SWITRS Overview

Introduction

The Statewide Integrated Traffic Records System (SWITRS) consists of data collected and maintained by the California Highway Patrol (CHP). SWITRS contains all crashes that were reported to CHP by local and governmental agencies. For more information regarding SWITRS or for detailed summaries by year, please visit the CHP website.

If you are interested in SWITRS crash rankings of cities or counties without investigating the actual datasets, the California Office of Traffic Safety (OTS) has an excellent resource at: OTS Crash Rankings. Their site allows you to choose a county or city and see crash rankings by total count, vehicle miles travelled, or by average population.

Final vs. provisional

We determine the dataset is final when we process the data after CHP announces the report of the specific year. CHP hasn't released the 2023 SWITRS Annual Report on CHP SWITRS website yet so we say provisional for 2023 and later datasets.

Working with the relational database structure

SWITRS has a hierarchical structure.

• The CRASH table contains information on each crash, one line per crash. 
• The PARTY table contains information from all parties involved in the crash, one line per party. Parties are the major players in a traffic crash - drivers, pedestrians, bicyclists, and parked vehicles. The information includes personal descriptors and vehicle descriptors.
• The VICTIM table contains information about the victims - persons associated with each party.  For example, a motorcyclist and his passenger are each a victim. Injury severity is included in the VICTIM table.

The victims can be thought of as being nested within parties and parties can be thought of as being nested within crashes. Another way to conceptualize the data is to use the entity-relationship model common in relational database management systems. In this data model, CRASH, PARTY, and VICTIM are entities that have very specific relationships to each other. CRASH is a “strong” or “parent” entity because it can exist on its own. PARTY is a “weak” or “child” entity, because it logically depends on an instance of CRASH. In other words, a driver cannot be a party in a crash unless the crash exists and is in the crash table. Likewise, VICTIM is a weak entity, because it depends on PARTY for its existence.

An important aspect of the relationship between the PARTY and VICTIM tables is that some parties (i.e., drivers, bicyclists, pedestrians) will also appear in the VICTIM table.  Injured parties will appear in the VICTIM table; uninjured parties will not. 

A common way to work with data of this structure is to create a PARTY table or a VICTIM table that has data from all its parents joined to each row.  For example, if you plan to do analyses at the victim-level, you can start with a VICTIM table and join in values from corresponding PARTY and CRASH tables.  In the figure, the CRASH variables and the PARTY variables are joined to the VICTIM table.  Doing so provides a victim-level table that can be summarized (for example, count or sums can be generated), while considering PARTY or CRASH variables.  You must keep in mind that the PARTY and CRASH variable will repeat for as many victims as there are.

Data definitions

The structure and allowed values for every SWITRS field are documented in the SWITRS Codebook. Use the codebook when validating records, building queries, or aligning TIMS exports with other datasets.

Geocoded coordinates

Roughly 97% of SWITRS crashes are geocoded using SafeTREC’s workflow. The POINT_X and POINT_Y fields store the resulting longitude and latitude in WGS84. When geocoding is not possible, those values remain blank.

The original GPS coordinates reported by officers are stored separately in the LATITUDE and LONGITUDE fields. Please note, mapping tools within TIMS will only display crashes that have a SafeTREC geocoded location.

Using SWITRS in GIS tools

SWITRS downloads include both POINT_X and POINT_Y fields geocoded by SafeTREC as well as LATITUDE and LONGITUDE fields which were provided by the original crash report. You can load downloaded SWITRS data into most GIS or mapping applications using these fields. Steps vary by product, but generally you’ll import the CSV file and then specify the coordinate fields. ArcGIS users can follow the Add XY data help article for detailed instructions.

Why recent years have fewer crashes

CHP data entry typically lags by 12–18 months. TIMS posts the most recent extract available from CHP each quarter, which may still be incomplete. Expect number of crashes in provisional years to increase over time until CHP finalizes.