A report on participant sampling and recruitment for travel and physical activity data collection : final technical report, July 2009.

Moudon, Anne Vernez; Saelens, Brian; Rutherford, Scott; Hallenbeck, Mark · 2009 · ROSA P / Transportation Northwest (TransNow) UTC

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This technical report details the participant sampling and recruitment strategies for a longitudinal study investigating the impact of light rail transit (LRT) on travel behavior and physical activity. The research addresses a gap in transportation literature, where previous cross-sectional designs have limited the ability to draw causal inferences regarding how infrastructure investments affect mode choice and health outcomes. Motivated by the introduction of LRT in Seattle, the study aims to determine if residents living near LRT stations increase their transportation-related walking and total physical activity compared to those living farther away. The project is part of a larger 5-year, $3.3 million study funded by the National Institutes of Health, with this specific report covering baseline data collection efforts supported by bridge funding from Transportation Northwest (TransNow). The study employs a case-control prospective cohort design involving 1,000 adults in King County, Washington. Participants are divided into two groups: "cases" residing within one mile of an LRT station and "controls" residing more than one mile away. To ensure valid comparisons, cases and controls are frequency-matched on seven criteria: household income, race, home value, net residential density, housing type, proximity to neighborhood commercial services, and levels of bus ridership. Sampling was conducted using parcel-level data to delineate geographic areas, followed by random selection of residential addresses. Recruitment involved mail and phone contact, with strict inclusion criteria such as residency duration and ability to walk. Data collection methods include portable GPS loggers and seven-day travel diaries to assess travel behavior, and accelerometers to objectively measure physical activity. Surveys also capture demographic data and perceptions of the built environment. The report presents the results of the baseline sampling and matching process. The case population, drawn from 100 census block groups within one mile of seven LRT stations, comprises 46,000 households characterized by high ethnic diversity and lower median income compared to the county average. The control population was selected from 673 census block groups outside the one-mile buffer, stratified to match the case population’s socioeconomic and environmental profiles. The matching process successfully aligned the two groups across all seven criteria, including housing type, assessed property value, and residential density. However, the report notes challenges in recruitment logistics, specifically low match rates (less than 50%) when linking residential addresses to telephone numbers via vendor databases, with controls achieving higher match rates than cases. The significance of this work lies in its rigorous methodological approach to isolating the effects of LRT on human behavior. By using a longitudinal natural experiment design with carefully matched controls, the study offers a potent test of how transportation investments influence physical activity and mode choice. The successful establishment of a representative baseline cohort allows for future causal analysis of behavioral changes before and after LRT implementation. This research contributes to scientific understanding of the health benefits of transit-oriented development and provides evidence to inform transportation policy regarding the public health impacts of infrastructure investments.

Key finding

The study design successfully delineated and matched case and control populations using parcel-level geographic data and stratified sampling to ensure comparability in socioeconomic and environmental characteristics.

Methodology

dataset

Sample size: 1000

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• exposure measurement