Why kids stopped walking and biking to school

In 1969, 47.7% of American K-8 students usually walked or biked to school. By 2009, 12.7% did.

Those two numbers come from the same survey question, asked the same way, forty years apart, and they are the closest thing this subject has to a fixed point. Call it 48% and 13% in conversation. Both are the share of kids aged 5 to 14 who usually got to school under their own power.

One thing to get out of the way first, because it matters on a site about bike buses. That 48% was almost entirely walking. Biking has never been more than about 1% of American school trips in any year anybody measured. The 1969 survey did not even separate the two, which tells you how small the bike share was considered at the time. So this is not a story about a country that used to bike to school and stopped. It is a story about a country where getting to school on your own feet used to be normal.

I expected the explanation to be distance and traffic. That is about half right, and the half that is wrong is the interesting part.

The trip got longer

Start with the thing that is unambiguously true: school is farther away than it used to be.

In 2009 the average American student lived 4.4 miles from school, and the average elementary student 3.6 miles. Fewer than a third of K-8 students, 30.6%, lived within a mile of theirs.

Distance decides almost everything about how a kid gets to school. In the 2009 data, 55.3% of K-8 students living within a quarter mile usually walked. At half a mile to a mile it was 15.1%. At one to two miles, 4.0%. Living a quarter mile from school makes a child roughly fourteen times more likely to walk than living one to two miles away.

Noreen McDonald, the transportation researcher whose work anchors most of this piece, put a number on it. Comparing 1969 to 2001 and holding the changing spread of students constant, she found that changes in how far kids live from school may account for about 47% of the decline. Her hedge, not mine.

It is worth being precise about why the trip got longer, because the usual explanation is wrong. People say schools consolidated. They did, but mostly before this story starts. The United States went from 248,117 public schools in 1929 to 89,372 by 1970, and the average school roughly quintupled in size in that window. Since 1970 the number of public schools has gone up, not down, and average size has been roughly flat.

What changed after 1969 was not how many schools there were. It was where they got built.

From 1964 onward, the widely used national guidance recommended a minimum of 10 acres for an elementary school, 20 for a middle school and 30 for a high school, plus an acre for every 100 students. By the late 1950s every state except Wyoming had adopted some form of minimum size guidance. Those numbers were advisory, but school boards often read them as requirements, and a district looking for 30 contiguous acres does not find them in an existing neighborhood. It finds them at the edge of town, across an arterial, on land that was cheap because nobody lived there yet.

The same guidance, for decades, recommended that elementary students walk no more than three quarters of a mile. Both rules were in the same documents. One of them won.

That still only explains about half

Here is the sentence in McDonald's 2007 paper that reframed this piece for me:

Factors besides distance account for half of the decline in walking during the study period and most of the decline since 1977.

Most of the decline since 1977 is not about distance at all.

The clearest way to see it is to hold distance still and look only at the kids who live close enough to walk. In 1969, roughly nine in ten K-8 students who lived within a mile of school walked or biked there. By 2009 it was about a third.

Read those two sentences together. The share of kids living within a mile fell by about ten points. The share of those kids who walked fell by more than fifty. The trip changed, and then, at every distance, what families do with that trip changed more.

So what happened on the short trip?

Ask a parent why they drive

There is a study I keep coming back to, because it asks a better question than the usual survey does.

Most research asks parents what the barriers to walking are, and gets back the answers you would expect: it is too far, the traffic is bad. McDonald and Aalborg instead called 403 families in walkable East Bay neighborhoods and asked the parents who drive their kids under two miles why they, personally, drive.

About three quarters said convenience. Time savings, dropping off on the way to work, everybody sleeps later. Just over 30% mentioned safety of any kind.

And about half of the parents driving under two miles did not allow their child to walk to school without an adult along.

That is a different problem than the one most programs are built for. It is a morning-logistics problem and a supervision problem. A parent who is already in the car at 7:40, already going that direction, and would not send an eight-year-old alone regardless of how good the sidewalk is, does not need a sidewalk. They need the eight-year-old to have somewhere to go with an adult already attached.

Worth saying plainly: this is one study, 403 families, in unusually walkable neighborhoods, and the kids were 10 to 14. It is not the whole country. It is the sharpest available answer to the question of why short trips get driven.

What about traffic?

This is where the research surprised me most, and where I want to be careful, because none of it means a parent's caution is silly.

American streets did not get more dangerous for children in the way the story usually goes. Pedestrian deaths among children under 13 fell from 1,632 in 1975 to 138 in 2024. Child bicyclist deaths fell from 446 to 37. Both are down about 92%.

That decline held even through the recent surge in adult pedestrian deaths. Total US pedestrian deaths rose about 80% between 2009 and 2023, while all other traffic deaths rose 13%. Over that same stretch, child pedestrian deaths kept falling, down about 27% between 2015 and 2024.

