First Advisor

Jennifer Dill

Term of Graduation

Spring 2024

Date of Publication

5-30-2024

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.) in Urban Studies

Department

Urban Studies and Planning

Language

English

Physical Description

1 online resource (xi, 192 pages)

Abstract

Despite advanced policies, plans, and facilities, many pedestrians are still injured and killed in traffic crashes in the United States. To improve pedestrian safety and the walking environment, the relationship between surrounding crash risk factors and perceived safety that influence people's behavior needs to be studied. This study aims to examine pedestrian crash risk factors, the relationship between crash risk factors and perceived safety, measured as threatened experiences and safety attitudes, and the relationship between safety attitudes and walking behavior.

The analysis used data from three primary sources: (1) an original survey of 551 residents in 10 neighborhoods in Oregon conducted in 2023; (2) pedestrian crash data that occurred in Oregon for 2018 - 2022; and (3) pedestrian count data collected at 65 sites in 2022. 729 pedestrian crashes occurred in the census block groups surrounding the 65 sites over five-year periods. These were complemented with built environment data.

The dissertation first tested whether crash risk factors predict actual pedestrian crashes in the study areas. One of the results shows that the pedestrian volume measured as pedestrian count data has better predictive power to explain the pedestrian crashes, cumulated for a shorter period of time than the pedestrian volume measured as population density. Though the count data was collected only for two days, it was more accurate than the population density. This result supports the need to collect pedestrian volume data in various places to develop road safety plans and policies. In addition to pedestrian volume, crash risk factors in macro-level areas, including mixed-use land areas, commercial land areas, and public transit stops, are significant in predicting pedestrian crashes. However, in this study, the number of intersections, speed limit, and vehicle speed were not statistically significant in predicting crash cases.

The structural equation model (SEM) results for the second research question show that the threatened experiences influenced by the surrounding environment significantly affect safety attitudes. Pedestrians feel more threatened in areas with higher intersection density and mixed-use land areas after controlling other risk factors, including speed and pedestrian and traffic volumes. However, intersection density is not significantly related to the cumulated pedestrian crashes. This may be because vehicle speeds decrease as the density of intersections increases. This implies that when pedestrians encounter intersections more frequently, they perceive that they have had more threatened experiences, even though the environment is not significantly riskier. Pedestrian crashes did not affect pedestrians' threatened experiences and safety attitudes in the SEMs. This can be interpreted as pedestrians' attitudes being mainly determined by their perceived experiences in a given environment rather than an actual crash risk.

In modeling results for the third research question, positive safety attitudes and nearby sidewalks increase walking frequency. On the other hand, larger commercial areas, faster vehicle speeds, and more vehicles in their households significantly reduce walking frequency. One likely reason for the negative relationships with having commercial areas nearby is that most survey respondents were walking primarily for exercise, to walk their pets, or for entertainment rather than to visit specific destinations such as work, school, or restaurants.

This study has several limitations despite meaningful findings. One limitation relates to the unit of analysis. For this analysis, crash risk factors, such as intersection types, road classification, weather conditions, or street lights, were aggregated around each respondent's home. In addition, pedestrian count data was available only for one or two major intersections within that area. Such aggregation does not account for details of the risk of each crash event or in micro-level places. The other limitation relates to cross-sectional analysis. If panel data can be collected and time-series analysis is possible, it would likely investigate how people's attitudes toward safety and behaviors change due to surrounding crash risk factors and threatened experiences.

Rights

© 2024 Kyu Ri Kim

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

Persistent Identifier

https://archives.pdx.edu/ds/psu/42081

Download

Share

COinS