On August 8th and 9th, I had the opportunity to attend the 2018 National Household Travel Survey Workshop (NHTS) in Washington, DC. I presented a poster called Evaluating Alternative Transportation Revenue Regimes Using the NHTS Transferability Statistics, based on work EDR Group has done for the Western Road User Charge Consortium (RUC West). This NHTS data product from the 2009 survey allows us to estimate the travel behavior of households at a census tract scale across the U.S. To date we’ve examined 10 states to understand how revenue-neutral conversion from a gasoline excise tax to a road usage charge would affect different groups of households.
While I’ve used the NHTS for several different analyses at EDR Group—from comparing modal choices and trip characteristics to constructing time of day and weekly travel distributions—having only 2009 survey results was starting to be limiting. This conference was a great opportunity to discuss innovative uses of the data with other experts in the travel behavior field and learn about what’s new in the 2017 vintage data. Wednesday presented many opportunities to dive into the survey and its findings. Thursday focused on plans to make the NextGen NHTS more timely than past surveys while integrating with new non-survey data sources.
There are a couple observations from the 2017 NHTS that are interesting in regard to our work connecting transportation and the economy. Trip rates continue to decline (by another 10% from 2009) following the trend since the 1995 survey, but almost all the change is concentrated among shopping and errands trip purposes. Surprisingly, errands represent the much more significant decrease over time with shopping remaining relatively stable. This suggests that so far the digitization of services (e.g. e-banking, travel sites, etc.) has had a much bigger impact on travel behavior than has e-commerce.
One of the drivers of lower trip-making rates is higher levels of immobility among younger age cohorts, which could potentially reflect greater work flexibility offered by teleworking and the continued trend of online interactions replacing social and recreational trips. Trips have also fallen much more for higher income households.
We’re quite excited that the NHTS tool took advantage of network routing tools this year instead of relying on self-reported travel distances since this results in much more reliable trip distance estimates. After all, except for my regular commute, I rarely know how many miles a trip takes. If I commuted by transit instead of biking, I don’t even think I’d know that trip length. If even transportation professionals have trouble estimating trip lengths, it is surprising that responses were as accurate as they were in the past. However, these methodology changes could make it difficult to compare vehicle miles traveled (VMT) trends over time.
Looking to the future, the NHTS program is discussing the use of GPS probe, cellular, and/or Location-Based Services data to curate a national origin-destination (O-D) database that could be combined with survey findings from a smaller annual service. This O-D database will likely do a better job capturing long-distance travel than the current survey instrument but will struggle to reveal smaller trips–especially the quick walk to the ATM and similar types of travel. The plans to include big data in the NHTS program seem relatively far along, although the NextGen survey instrument and methods for combining these data sources seems less certain.
Annual collection of surveys will be very helpful in terms of the timeliness of results, since transportation changes are occurring so fast that there are many reasons to consider an eight-year wait to be too long. However, the smaller sample size will reduce the opportunity to do sub-national analysis or examine as many segments of population. The federal government has hopefully learned a lot at this point through the U.S. Census Bureau’s American Community Survey about how to collect annual surveys in a way that allows broader insights.
Overall, I’m excited to be able to apply this new data resource to future work and hope it can provide additional answers about how and why people are traveling.