doing a public transit service or add other types of disability depending upon the market segments you plan to serve. • Poverty Status in the Past 12 Months by Sex by Age (AIAN alone). Select the population with income below poverty line. • Tenure by Vehicles Available (AIAN alone, by householder). Select data from both owner- occupied and renter-occupied housing units with no vehicles available. The ACS is conducted yearly, so it has more current information, but the margin of error is high compared to the decennial census. You can access information from the census website and select the most recent year for which information is available. For most rural areas, the 5-year estimates by tribal block group/block group level are the most appropriate. Geographic Information Systems Geographic information systems (GIS) provide another effective tool for displaying the spa- tial distribution of population and other transit-dependent population segments. GIS can also help with mapping major transit trip generators like hospitals, clinics, senior centers, shopping centers, recreation areas, and employment centers in order to analyze the linkages between residential areas and the major employment centers on and off the reservation. Transit Demand Methodologies This section of the chapter examines various transit demand methodologies. Where appli- cable, links to the original documents are provided for additional reading and access to other tools to aid in demand estimation. Not all methods discussed will be needed for every system. Program Trips A methodology developed in TCRP Report 3 and currently being updated in TCRP Proj- ect B-36 (with a project report to come) allows users to forecast program trips for market segments of the population. The project report, tentatively titled Methods for Forecasting Demand and Quantifying Need for Rural Passenger Transportation, is expected to be published in 2012. Program trips occur because of the presence of specific social service programs, including Head Start, day habilitation services, and programs offered by senior living centers. A two-step process is used to generate demand estimates for program trips. The first step of the process uses census information to estimate the number of participants within a specific program (i.e., mental health counseling, group homes, meal programs, etc.). The second step is to apply a trip rate to the number of participants in that program. This allows for an annual number of trips to be calculated for each individual program. Although this model is fairly straightforward, it requires a significant amount of background demographic information. Table 4.1 presents an

34 Developing, enhancing, and Sustaining tribal transit Services: a Guidebook example of the kind of data accumulated using this methodology, as applied to a project for the Eastern Band of Cherokee Indians: More information on this model, including a workbook to aid in the calculations, can be retrieved from TCRP Web-Only Document 49: Methods for Forecasting Demand and Quantifying Need for Rural Passenger Transportation, available online at: http://onlinepubs.trb.org/onlinepubs/ tcrp/tcrp_webdoc_49.pdf. ADA Demand Model Estimating the demand for Americans with Disabilities Act (ADA) complementary paratran- sit service is an important part of the transit demand process. TCRP Report 119: Improving ADA Complementary Paratransit Demand Estimation established a demand estimation tool developed from statistical analysis of transit systems across the country. The model uses peer-comparison data along with multiple factors to predict paratransit ridership. The input variables include population, percentage of households below the poverty line, and fare. More information about this model, including an Excel spreadsheet that aids in the comple- tion of the data, can be found in TCRP Report 119, accessible at http://www.trb.org/Main/Public/ Blurbs/159293.aspx. Rural Public Transit Demand A technique for estimating rural public transit demand was presented in a report from TCRP as part of Project B-3. TCRP Report 3: Workbook for Estimating Demand for Rural Passenger Transportation presents the methodology described in this guidebook. The study was based on a database of 185 transit agencies across the country. The model requires inputs for elderly persons, mobility-limited and low-income populations, as well as the assumed level of service (vehicle-miles). The model looks at the relationship between the level of service provided and the resulting trip rates of the three market segment populations. Table 4.2 illustrates this model. Once again, results are provided for the Eastern Band of Cherokee Indians. The assumed level of service for this model was 2,500 vehicle-miles per square mile of study area annually. Program Type Estimated # of Participants/Clients Annual One- Way Trips Developmental services Adult 20 participants 7,071 Case management 20 participants 773 Pre-school – 3 to 5 yrs 13 participants 2,831 Head Start 70 participants 18,431 Job training 36 clients 4,901 Mental health services 70 clients 24,117 Nursing home 13 participants 122 Senior nutrition 34 participants 8,880 Sheltered workshop 30 participants 11,617 Group home 10 participants 6,453 Program trips 85,196 Source: LSC, 2010. Table 4.1. TCRP methodology, program trips.

