Written by Barnaby Prince Police Community Support Officer at Lincolnshire Police Dementia Safeguarding Lead & Member of the force Safeguarding Network Police lead for the Atlantis Alliance – Collaborative Coastal Working Every summer, the Lincolnshire coastline attracts millions of visitors. Families arrive for rest, recreation, and connection but behind the scenes, policing, rescue and community services prepare for intense demand, heightened risk, and devastating consequences if things go wrong. This blog tells the story of Operation ATLANTIS, a multi‑agency, evidence‑led response to missing children on Lincolnshire beaches. It is not a story of a single intervention or a “silver bullet”. Instead, it is a practical example of how evidence‑based policing, structured problem solving, and true partnership working can reduce harm, save lives, and deliver real value for policing and the public. The Catalyst: Why Change Was Necessary In August 2022, a child was reported missing on the Lincolnshire coast. A large‑scale multi‑agency search was launched. After five hours, the child was tragically located in the water and pronounced dead at the scene. No agency failed. Every service responded professionally and with commitment. Yet the incident exposed a difficult truth: the system was almost entirely reactive. That moment became the catalyst for change. SCAN: Understanding the True Scale of Harm Following the 2022 season, neighbourhood teams at Mablethorpe and Skegness conducted a detailed review with partner agencies. What the data revealed Police systems recorded 22 lost or missing children from beaches. RNLI lifeguard data recorded 56 reports, including 39 multi‑agency searches. On the day of the fatal incident, officers recalled around 10 missing children, yet only four were formally logged. The conclusion was clear but uncomfortable: despite best efforts, policing did not fully understand the scale or nature of the problem. Patterns emerged even within incomplete data: Two consistent hotspots: Skegness Central Beach and Mablethorpe Central Beach Peak risk hours between 1300–1900, with highest risk 1600–1900 All children were under 13, non‑local, and visiting the county National RNLI data confirmed this was not unique, Skegness consistently ranked among the busiest beaches nationally for multi‑agency missing‑child searches. ANALYSE: Why Was This Happening? Partner analysis and frontline debriefs identified multiple contributory factors: Lost‑child searches are extremely resource‑intensive, often pulling officers from core policing duties Officers were deployed close to water without appropriate rescue equipment Preventative engagement was minimal and largely shouldered by policing alone Agencies routinely worked in parallel rather than together Communication relied on separate radio systems, creating risk during live incidents The working hypothesis was that children became lost due to: Crowd density and spatial disorientation Lack of local knowledge “Holiday psychology” - a perception that beaches feel inherently safe Across all findings, one gap stood out: prevention. RESPOND: Operation ATLANTIS Using the SARA problem‑solving model, partners redesigned their approach from the ground up. Building the Alliance Agencies brought together included: Lincolnshire Police RNLI Lifeguards and Lifeboats HM Coastguard National Coastwatch Local councils Holiday parks and community businesses Joint training, tabletop exercises, and honest debriefs clarified roles, formalised shared objectives, and embedded daily points of contact. A Visible, Preventative Model The Sandi Starfish wristband scheme became central: Free wristbands for children and vulnerable adults Highly visible safe‑point flags staffed by trained partners Distribution by police, lifeguards, Coastwatch, businesses and holiday parks Children knew where to go. Adults knew what to do. Prevention became everyone’s responsibility. Smarter Responses On operational days: Teams staffed hotspots between 1100–1900 Every incident logged via police CAD and a shared missing‑child log Single points of contact coordinated searches Early dynamic risk assessments prevented escalation In 2024, partners trialled shared Airwave and VHF communications, significantly improving live incident coordination. An anonymous staff survey later showed 79.16% of respondents agreed or strongly agreed that the radio trial had a positive impact on their work. ASSESS: What Changed? Across 2023–2025, the results were clear: 96% reduction in reports of lost and missing children Minimum policing savings of £85,000 This saving was calculated using national average investigation costs: Low/medium risk incidents: £2,500 High‑risk incidents: £8,500 Based on reduced demand, this equated to: £72,500 saved between 2022/23 £12,500 saved between 2023/24 Additional outcomes included: 607 recorded staff engagement hours across 44 days in hotspot areas 22,000 pieces of crime‑prevention and water‑safety literature distributed Frontline and partner feedback echoed these findings: “In my 15 years with the RNLI, this is the closest I’ve seen our organisations work together and it’s translated directly into better outcomes.” An independent review by Lincoln University corroborated the findings and the evaluation report can be provided on request. Learning Forward Success did not end the learning cycle. New risks emerged including paddleboards drifting, weaver fish injuries, and UXO washing ashore, all now addressed through partnership planning. Future priorities include: Reaching families before they travel Better data capture on public engagement Joint, synchronised media messaging Expanding interactive engagement methods Continuing regular multi‑agency debriefs year‑round Why This Matters Operation ATLANTIS demonstrates that: a tested approach (POP) can be practically applied Prevention saves lives, money, and wellbeing Collaboration requires structure, trust and accountability The SARA model is not abstract - it is deeply practical Most importantly, it shows that complex safeguarding problems cannot be solved by policing alone. If there is one question this project leaves for all of us, it is this: “What are we only reacting to that we could prevent together?”

