Chief statistician’s update: understanding road collision and casualty statistics

Darllenwch y dudalen hon yn Gymraeg

On 6 June 2024 we will be publishing provisional data on police recorded road collisions and casualties for October to December 2023. Since the introduction of the 20mph default speed limit on restricted roads in September 2023 there has been considerable interest in collisions data.

This blog post aims to help users to better understand the data we publish, including:

what will be published and where
strengths and limitations
guidance on how the data can be interpreted
how this data will feed into 20mph monitoring and evaluation plans

Statistics to be published

A number of different outputs will be published and updated, including:

A statistical bulletin covering provisional data on collisions during 2023. This bulletin will present long term trends, accompanied by commentary and charts showing different breakdowns of the data (such as severity, location, speed limit of the road).
An interactive dashboard. This allows users to explore the data by different time periods, severity and road types. The dashboard also contains information on casualties and the vehicles involved.
Data tables on our StatsWales open data service. These tables present quarterly totals of collisions by different breakdowns, such as totals by the speed limit of the road where the collision occurred, severity and road user type.
Collision level data. These spreadsheets will contain information on individual collisions, including location, time of collision and the number of vehicles and casualties involved allowing users to carry out their own more detailed analysis.

We routinely publish data quarterly on our StatsWales website with around a six month time lag. This year we will introduce some additional commentary and context to our bulletin to help improve interpretation and understanding. We will also update our quality report to summarise sources and methods, and review the quality of data.

Strengths and limitations

The data we collect from the four police forces in Wales relate to collisions which resulted in personal injury and for which information was reported by the police. These statistics are referred to as STATS19.

STATS19 is a comprehensive collection of data on police recorded road collisions and is collected across the UK. The data provided by each police force is consistent in terms of definitions and coverage and can be compared across the UK. Figures for fatalities are broadly comparable internationally.

The data that is collected as part of STATS19 is detailed, and we make as much of this data available to users as possible without risking breaching data protection rules.

Whilst the data is rich, the coverage is limited and does not include collisions:

that were not reported to the police
that occurred on private land i.e. car parks or fields
where no personal injury was recorded
where the collision was later confirmed by a medical professional or coroner to be a medical episode or suicide

Some elements of the data are limited by the knowledge of police officers at the time of the collision. For example, we present the contributory factors that were recorded by officers at attendance of the scene. We know that there are limitations to this data, and research carried out by the Department for Transport shows that:

over a third of fatal collisions had a speed-related factor assigned after further investigation, compared to a quarter of fatal collisions recorded through STATS19 (the process of collecting police recorded road collisions)
some factors, particularly those related to speed and impairment by drink or drugs appear more frequently when based on information available after further investigation, meaning that STATS19 data potentially understates the impact of these factors.

While presenting contributory factors gives us a good idea of actions that led to a collision, in the most serious of collisions it is possible that additional factors were at play that were not identified at the scene of the collision.

There are also some data quality issues relating to the recorded speed limit of the road where collisions occurred. Additional validation carried out this year highlighted some historical inconsistencies together with potential reasons for these inconsistencies, such as:

police officers recording the observed speed limit rather than the permanent speed limit
inconsistencies between local authorities and road speed limit signs
errors in recalling the road speed limit

Based only on a small sample, it appears that the road speed limit recorded at the time of the collision is more accurate for 20, 30 and 60mph roads than for 40, 50 and 70mph roads. In 2023, over 80% of collisions occurred on 20, 30 and 60mph roads.

Our road collisions and casualties quality report contains more detailed information on the coverage and limitation of these statistics.

Drawing conclusions from the statistics

The number of collisions over the short term tends to be volatile and can be impacted by things such as the weather. The COVID-19 pandemic saw significant changes in road collisions as a result of restrictions placed on how and where people could travel. Longer term trends give a better indication of patterns and changes in police recorded road collisions.

Currently, there are no official statistics available for the volume of traffic on roads with different speed limits. This is something that we are continuing to explore.

20mph monitoring and evaluation plans

On 17 September 2023, the law changed the default speed limit on restricted roads from 30mph to 20mph in Wales. These are usually residential or busy pedestrian streets with streetlights. The change affected most roads that were 30mph before 17 September, but not all. We have published a map on DataMapWales that shows which roads stayed at 30mph.

Transport for Wales (TfW) are responsible for monitoring the introduction of the 20mph default speed limit, and their latest release looks at preliminary weighted mean speed changes (TfW) following the national roll out. TfW have also published a summary of the monitoring data of the phase 1 roll out (TfW).

TfW’s monitoring framework for 20mph includes key performance indicators (KPIs). Three of these relate to the number of casualties that occur in police recorded road collisions. For a meaningful comparison to be made, typically at least three years’ worth of collision data would be required post-implementation to compare with at least a three year pre-implementation period.

Welsh Government is in the process of developing a specification for the evaluation of the 20mph policy which will be conducted by an independent evaluator. The evaluation will build on data available from the TfW monitoring framework, and will also explore the implementation of the policy and its wider impact.

Additional challenges

The way in which police forces collect data on road collisions is changing. Last year one of the police forces in Wales migrated to a new standardised reporting tool (called CRaSH – Collision Recording and SHaring) which is designed to provide a common way for police forces to collate and submit data.

CRaSH is an injury-based reporting system, and the Department for Transport have found that police forces using these types of systems are likely to see an increase in more collisions being classed as severe (Department for Transport) compared to the previous data collection.

This is further explained in our quality report, and we will continue to review how this impacts our data.

Future publications

We will continue to publish our quarterly data on police recorded road collisions. Transport for Wales have pre-announced their next publication for June 2024 which will present data for speed limit compliance, speed distribution of vehicles and 85th percentile speeds following the introduction of the national default 20mph speed limit on restricted roads for September 2023 to February 2024.

Contact us

We’d welcome any feedback to stats.transport@gov.wales

Stephanie Howarth
Chief Statistician