A two-dimensional propensity score matching method for longitudinal quasi-experimental studies: A focus on travel behavior and the built environment
- Resource Type
- Authors
- Haotian Zhong; Marlon G. Boarnet; Wei Li
- Source
- Environment and Planning B: Urban Analytics and City Science. 48:2110-2122
- Subject
- 050210 logistics & transportation
Focus (computing)
business.industry
Computer science
05 social sciences
Geography, Planning and Development
Management, Monitoring, Policy and Law
Machine learning
computer.software_genre
0506 political science
Urban Studies
Travel behavior
Causal inference
0502 economics and business
Architecture
Propensity score matching
050602 political science & public administration
Artificial intelligence
business
computer
Built environment
Nature and Landscape Conservation
- Language
- ISSN
- 2399-8091
2399-8083
The lack of longitudinal studies of the relationship between the built environment and travel behavior has been widely discussed in the literature. This paper discusses how standard propensity score matching estimators can be extended to enable such studies by pairing observations across two dimensions: longitudinal and cross-sectional. Researchers mimic randomized controlled trials and match observations in both dimensions to find synthetic control groups that are similar to the treatment group and to match subjects across before- and after-treatment periods. We call this a two-dimensional propensity score matching method. This method demonstrates superior performance for improving treatment effect estimation based on Monte Carlo evidence. A near-term opportunity for such matching is identifying the treatment effect of transportation infrastructure on travel behavior.