Here are the studies we found this month that may be worth your attention. As ever, if you spotted something we missed, let us know and we'll add it into the next brief. TL;DR Summary Strip searches in police custody and ethnicity Black detainees face significantly higher odds of being strip-searched in custody, even after controlling for offence type, vulnerability and use of force. What shapes police officers' attitudes toward AI-assisted policing Organisational justice and supervisory support significantly shape whether officers accept AI tools. A community policing communication field experiment A brief collaborative message during door-to-door visits significantly improved citizen perceptions and reporting intentions. School exclusion, missing children and serious harm School exclusion, exploitation and missing incidents frequently overlap in the most seriously harmed child cases. Automated feedback and AI body-worn camera review Automated positive feedback emails improved officers' acceptance of AI-driven body-worn camera review. Strip searches in police custody and their association with ethnicity: evidence from an English force Authors: Ali and Dymond Criminology and Criminal Justice Study design: Quantitative analysis of administrative custody data (binary logistic regression, N=25,676) Setting: England and Wales Summary: Using four years of custody records from an English force, this study finds that Black detainees face over twice the odds of being strip-searched compared to White detainees, even after controlling for offence type, vulnerability indicators and use of force. Black men and Black children face disproportionately higher odds than either characteristic alone would predict, providing evidence of racialised policing in the custody suite. You'll be interested if: you work in custody, equality, professional standards, inspection or research, or if you are responsible for ensuring policing is fair and lawful. Relevant to all forces given proposed PACE Code C changes. What shapes police officers' attitudes toward AI-assisted policing Authors: Jian, Sun, Wu et al Journal of Experimental Criminology Study design: Vignette experiment Setting: Taiwan Summary: This vignette experiment tests what contextual factors shape police officers' willingness to accept AI-assisted tools. It finds that organisational justice, supervisory support, and the perceived purpose of the AI system all significantly influence officer attitudes. Officers are more accepting when they believe the technology supports rather than replaces professional judgement. You'll be interested if: you are implementing AI tools in a force, leading technology change, or researching the human side of AI adoption in policing. How police communication influences citizen perceptions and cooperation: a community policing field experiment Authors: Canales and Santini Journal of Experimental Criminology Study design: Randomised field experiment (N=980 households) Setting: Mexico Summary: Officers conducting door-to-door visits were randomly assigned to vary their uniform type and whether they delivered a brief collaborative safety message. The message condition significantly improved citizen perceptions of police and their willingness to report crime. Uniform type alone had little effect. Findings support procedural justice theory and suggest low-cost communication changes can meaningfully improve public cooperation. You'll be interested if: you work in neighbourhood policing, community engagement or public confidence. Note: Mexican context. School exclusion, missing children and serious harm: disentangling the interlinking factors Authors: Fox CrimRxiv (preprint — not yet peer reviewed) Study design: Mixed-methods analysis of Child Safeguarding Practice Reviews using machine learning-assisted content analysis Setting: UK Summary: This preprint analyses Child Safeguarding Practice Reviews to examine how school exclusion, missing child incidents and exploitation co-occur in the most seriously harmed cases. Findings show these factors frequently intersect, with exclusion functioning as both a vulnerability indicator and a pathway to further harm. Uses accessible machine learning methods to analyse sensitive text data at scale. You'll be interested if: you work in child protection, exploitation, vulnerability, or multi-agency safeguarding. Also relevant for those interested in how AI-assisted text analysis can support research with sensitive case review data. Automated feedback and the acceptability of AI-generated body-worn camera review: an implementation science natural experiment Authors: Watts, del Pozo, White and Malm Journal of Experimental Criminology Study design: Natural implementation science experiment (within context of two randomised controlled trials) Setting: United States Summary: Automated emails highlighting positive professionalism scores from AI-driven body-worn camera analysis were associated with significantly improved officer acceptance of the technology. This is the first implementation science study in a police setting and finds that feedback design, not just tool validity, matters for successful adoption. You'll be interested if: you are implementing body-worn camera AI, leading technology adoption, or interested in what implementation science can offer to evidence-based policing.

