Randomised control trial testing the impact of a pilot training package for police officers on stop and search.
Dr Paul Quinton
|Collaboration and partnership||
The Equality and Human Rights Commission, Research Advisory Service and RAND Europe.
|Level of research||
|Project start date||
- Trained officers would have greater knowledge and feel more prepared for stop and search.
- Trained officers would have attitudes more favourable to good practice in police-public interactions.
- Trained officers would say they will behave in line with training standards on stop and search decision making and practice.
- Trained officers’ practice would be more professional and effective.
Six pilot forces were involved:
- British Transport Police
- Cleveland Police
- Greater Manchester Police
- Metropolitan Police Service (four Operational Command Units)
- Sussex Police
- Thames Valley Police
Target sample size
The total sample size was 1,323. This equated to approximately 220 participants in each pilot force.
- 110 were randomly assigned to the treatment group.
- 110 to the control group.
Participants - inclusion criteria
Serving police officers who were:
- in operational roles likely to carry out stop and search (for example, response and neighbourhood officers)
- available for training and ordinary duties during the evaluation period
- regular users of stop and search in 2014/15
The pilot training consisting of:
- a pre-read and knowledge check
- a one-day face-to-face classroom session designed and delivered locally by the pilot forces based on Guidance for Trainers produced by the College
The intended focus of the pilot training was:
- practical knowledge of the law and decision-making
- unconscious bias
- procedural justice
The randomised controlled trial used a stratified parallel-groups design, with officer-level randomisation.
The study included:
- an impact evaluation (carried out with the Research Advisory Service) that examined the effect of the pilot training, comparing the treatment and control groups post-test (stop and search data was compared pre- and post-test)
- a process evaluation (carried out with RAND Europe) that examined the nature and quality of implementation in the pilot forces
- Officers’ knowledge (online survey)
- Officers attitudes (online survey)
- Officers’ anticipated behaviours in written scenarios (online survey)
- Arrest rates from recorded searches (police data)
- The quality of recorded grounds for searches (blind-coded police records)
- Monitored data (direction of change was not hypothesised)
- The number of recorded searches
- The ethnic profile of recorded searches
Summary of findings
The impact of the pilot was mixed overall. The pilot training had some small positive effects on officers’ knowledge, attitudes and anticipated behaviours immediately after the training was delivered. Some of these effects were sustained at a three-month follow-up.
No effects were found on recorded stop and search practices (namely the quality of written grounds for search or arrest rates). Substantial variation was found in training delivered by the pilot forces, although this was possibly because the trainer guidance was not sufficiently prescriptive.
Paul Quinton and Daniel Packham; College of Policing stop and search training experiment: An overview; 2016
Paul Quinton; College of Policing stop and search training experiment: Design of the randomised controlled trial; 2016
Joel Miller and Banos Alexandrou; College of Policing stop and search training
experiment: Impact evaluation final report; 2016
Chris Giacomantonio, Tal Jonathan-Zamir, Yael Litmanovitz, Ben Bradford, Matthew Davies, Lucy Strang, Alex Sutherland; College of Policing stop and search training experiment: Process evaluation; 2016