A two-dimensional propensity score matching method for longitudinal quasi-experimental studies: A focus on travel behavior and the built environment

Abstract

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.

Mathematical Modeling of COVID-19 Cases and Deaths and the Impact of Vaccinations during Three Years of the Pandemic in Peru

“…To determine the causality effect between two variables, some statistical strategies have been developed; for example, propensity score matching (PSM) is widely used in the epidemiological context when it is desired to minimize the bias of non-randomized studies and to determine the effects of control measures or treatments on populations of interest; however, this methodology is essentially applicable only to cross-sectional studies and requires cofounders or covariables to perform a robust study [44][45][46][47][48]. In addition, some methodologies have been implemented in time-varying treatment or exposure, but this methodology could be inappropriate [49,50].…”

Examining neighborhood effects on residents’ daily activities in central Shanghai, China: Integrating “big data” and “thick data”

“…Empirical studies on neighborhood effects on daily activities and travel behavior can be decomposed into the effects of physical environments, sense of community, social cohesion and control, relative deprivation, social network, cultural norms and values. Empirical evidence indicates that people’s physical activity and travel behavior are associated with several characteristics of the physical neighborhood such as residential density, street connectivity, land use patterns, local accessibility, and pedestrian-friendly features (Aditjandra et al, 2012; Zhong et al, 2020). While the physical planning dimension has been spotlighted in the literature, other dimensions have received inadequate academic attention.…”

Building Sustainable and Connected Communities by Addressing Public Transportation’s First-Mile Problem: Insights from a Stated Preference Survey in El Paso, Texas

“…To address this access barrier, an increasing number of public transit agencies are developing innovative strategies to promote the first-mile connection. These strategies focus on introducing new mobility services, such as autonomous vehicles, ride-sharing services, electric scooters, and dockless bicycles to connect homes and bus stops, and the evaluation of these strategies likely needs to involve emerging analytical methods [37][38][39][40][41][42][43][44].…”