School of Civil and Environmental Engineering

Ph.D. Thesis Defense Announcement

User Preferences for Bicycle Infrastructure and the Impact of Facilities on Perceptions of Bikability

By

Calvin Clark

Advisor:

Dr. Kari Watkins (CEE)

Committee Members:

Dr. Pat Mokhtarian (CEE), Dr. Giovanni Circella (CEE), Dr. Shatakshee Dhongde (Econ)

Date & Time: Monday, November 11th, 2019, at 3:00 PM

Location: Sustainable Education Building (SEB), Room 122

     Bicycling for transportation is experiencing a resurgence in much of the U.S. Consequently, the question of facility design has become a hot research topic. However, most such research is conducted in areas with strong bicycling cultures, which misses a critical link of how facility design can help shape bicycling culture. This dissertation contains analyses on data from a dual-wave survey deployed between 2016 and 2018 in communities in the Southern United States. To varying degrees, these are communities where cycling is not (yet) popular and/or widely adopted, a setting that is much more representative of the nation at large. 
     The first analysis is on a subset of the data from the first-wave survey (N=1,178), in which quantitative analysis including linear regression models are used to estimate perceived comfort and safety of and willingness to try bicycling facilities. Facilities that limited interaction with automobile through, turning, and parking traffic were found to be perceived as more desirable among cyclists. The second analysis contains the full first-wave sample (N=2,157) including respondents in neighborhoods in Atlanta, GA. Latent class models were estimated with attitudinal factors such as bicycle enjoyment and risk tolerance as class membership covariates, with results indicating the presence of a latent class of pro-bicycling but risk-cautious respondents whose perceptions differ from those of their pro-bicycling, risk-embracing counterparts by the relatively greater impact of protected bicycle facilities. The final analysis of the dissertation is conducted on before-and-after survey responses (N=807) from all ten sites, with five sites serving as treatment communities (where bicycling facility treatments were implemented over the course of the study) and five serving as control communities (where no such bicycling facility treatments were implemented). Linear regression and order logit models are estimated respondents’ perceptions of changes in bikability in their neighborhood. Those in the treatment group were more likely to rate observed changes as improvements, though the effect of on-street facilities diminished for those farther from the treatment. These results provide key takeaways regarding expectations for the impacts of bicycling facilities in places where bicycling for transportation is viewed as rare, which can provide an additional tool for planners and engineers in making the case for the implementation of bicycle facilities.