transportation Needs assessment 35 An alternative technique for estimating rural public transit demand was developed as part of TCRP Project B-36 to give a rough estimate based on demographic characteristics alone using the formula Non-program Demand trips per year Po( ) = ×2 20. pulation Age 60 Mobility Limited P +( ) + ×5 21. opulation Age 16 to 64 Residents of ( ) + ×1 52. Households Having No Vehicle( ). Table 4.3 illustrates the demand estimate using this model from the TCRP B-36 project. Urban Public Transit Demand A fixed-route demand model can be used to analyze whether existing transit service is meet- ing the community’s needs based on the type of service. The model is based on household vehi- cle ownership, average walking distance to bus stops, and frequency of operation. The basic approach is described in Demand Estimating Model for Transit Route and System Planning in Qualla Boundary: Estimated Annual Passenger-Trip Demand Elderly + Mobility- Limited Estimated Daily Census Block Mobility- Limited Transit Demand County Tract Group Elderly Income Total # % Jackson 9501 1 2,360 1,460 3,820 2,550 6,370 25 23.8% Jackson 9501 2 1,610 320 1,930 1,370 3,300 13 12.3% Jackson 9501 3 2,470 130 2,600 940 3,540 14 13.2% Swain 9601 1 870 850 1,720 2,580 4,300 17 16.0% Swain 9601 2 1,240 200 1,440 2,730 4,170 16 15.6% Swain 9601 3 2,390 620 3,010 2,110 5,120 20 19.1% TOTAL 10,940 3,580 14,520 12,280 26,800 105 Source: U.S. Census, 2000; Office of State Budget and Management, 2009 estimate; LSC, 2010. Table 4.2. 2010 estimated public transit demand using the TCRP B-3 method. Qualla Boundary County Census Tract Block Group Elderly Mobility- Limited Zero- Vehicle Households Average Household Size Residents of Households Having No Vehicle Estimated Annual Transit Demand (#) Jackson 9501 1 326 273 69 2.6 179 2,409 Jackson 9501 2 223 60 81 2.7 215 1,128 Jackson 9501 3 341 24 32 2.4 78 996 Swain 9601 1 121 158 51 2.5 126 1,283 Swain 9601 2 172 38 83 2.3 189 861 Swain 9601 3 330 115 35 2.7 97 1,474 Total 1,513 668 351 883 8,151 Source: U.S. Census, 2000; Office of State Budget and Management, 2009 estimate; LSC, 2010. Table 4.3. 2010 estimated public transit demand using the TCRP B-36 method.

36 Developing, enhancing, and Sustaining tribal transit Services: a Guidebook Small Urban Areas (Golenberg and Pernaw 1979). This model incorporates factors for walking distance, distance traveled on the bus, and frequency of service or headway. The multidimensional approach used for this model yields fairly accurate results. The model works best when an existing service level is established so that the model may be calibrated to the current level of service. All future service changes can then be evaluated using the general trip rate established by the model. Although the source document for this model is out of print, photocopies can be ordered. Information on ordering copies and finding library sources of the document is available at: http://www.trb.org/Publications/Pages/262.aspx. Intercity Demand Model To estimate demand for intercity bus service, a model was taken from the report Planning Techniques for Intercity Transportation Services. In general, this model represents the number of passengers traveling one-way on a given route as a function of the frequency of service, the population served, the cost to the rider, and the distance of the trip. The format of the model that proved to be appropriate is: PASS MO CONST RTFREQ SERVPOP FARE MIa b c= × × × where: PASS/MO = the number of one-way passengers boarding per month for the route segment specified; CONST = a constant specifically derived for this equation; RTFREQ = scheduled round-trips per week on the route; SERVPOP = the population served (defined as the sum of the populations of villages, towns, and cities directly along the route, divided by 100); FARE/MI = fare per mile in cents, found by dividing the cost of a one-way fare between the end points of each route by the one-way distance between the end points of the route; a = the exponent for round-trip frequency; b = the exponent for service population; and c = the exponent for fare per mile. Intercity trips of varying lengths differ in terms of trip purpose and frequency. This equation can be applied to estimate the potential demand for services between the various communities. One advantage of this model is that the intercity bus demand model takes into account an actual route along with factors like frequency and cost. Three different models are used, depending on the distance of the trip. The three models are: 1) PASS/MO (20–60 miles) = 17.989 × RTFREQ(1.032) × SERVPOP(0.376) × FARE/MI(-0.645) 2) PASS/MO (20–120 miles) = 6.871 × RTFREQ(1.093) × SERVPOP(0.409) × FARE/MI(-0.352) 3) PASS/MO (121+ miles) = 1.510 × RTFREQ(0.415) × SERVPOP(0.726). Rural Intercity Demand Model A methodology recently developed as part of the TCRP B-37 project allows users to forecast demand for rural intercity bus service. The methodology is outlined in the TCRP Report 147: Toolkit for Estimating Demand for Rural Intercity Bus Services. A CD-ROM included with the print version of the report also is available for download from TRB’s website at http://www.trb.org/ Main/Blurbs/165858.aspx.