Now the honest caveat, and it is a real one. Nobody in the United States has computed the child pedestrian fatality rate per mile walked. Every number above is deaths per population, not deaths per trip. Since children walk far less than they used to, the counts falling is consistent with two very different worlds: streets became much safer for kids on foot, or kids simply stopped being on foot. The published literature says outright that the exposure data to settle this does not exist.

The one country where somebody did the arithmetic is England and Wales, where between 1985 and 2003 the child pedestrian death rate per ten million miles walked fell about 75% while the distance children walked fell only 19%. That is a genuine safety improvement, not an artifact of kids staying home. It is also a different country, ending twenty years ago, and I would not transfer it to Atlanta.

So I cannot tell you American streets got safer per mile walked. I can tell you the raw counts fell hard, and that the picture of a steadily more dangerous walk to school is not what the data shows.

None of which makes a parent's caution irrational. A parent is not managing a national fatality rate. They are managing one specific crossing, one specific arterial, one specific kid who is not paying attention. And in the studies, traffic danger is consistently one of the two most-cited barriers to letting a child walk. The feeling is real, it is widely shared, and a solution that ignores it is not a solution.

What the trip is actually worth

Federal guidelines say children and adolescents aged 6 to 17 should get 60 minutes or more of moderate to vigorous physical activity a day.

Kids who actively commute to school do get more activity than kids who don't. Accelerometer studies put the gap somewhere between about 3 and 40 minutes of extra activity a day, which is a wide enough range to be honest about.

But the research gets less flattering the better it is designed. In one large study, active travel was associated with about 7 extra minutes a day when you compared kids at a single point in time, and that association disappeared when the same kids were followed for two and a half years. Most studies find no difference in how active these kids are during the rest of the school day, which suggests the gain is the commute itself rather than some general activation of the child.

So the modest, defensible version: the ride is exercise, on the days it happens, and it is exercise that does not require anybody to schedule it. That is worth something. It is not a weight-loss program, and studies that looked for effects on body weight mostly did not find them.

Everything past that point, about confidence, about knowing your own neighborhood, about the specific pleasure of arriving somewhere under your own power with your friends, I believe from watching it. I am not going to dress it up as a finding.

One answer is riding together

A bike bus is a group of kids and adults who ride to school together along a planned route, on a set morning, picking up families at stops along the way. Adults lead and sweep. That is the whole idea.

Look at what it actually does against the list above. It supplies the adult supervision that about half of short-distance driving parents say they require. It removes the parent's time cost, because one adult covers many children instead of each family running its own escort. And it makes a group visible enough that the crossing everybody worries about gets crossed as a group, which is the part parents tell researchers they care about.

That is not me reverse-engineering a pitch. McDonald and Aalborg reached it themselves, proposing that a program providing adult supervision and removing parents' time costs could address the concerns of about 60% of the parents in their sample who drive under two miles. That figure is their projection from a survey, not a measured result, and they said themselves that such a shift is unlikely to happen in full.

There is also a nice fit with distance that took me by surprise. Biking's share of school trips peaks at 3.4% in the half-mile to one-mile band, which is exactly where walking starts to fail on time. The bike is the tool for the distances that walking lost.

Now the part where I tell you what the evidence does not say.

Bike buses have been studied once, properly. A 2017 pilot randomized trial in four Seattle schools put 54 fourth and fifth graders into bicycle trains for three to five weeks. Cycling to school rose 44.9 percentage points. Measured activity rose 21.6 minutes a day. Usefully, both the intervention and control groups got bikes, helmets and a safety course, so what the trial isolated was the group ride itself.

It has never been replicated. A systematic review of school cycling programs found exactly two papers on bicycle trains, and they were the same intervention.

The closest well-studied cousin is the walking school bus, and its arc is a caution. An early pilot produced large effects. When it was finally tested across 22 schools with 418 children over full school years, active commuting rose about 9 percentage points, roughly a third of what the pilot suggested, activity rose about 4 minutes a day, and the effect on body weight was zero.

Read any bike bus success story, including ours, with that in mind.

What a bike bus cannot do

It cannot shorten a 4.4-mile commute. If school is across a highway on the other side of a county, this is not your fix.

It cannot build a sidewalk, tame an arterial, or install a protected lane. Only 4% of kids living one to two miles from school walk, and closing that gap is infrastructure work with a budget and a timeline, not a group ride.

There is no research at all on whether bike buses reduce injuries. Not weak research: none. The single trial reported no adverse events among 54 children over about a month, which is nowhere near enough exposure to detect anything. Anybody who tells you a bike bus is statistically safer than riding alone is telling you something nobody has measured.

And it will not work for every family or every school. A group ride is a fixed time on a fixed morning, which is exactly the rigidity some families cannot absorb.

We don't need to go back to 1969

I want to be clear that this is not an argument for restoring some earlier America. The country that produced 48% was one with smaller schools closer to houses, fewer two-earner households, and a different tolerance for eight-year-olds crossing streets alone. Some of what changed is genuinely better.

