Critical Areas in Advanced Driver Assistance Systems Safety: Point of Sale and Crash Reporting (06-003)

Project Description: This dataset was developed for an exploratory analysis of ADAS features and safety outcomes. The dataset is built on NHTSA’s CISS database. Related tables from CISS was retrieved and imported.Variables describing crash injury severity, vehicle style, and advanced driver assistance features were derived from across the related CISS tables and stored at the crash level or vehicle level. This dataset includes reproductions of the crash and vehicle tables from 2017, 2018, 2019, and 2020 CISS, harmonized for consistency across years. Columns have been added to both the crash and vehicle table, representing the result of further calculations across multiple tables from the CISS dataset. This documentation describes these additional columns. For descriptions of original CISS variables, refer to CISS documentation. From these data, we observed improved safety outcomes associated with the presence of three ADAS features – lane departure warning, forward collision warning, and blind spot detection. In particular, vehicles with ADAS features were less likely to be involved in fatal and severe crashes than vehicles of a similar age and body style. Additionally, vehicles with ADAS features had a lower change in velocity pre- and post-crash, a key proxy for injury severity. Vehicles with ADAS were also less likely to be involved in head-on and rear-end crashes than similar vehicles without ADAS and were less likely to experience a severe outcome than vehicles without ADAS when involved in sideswipe collisions. Despite these promising findings, however, small sample sizes precluded robust within-category and multivariate analysis. More robust data is urgently needed to disentangle the safety effects of ADAS features from those related to confounding variables such as vehicle age and body type for vehicle occupants, pedestrians, and bicyclists. We conclude the paper with recommendations for improvements to the CISS database that could support future investigation. We hope this work provides a baseline for scholars and practitioners to deepen our understanding of the effects of ADAS features on transportation safety. Data Scope: Table 1: Crashes Number of observations: 11,119 Number of columns: 93 Description: This table includes 1 record/row per crash in the dataset. Notes/Processing: The 'caseid' field is a unique identifier. Columns 2 through 26 are directly inherited from the data source, with minor harmonization done to pool across 2017--2020 data years. Refer to CISS documentation for these variables. Columns 27 through 93 are calculated, derived, or recoded from this and other tables in the CISS database and are documented in the accompanying data dictionary. All binary, ordinal, and categorical variables are encoded with descriptive values (e.g., yes/no instead of 1/0) so no separate codebook is needed. Table 2: Vehicles Number of observations: 19,983 Number of columns: 190 Description: This table includes 1 record/row per vehicle in the dataset. Notes/Processing: Vehicles join to crashes in the crash table using the 'caseid' field. The combination of 'caseid' and 'vehno' uniquely identify each vehicle in the dataset. Columns 2 through 106 are directly inherited from the data source, with minor harmonization done to pool across 2017--2020 data years. Refer to CISS documentation for these variables. Columns 107 through 190 are calculated, derived, or recoded from this and other tables in the CISS database and are documented in the accompanying data dictionary. All binary, ordinal, and categorical variables are encoded with descriptive values (e.g., yes/no instead of 1/0) so no separate codebook is needed. Data Specification: Please see the Data Dictionary document below.