Editor’s note: This article originally appeared on City Observatory. Read the original here.
For all the time we spend talking about transportation, it’s surprising how little useful data there is. The Census Bureau asks a limited number of questions about our daily commute as part of its annual American Community Survey, but this misses out most trips that aren’t to and from work.
The US Department of Transportation periodically commissions the National Household Travel Survey to ask a richer, more detailed set of questions about all types of trips. The NHTS serves as an important benchmark for understanding travel patterns.
The latest survey, completed in 2022, will likely help answer questions about how the COVID-19 pandemic has changed our travel patterns. Last week, the U.S. Department of Transportation tweeted an infographic showing key findings from the latest NHTS.
Warning: Data from 2017 and 2022 cannot be compared.
The graph seems to show a dramatic, even devastating, decline in travel. The headline reads: “Daily Trips Decline.” The caption reads: “Family travel declined in 2022 compared to 2017.” The table shows that trips have fallen from 3.4 trips per person per day to just 2.
That’s a big change if true. The “fewer trips” message is clearly a take on the data from Active Transportation blogger Angie Schmidt, who echoes the top number on the graph and mused on our increasing isolation in a Substack essay titled “Nobody Leaves the House Anymore.”
In 2017, Americans made an average of three trips a day. They went to work, came home, and then went somewhere else. That’s three trips. They might have also gone to the store or a friend’s house every other day. Now, people make a little more than two trips a day…
Like I said, I think this is amazing. Let me explain why a little bit more. This is going to make a huge difference in people’s lives and daily activities. I’ve been working on sustainable transportation for over a decade, and I never could have imagined that the changes we’re making would have such an impact so quickly and on such a scale.
But Angie is right to describe such a shift as inconceivable, and on closer inspection, the decline in travel is likely an illusion, driven by faulty data (or, more accurately, faulty data inference).
There are problems with the data. Between 2017 and 2022, the U.S. Department of Transportation made major changes to the data collection methods for the NHTS. First, it moved to a primarily online data collection method; previous surveys relied on face-to-face interviews. Second, and more importantly, the survey did away with “travel diaries.” Previous versions of the NHTS sent participants travel diaries to contemporaneously record their trips. The 2022 online survey does not include a diary and relies on participants’ memory.
This change almost certainly accounts for much (perhaps all) of the decline in reported trips, according to a technical note from the study (posted on the website of the Oak Ridge National Laboratory, which conducted the study for the US Department of Transportation), which reviewed earlier findings and concluded that relying solely on memory (rather than contemporaneous diaries) misses about 20 percent of all trips.
error
Certainly, travel may be different (and less) now than it was pre-pandemic, but because methodologies vary so much from year to year, this data cannot be used to make reliable statements about the direction or magnitude of change.
In Raiders of the Lost Ark, just before Indiana Jones swallows a poisoned date, Sarah grabs it out of the air and yells, “Bad date!” Unfortunately, she wasn’t able to grab the bad data before it was swallowed.
It is important for users of this NHTS data to know that travel data from 2017 cannot be compared to travel data from 2022. This should be a warning on a cigarette pack, not an obscure footnote. Yet USDOT did the exact opposite with a cute infographic. In its tweet, it endorses the two data sets as directly comparable. Thanks to this misinformation, we are faced with a “BS asymmetric” problem. It takes very little effort to use social media to create erroneous (but interesting) findings, but it takes much more effort to correct the error. We hope that the US DOT will take steps to correct this false impression.
If it were true that overall travel was down by about one-third, this would have significant implications for the transportation system – it would call into question nearly all spending on expanding highway capacity, for example. It is unfortunate that the NHTS data is presented in such a sloppy and inaccurate manner by an agency called the US Department of Transportation.
That doesn’t mean the National Household Travel Survey is useless. The key point here is that you can’t directly compare 2022 data to previous data to make plausible claims about changes in travel patterns. In fact, that’s one of the big questions we want to know: There seems to be no doubt that the pandemic and the advent of working from home have caused fundamental changes in travel patterns.
