Neighborhood Characteristics that Support Bicycle Commuting

This study examined American Community Survey journey-to-work data from 2008 to 2012 to identify the characteristics of neighborhoods with the highest levels of bicycle commuting in the United States. The 100 census tracts with the highest bicycle commute mode shares (top 100 census tracts) were identified and paired with 100 other randomly selected census tracts from the same county (100 comparison census tracts). As a whole, the top 100 census tracts had a bicycle commute mode share of 21%. Seventy of the top 100 census tracts were in locations that had fewer than 10 days per year with high temperatures below 32°F (0°C), and 68 were within 2 mi (3.2 km) of a college or university campus. Seventeen had relatively low college populations and were in high-density neighborhoods close to large city central business districts. Conditional logistic regression was used to estimate the likelihood of a paired tract being in the top 100 rather than the comparison tract. After climate and topography were controlled for, being a top 100 census tract was associated with several socioeconomic and local-environment characteristics, including being located closer to a university and having more households without automobiles, more people born in other states and countries, higher population density, more housing constructed before 1940, and greater bicycle facility density. The results suggest that policies to model employment centers after university campuses; design neighborhoods that support routine, multimodal travel; and reduce barriers to bicycling in bad weather may help create more local areas with high rates of bicycle commuting.