Methodology
Return to 2009 Local Area Transportation Characteristics by Household (LATCH) methodology
Census Region/Division Groups
The increased number of households sampled in the 2009 NHTS, a national sample of 25,000 and 125,000 ‘add-on’, or oversampled, geographic areas, allows the sample to be divided up into six census region/division groups, and then subsetted into three urban groups, for a total of 18 separate categories. Each category is estimated separately. The geographic disaggregation was done to make more homogenous groupings of the households for the regression equations. The specific groupings were defined to keep an adequate sample size for estimation purposes. The census region/division groupings used were the following:
1. Northeast Region
2. Midwest Region
3. South Atlantic Division
4. East South Central Division and West South Central Division
5. Mountain Division
6. Pacific Division.
Development of Urbanicity Index
The 2009 NHTS includes the set of Census tract and block group variables known as the Claritas Variables that describe the characteristics of the areas where the NHTS were surveyed. The urban-rural continuum variable, contained within this set of Census tract and block group descriptors, describes the level of urban development within an area in terms of the population density in the geographic area itself and in the surrounding areas. This overall measure of population density is used in labeling Census block groups as: urban, suburban, second city, and town/country. Block groups where one or more households were surveyed are assigned to one of these groups. Households were not surveyed in all block groups and consequently, a complete urban-rural continuum dataset, using the Claritas Variable, for Census block groups cannot be extracted from the NHTS[1].
The present analysis requires all U.S. Census tracts to be labeled in terms of their urbanicity. If a complete urban-rural continuum dataset could be obtained for Census block groups, a dataset for tracts could be created by calculating the mean or median value assigned to block groups contained by a given tract and assigning that value to the tract. Since a complete block group dataset cannot be obtained, a dataset cannot be created for tracts in this manner. For this reason, a new urbanicity measure was created and calculated for all tracts. This measure is based on the population density of a Census tract (converted to a centile score) and on whether the Census tract is in an urban area or urban region/division. The 2010 Census tract and urban boundaries were used in combination with 2010 Census information on population and land area[2]. The following shows the assignment of Census tracts to the following categories: urban, suburban, and rural (Table 5).
The number of NHTS households in each census region/division and urban group is given in the following table (6):
Mean and Confidence Intervals of Travel Variables
The objective of dividing the NHTS households into these 18 groups is to improve the accuracy and usefulness of the regression estimates. One way to assess the groupings is to look at the differences in means and confidence intervals for each travel variable. The means and confidence intervals are shown in Figures 1, 2, 3 and 4. Considerable variation can be observed, both across geographical divisions and between urban groups.
Definition of Travel Variables and Exclusions of Certain Households
Certain NHTS households were excluded in the analysis. Only households with a weekday travel day were used, all weekend trips were excluded. This follows the methodology used in the previous 2001 NHTS Transferability Project (Hu et al. 2007) and is a common assumption in urban planning models. As in the previous study, all households in Manhattan, New York, were excluded, due to the unique travel patterns and relationships to household characteristics of that area. There were also several outliers identified in some household trip data. In order to reduce the distortions of those outliers, approximately 1 percent of the trips in the upper tail of each distribution were excluded.
Person trips include all trips, except that using an airplane mode. Vehicle trips include trips using cars, vans, SUVs, pickup trucks, other trucks, RVs, motorcycles, and light electric vehicles. It includes only trips taken by the driver of the vehicle. Household trips represent the sum of all trips taken by members of the household.
Four travel variables were estimated:
1. Total household person miles traveled, excluding outliers > 500 miles
2. Total household number of person trips, excluding outliers > 30 trips
3. Total household vehicle miles traveled, excluding outliers > 310 miles
4. Total household number of vehicle trips, excluding outliers > 20 trips
Explanatory Variables
The selection of explanatory variables to be used in the analysis relied partially on previous work in the 2001 NHTS Transferability Study (Hu et al. 2007). In addition other NHTS household variables were examined for potential inclusion. This examination also included the requirement that comparable data be available in the Census ACS public data tables at the census tract level. This became a significant constraint in developing the life-cycle household variables. The NHTS defined life-cycle variables do not have equivalent counterparts in the ACS data tables. As a result, alternative life-cycle variables were constructed that could be used with the available ACS data tables. The final set of explanatory variables used in the analysis includes:
- Household income [HH Income or Nat. Log (HH Income)] This variable was converted from the household income categories in the NHTS data to a point estimate, using the mid-point of each category range. For the last category, household income above $100,000 , more detailed Census household income tables were used to derive a weighted average of $147,500 for that category.[3] The natural log of household income was also used in some cases to reflect the non-linear relationship sometimes observed between higher household income and trips taken. Household income is the best available proxy for household wealth, which is assumed to the primary driver of discretionary travel expenditure.
- Count of household vehicles [Count of HH Vehicles]
- Number of members in household [Count of HH Members]
- Homeowner (yes or no) [Homeowner]
- Number of workers in household [Number of Workers]
- Life-Cycle, 1 or more children in household, less than 18 years old [Life Cycle (1+C<18)]
- Life-Cycle, 1 person household, less than 65 years old [Life Cycle (1P hh<65)]
- Life-Cycle, 2 or more person household, all less than 65 years old [Life Cycle (2+P hh, 0 65+)]
- Life-Cycle, 2 or more person household, at least one 65 or more years old [Life Cycle (2+P hh, 1+65+)]
[1] Claritas data for all Census Tracts can be purchased, but that was not an option for this study. For complete information on the Claritas variables included in the 2009 NHTS, see: http://nhts.ornl.gov/2009/pub/UsersGuideClaritas.pdf
[2] 2010 Census Tract boundaries obtained from National Historical Geographical Information System: https://www.nhgis.org/ 2010 urban boundaries obtained from the 2010 Census TIGER/Line Shapefiles: https://www.census.gov/geographies/mapping-files/2010/geo/tiger-line-file.html. 2010 population and land area data obtained from the Census 2010 Census Tract Relationship File: https://www.census.gov/geographies/reference-files/2010/geo/relationship-files.html
[3] U.S. Bureau of Census, Current Population Survey, Table HINC-01, Selected Characteristics of Households, by Total Money Income in 2009. https://www.census.gov/data/tables/time-series/demo/income-poverty/cps-hinc/hinc-01.html