Using the NHTS
The NHTS user guide is available here.
The Codebook is available here.
The Questionnaire is available here.
Here is the scheme for how the files relate to each other:
Here are the IDs that correspond to different entities.
| Variable | Contents | Notes |
|---|---|---|
| HOUSEID | Household ID number | |
| PERSONID | Person ID number | |
| VEHID | Vehicle ID number | |
| TDTRPNUM | Trip number |
From the NHTS user guide:
Each unit (e.g., households, persons) in the survey has its unique identification number (ID). Specifically, each household is identified by a unique eight digit household ID (HOUSEID). Within each household, household members are identified by a two digit person number (PERSONID) and, similarly, household vehicles are identified by a two digit vehicle number (VEHID). Again, trips taken by an individual are numbered by a trip number (TDTRPNUM) for a travel day trip.
An example of how to link files:
| Code | Value | Notes |
|---|---|---|
| Legitimate skip | -1 | |
| Refuse | -7 | |
| Don't know | -8 | |
| Not ascertained | -9 |
Other:
Data type | Notes | ----|----|---- Times | Military times 0000 to 2359 Yes/No | 1=yes, 2=no
Chapter 5 described how the weights were calculated for the 2009 survey. The weights reflect the selection probabilities and adjustments to account for nonresponse, undercoverage, and multiple telephones in a household. To obtain estimates that are minimally biased, weights must be used. Note that the 2009 NHTS sample was designed to oversample households in 14 states and 6 metropolitan planning areas. The weighting process adjusted that oversampling to provide correctly balanced estimates by geographic area. Because the weighting also involved adjustments for demographic factors, such as household size, race and ethnicity, tabulations without weights may be significantly different than weighted estimates and may be subject to large bias. Estimates of the totals are obtained by multiplying each data value by the appropriate weight and summing the results. Which Weight to Use. There are several different weights, and it is important that the appropriate weight is used for a particular estimate. There are household weights, person weights and travel day weights.
- Household weights (WTHHFIN) are used whenever one is tabulating an estimate at the household level as opposed to the person level, such as number of households by household vehicle ownership and distribution of households by number of household drivers.
- Vehicle weights (WTHHFIN) are the same as the household weight since the vehicle is considered a household attribute. Use the vehicle weight for items such as vehicles by vehicle type or by vehicle age.
- Travel day weights (WTTRDFIN) are used for estimates involving numbers of trips or miles of travel, for example, number of vehicle trips by trip purpose. Only trips in privately-owned vehicles that are reported by the driver should be counted in estimating vehicle trips. For example, if a person reports being a passenger in a vehicle driven by another household member, that trip would not be counted.
- Person weights (WTPERFIN) are generally used for estimates of non-household and non-travel day items of interest, for example workers by gender, drivers by annual miles estimated, etc.
Note that for some estimates requiring ratios, different weights should be used for the numerators than for the denominators. For example, for estimates of daily trips per household, travel day weights are used for the numerator (since the numerator involves person trips) and household weights are used for the denominator (since the denominator is the weighted number of households). As a second example, for estimates of average time spent driving by all drivers, travel day weights are used for the numerator and person weights are used for the denominator (since drivers are a subset of persons).