SEBP is pleased to highlight an important update from the All‑Wales Policing Academic Collaboration (AWPAC) - a partnership that exemplifies what can be achieved when policing and academia work side by side with a shared purpose. In a recent reflection, Temporary Assistant Chief Constable Ross Evans, Co‑Chair of AWPAC, set out the scale of progress made across Wales and the growing impact of collaborative, practice‑focused research. His message is clear: Wales has built something genuinely pioneering, and its momentum must continue. Where policing meets research: Transformative impact across Wales By Temporary Assistant Chief Constable Ross Evans, Co-Chair of AWPAC "As I prepare to step down from my role as co-chair of the All-Wales Policing Academic Collaboration (AWPAC), I’ve been reflecting on just how far we have come - and how vital it is that this work continues to grow. What began as a shared ambition to strengthen ties between policing and academia in Wales has evolved into the nation’s first collaboration of its kind: a nationally recognised model of evidence‑based partnership that is shaping practice, policy, and future research capacity across the country. AWPAC’s core purpose has always been clear: to bring policing and academia together to advance evidence-based change and ultimately keep our communities safer. Brought together by the Wales Innovation Network (WIN), and supported by Policing in Wales, AWPAC has spent the past several years transforming that purpose into meaningful, tangible results. AWPAC has funded fifteen research projects so far, each a collaboration between researchers and police officers, producing recommendations now being embedded into policing practice across Wales. These aren’t abstract academic exercises - they are shaping the way officers engage with communities, respond to harm, and address emerging challenges. Some examples include: Understanding misogynistic behaviour and predicting domestic abuse risk A Bangor University project with North Wales Police and Dyfed-Powys Police is exploring how indicators of misogyny can help predict domestic abuse risk, with the aim of feeding these findings into training and operational practice. Strengthening specialist units tackling violence against women and girls Cardiff University researchers partnered with all Welsh police forces to evaluate violence against women and girls specialist units. This work has provided evidence to support future policing plans and improvements. Improving community engagement and training Findings from a Bangor University–led project, developed with North Wales Police and university partners, have been incorporated directly into North Wales Police’s communications strategy, strengthening how the force connects with the public. Boosting student volunteering A multi‑agency project involving South Wales Police, University of South Wales, University of Wales Trinity Saint David, and Cardiff Metropolitan University identified the barriers and enablers to volunteering within policing. The resulting recommendations are already informing new approaches to student engagement. These projects demonstrate what is possible when police and academic partners work closely and openly together: research with immediate relevance, shaped by policing needs, to improve practice today while shaping the policing of tomorrow. One of AWPAC’s greatest achievements is how it has built on existing local connections to establish networks that now operate at an all-Wales level. For the first time, policing research in Wales is connected, collaborative, and strategically aligned. Through AWPAC, policing and academic colleagues have: formed national research networks that strengthen collaboration connected with major funders such as UKRI and NIHR, opening doors for wider research investment into Wales built links with UK and international networks including the College of Policing, the Welsh Centre for Crime and Social Justice, the N8 Policing Research Partnership, and violence against women, domestic abuse and sexual violence networks pursued global opportunities, such as work between three Welsh universities and the National Forensic Sciences University in India, supported through Global Wales This connectivity is invaluable. It raises Wales’ profile, increases our influence, and ensures we are part of the conversation about the future of policing research. A particular point of pride for the group has been AWPAC’s support for early career researchers and PhD students. Three AWPAC-funded projects were led by early career academics, helping build essential skills, confidence, and connections within the policing research community. We have also supported wider academic projects not directly funded by AWPAC, such as facilitating connections for Project DRAGONS and DRAGONS+, which focus on countering online child sexual exploitation. This commitment continues with AWPAC now funding the development of a training resource to prevent online grooming. This work lays the foundations for the next generation of Welsh policing research, built through genuine collaboration between universities, police forces and police and crime commissioners. By co‑designing research and sharing expertise, we ensure that innovation is driven jointly by academic insight and frontline experience. As my tenure as co-chair comes to an end, one thing is abundantly clear: AWPAC’s work is vital and we must continue to build on the strong foundations that we’ve established. The challenges facing policing are complex and evolving - from digital crime and technological change, to vulnerability and community cohesion. Using evidence to guide policing practice is now fundamental to effective policing. AWPAC has proven what sustained collaboration can achieve. We have shown that well-designed research, rooted in real-world need, can transform practice, shape strategy, and improve outcomes for the public. But there is more to do. More questions to answer. More innovation to unlock. More opportunities to connect Wales with national and international best practice. That is why continued and expanded investment in AWPAC is so important. With the right support, this collaboration can reach even further, bringing greater insight, stronger partnerships, and deeper impact for policing across Wales. AWPAC is a powerful example of what happens when policing and academia join forces with a shared purpose. I look forward to seeing this collaboration continue to grow, evolve, and make Wales a leading voice in evidence-based policing. To all our academic partners, police colleagues, funders, and supporters of AWPAC: thank you."

