Objective Driving Data in the LongROAD Study: An Overview of Changes to Data Collection Procedures from a Datalogger to a Travel App
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 research brief details the transition of the Longitudinal Research on Aging Drivers (LongROAD) study from using vehicle-mounted Danlaw dataloggers to a smartphone-based travel application for collecting objective driving data. The study aims to understand the driving behaviors, exposure, and safety of older adults (ages 65–79), a demographic at high risk for crash-related deaths. While self-reported surveys are common, they suffer from recall bias; thus, objective measures are critical. Initially, the study used Danlaw dataloggers plugged into vehicles’ OBD ports. However, as smartphone technology improved, the study shifted to a travel app developed by Tourmaline Labs to allow for portable data collection across any vehicle the participant drives, rather than just their primary vehicle. The paper outlines the technical and procedural differences between the two methods. The Danlaw datalogger collected data at a 4 Hz sampling rate and defined trips based on engine start/stop cycles. In contrast, the LongROAD travel app utilizes smartphone GPS and accelerometers at a 100 Hz sampling rate, capturing more granular data, including rapid deceleration events that the datalogger might miss. The app defines trips based on vehicle stop duration, considering stops under five minutes as part of a single trip. To address the challenge of identifying whether the participant was driving or riding as a passenger, the study developed a deep-learning driver detection algorithm. This algorithm creates a unique "driving signature" for each participant based on turning characteristics, acceleration, and angular momentum, validated against historical datalogger data. A pilot study involving 32 participants was conducted in January 2019 to assess willingness to transition, data quality, and phone-carrying habits. Results indicated that participants carried their phones while driving more than 90% of the time, a prerequisite for data collection. Following the pilot, the app was gradually rolled out across five study sites between April and August 2019. As of June 2021, 1,273 of the 2,990 enrolled participants had successfully transitioned to the app. Some participants could not transition due to lack of smartphone access, inadequate phone plans, privacy concerns, or refusal to consent. The brief concludes that while the transition modernized data collection, it introduced challenges. Data from the two methods are not directly comparable due to differences in granularity, trip definitions, and sampling rates. Additionally, the transition caused gaps in objective driving data for participants who did not adopt the app. The authors emphasize the importance of clear communication among study sites, developers, and participants, as well as robust troubleshooting procedures. The shift to smartphone apps represents a more flexible approach to collecting naturalistic driving data, though it requires addressing issues related to device accessibility and privacy to ensure comprehensive data coverage.
Key finding
The transition from dataloggers to a smartphone app enabled the collection of more granular driving data but introduced challenges in data comparability and required a new driver detection algorithm to ensure accurate participant identification.
Methodology
mixed_methods
Sample size: 2990
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_aaa_foundation on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | aaa_foundation | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
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.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: observational prevalence
- Methodological Resource: dataset resource, tool software