transportation Needs assessment 37 The CD-ROM is self-contained and includes all the required census data and directions. It does not require access to specialized data or software like GIS and is sensitive to a variety of factors that could affect rural intercity bus demand. The model requires a computer capable of running Microsoft Excel. The report presents two methods—the regression model and the trip rate model. The regression model uses the following basic formula: Annual Ridership = –2,803.536 + 0.194 (average origin population) + 314.734 (the number of stops on the route) + 4971.668 (yes to airport service/connections) + 5783.653 (yes to service provided by an intercity provider). R Adjusted R2 2= =0 712 0 690. , . . The trip rate model uses data from the National Household Travel Survey. The base data uses information about long-distance trips of 50 one-way miles or more. It provides information by urban and rural categories, by region (census divisions), and by three categories of income groups—persons earning under $30,000, under $75,000, and over $75,000 per year. Commuter Demand The demand estimation technique established by the B-36 project involves applying a trip rate to the number of workers traveling between counties for work. The formula was developed using data from numerous commute locations. The resulting formula is as follows: Commuter trips by transit from County A to County B person work trips.= ×0 012. This formula breaks down to roughly one transit trip for every 83 trips into the region for work. Because most individuals need to make two commuter trips daily, this breaks down to roughly one transit trip for every 42 workers traveling to the county for work. Infor- mation on the commuting patterns of individuals can be obtained from the U.S. Census Bureau’s Longitudinal Household Employment Dynamics webpage at http://lehd.did.census. gov/led/. Need Assessment Survey Tool Reasons vary for launching surveys. Some surveys are created to do a community need assess- ment, some are focused on a specific passenger transit service, and still others are designed to inventory existing transportation resources and measure interest in coordinating transporta- tion services. The community survey may focus on the interest of the community in supporting public transportation or may focus on specific transportation for employment, medical, or other purposes. The passenger transit survey is meant to capture levels of customer satisfaction and get input on how transit services can be improved. The passenger transit survey can be in the form of intercept or onboard surveys. TCRP Synthesis 63: On-Board and Intercept Transit Survey Techniques describes how to administer onboard and intercept surveys. The transportation pro- vider survey can be used to get input from agencies that have a transportation service or those that have a need for transportation services. These surveys help in determining the community needs demographics and travel patterns. A community needs assessment survey is useful for getting input but not for estimating demand. Such a survey can be conducted by phone, mail, Internet, in-person interview, or through distributed handouts. No single delivery method works best. The method depends on whether most of the people surveyed have Internet access or if the information could be given out at a common place or event like a community center, Pow-Wow, or other community event. In some cases it may be necessary to combine several delivery methods to get the information.

38 Developing, enhancing, and Sustaining tribal transit Services: a Guidebook It is important to test the questionnaire on a small group of people before distributing it to the community to see whether people understand the questions or to identify mistakes in the survey design. Figure 4.2 presents a sample community needs survey. For More Information SG Associates, Inc., with Leigh, Scott & Cleary, Inc., and C. M. Research, Inc. TCRP Report 3: Workbook for Estimating Demand for Rural Passenger Transportation. TRB, National Research Council, Washington, D.C., 1995. http://www.trb.org/Publications/Blurbs/153843.aspx. Koffman, D. et al. TCRP Report 119: Improving ADA Complementary Paratransit Demand Estima- tion. Transportation Research Board of the National Academies, Washington, D.C., 2007. http:// www.trb.org/Publications/Blurbs/159293.aspx. Figure 4.2. Sample community survey.

transportation Needs assessment 39 Spielberg, F., A. T. Stoddard, and J. Erickson. TCRP Web-Only Document 49: Methods for Forecasting Demand and Quantifying Need For Rural Passenger Transportation. Transporta- tion Research Board of the National Academies, Washington, D.C., 2009. http://www.trb. org/Publications/Blurbs/163867.aspx. Fravel, F. D., J. E. Burkhardt, and R. E. Menzer. Planning Techniques for Intercity Transporta- tion Services. Office of Technology and Planning Assistance, Office of the Assistant Secretary for Government Affairs, U.S. Department of Transportation, 1987. http://isddc.dot.gov/OLPFiles/ FTA/015511.pdf. Fravel, F. D., et al. TCRP Report 147: Toolkit for Estimating Demand for Rural Intercity Bus Services, Transportation Research Board of the National Academies, Washington, D.C., 2011. http:// www.trb.org/Publications/Blurbs/165858.aspx. Golenberg, M., S. Pernaw. Demand Estimating Model for Transit Route and System Planning in Small Urban Areas. In Transportation Research Record 730, TRB, National Research Council, Washington, D.C., 1979, pp. 14–23. Schaller, B. TCRP Synthesis 63: On-Board and Intercept Transit Survey Techniques: A Synthesis of Transit Practice. Transportation Research Board of the National Academies, Washington, D.C., 2005. http://www.trb.org/Publications/Blurbs/156542.aspx.