But there is one more number worth knowing, and it is the hopeful one. The decline stopped.

The federal survey has a second question, asking how a child traveled on one specific day rather than how they usually travel, and it covers ages 5 to 17. It is a different measure than the 47.7% and 12.7% above, so do not line them up side by side. On that measure, walking and biking to school was 13.7% in 2001, 10.6% in 2009 and 10.4% in 2017, differences small enough that the agency itself calls them statistically indistinguishable. The forty-year slide was a 1969 to 2001 event. For twenty years the number has been flat.

A thing that fell for thirty years and then stopped is not a law of nature. It is the sum of decisions about where schools go, how streets are built, and what a morning looks like. Decisions can be made differently.

The serious version of making them differently is a portfolio: safe routes programs, sidewalks, crossings, traffic calming, protected lanes, school siting that stops treating 30 acres at the edge of town as the default. Federal Safe Routes to School funding, evaluated across 801 schools, was associated with walking and biking rising about a percentage point per year at participating schools, with engineering improvements worth about 3 points on their own. Associated, not caused. That is the honest verb for the study design.

A bike bus is one tool in that portfolio, and a cheap one. It is the tool a few parents can pick up next Friday without waiting for a capital budget. It is not the whole answer, and the people who will tell you that most clearly are the ones already organizing them.

Give your bike bus a home

If your school already has a bike bus, an organizer can add it here for free, and families searching for your school will find the route, the stops, the ride times, a link to share and a QR code for the flyer.

If your school doesn't have one, somebody has to be first. Find your school, or read how to start one before you decide.

If you want the shorter version of what a bike bus is before any of that, the shorter explainer. If you want to know what your local Safe Routes program can do alongside it, we wrote that up too.

Sources

• McDonald NC, Brown AL, Marchetti LM, Pedroso MS. "U.S. School Travel, 2009: An Assessment of Trends." American Journal of Preventive Medicine 2011;41(2):146-151. PubMed 21767721 · doi:10.1016/j.amepre.2011.04.006
• McDonald NC. "Active Transportation to School: Trends Among U.S. Schoolchildren, 1969-2001." American Journal of Preventive Medicine 2007;32(6):509-516. PubMed 17533067 · doi:10.1016/j.amepre.2007.02.022
• McDonald NC, Aalborg AE. "Why Parents Drive Children to School: Implications for Safe Routes to School Programs." Journal of the American Planning Association 2009;75(3):331-342. doi:10.1080/01944360902988794
• McDonald NC. "School Siting: Contested Visions of the Community School." Journal of the American Planning Association 2010;76(2):184-198. doi:10.1080/01944361003595991
• McDonald NC, Steiner RL, Lee C, Rhoulac Smith T, Zhu X, Yang Y. "Impact of the Safe Routes to School Program on Walking and Bicycling." Journal of the American Planning Association 2014;80(2):153-167. doi:10.1080/01944363.2014.956654
• Mendoza JA, Haaland W, Jacobs M, et al. "Bicycle Trains, Cycling, and Physical Activity: A Pilot Cluster RCT." American Journal of Preventive Medicine 2017;53(4):481-489. PubMed 28668251 · doi:10.1016/j.amepre.2017.05.001
• Johnson AM, Zhou M, Haviland MJ, Mendoza JA. "Evaluation of a walking school bus program: a cluster randomized controlled trial." International Journal of Behavioral Nutrition and Physical Activity 2024;21(1):55. PubMed 38730407 · doi:10.1186/s12966-024-01602-w
• Schönbach DMI, Altenburg TM, Marques A, Chinapaw MJM, Demetriou Y. "Strategies and effects of school-based interventions to promote active school transportation by bicycle among children and adolescents: a systematic review." International Journal of Behavioral Nutrition and Physical Activity 2020;17(1):138. PubMed 33183331 · doi:10.1186/s12966-020-01035-1
• Sonkin B, Edwards P, Roberts I, Green J. "Walking, cycling and transport safety: an analysis of child road deaths." Journal of the Royal Society of Medicine 2006;99(8):402-405. PubMed 16893940 · doi:10.1177/014107680609900817
• National Center for Safe Routes to School. "How Children Get to School: School Travel Patterns from 1969 to 2009." 2011. pedbikeinfo.org
• Federal Highway Administration. "Travel to School: The Distance Factor" and 2017 National Household Travel Survey school travel data. nhts.ornl.gov
• Insurance Institute for Highway Safety. "Fatality Facts: Children," from NHTSA Fatality Analysis Reporting System data. iihs.org
• U.S. Department of Health and Human Services. "Physical Activity Guidelines for Americans, 2nd edition." 2018. health.gov
• National Center for Education Statistics, Digest of Education Statistics, Table 214.10 (public schools) and Table 201.10 (enrollment). nces.ed.gov