Event library Webinars Conferences National Evidence Based Policing (NEBP) Conference 2025 10-11 September 2025 (University of Huddersfield, UK) SEBP Midlands Conference 2025 24 June 2025 (De Montfort University, Leicester, UK) Fighting crime with legitimacy and procedural justice 14 June (The Royal Society, London, UK) BETA Contact us to help improve this site 2018-2024 What Works for the Frontline? SEBP Annual Conference, 11-12 September 2024 The Importance of Focus: How EBP can Drive Precision Policing. SEBP Annual Conference, 9-10 November 2023 10 Challenges to Policing. Weekly webinar series, 24 March - 26 May 2021 SEBP Annual Conference, 1-2 March 2018

BETA Contact us to help improve this site Our people Board of Trustees Ben Linton Chair Ben held a wide variety of leadership, operational and strategic roles over nearly 17 years at the Metropolitan Police Service, ending his time as a Chief Inspector. Ben became Chair of Trustees in 2026 Caroline Fiennes Trustee Caroline is an advocate for evidence-based philanthropy, and has worked on this for over 20 years. She has advised many donors and foundations on many issues in many countries. Lewis Linton Treasurer Lewis has helped various businesses navigate their finances over the years, and has supported the SEBP since it was founded. Alex Murray OBE Trustee Alex is the founder of SEBP. He was previously Chief Constable at West Mercia Police, Commander for Serious Violence at the Metropolitan Police Service and Assistant Chief Constable at West Midlands Police. He is currently Director of Police AI. Simon Ruda Trustee Simon has worked for and with police for over a decade, including as Director of Home Affairs & International Programmes at BIT, and Head of the Strategic Insight Unit at the Metropolitan Police Service. He is currently an independent consultant. Executive Committee The Executive Committee (ExCo) has powers delegated to it by the Board of Trustees. It is responsible for the day to day delivery of SEBP's strategy. Alex Murray OBE Chair of Executive Committee As Chair of ExCo, Alex leads the operational delivery of SEBP. He is the founder of SEBP, formerly T/CC at West Mercia, Director at the NCA and is currently Director of Police AI. C/ Supt Neil Ralph Vice Chair: Leadership As Vice Chair: Leadership, Neil is responsible for co-ordination of the regional volunteer network embedded within policing. Dr. Matt Bland Chief Operating Officer Responsibility for the operational delivery of the SEBP's mission and objectives. He has worked previously in policing and academia. Matt is a full-time employee at SEBP. Helen Khezrzadeh Head of Engagement Helen is a full-time employee. She is responsible for leading on SEBP's engagement with the policing profession and leads the volunteer support teams for communications, events and programmes James Watson Communications James is the Lead Service Designer for Policing and Security at the Home Office. He advises SEBP on design principles for its digital assets and conferences. Suasnne Knabe-Nicol Communications Responsibility creating for educational and promotional media content for the SEBP. Susanne is a Lecturer and Programme Leader at London Middlesex University and the Police Science Dr. T/ACC Kate Anderson Regional Co-ordinator Kate was appointed temporary Assistant Chief Constable in September 2025. Kate has worked in Lincolnshire Police since 2021, having previously worked in Cambridgeshire Constabulary and for Northamptonshire Police. Dr Jennifer Norman Regional Co-ordinator Jen is Head of Policing at the Open University and co-lead for our South East region. Regional Co-ordinators Our volunteers are at the heart of everything we do and our Regional Teams span the whole of the UK. Each co-ordinates evidence-based practice requirements and developments in their constituent organisations across policing, academia and beyond. C/ Supt Neil Ralph Vice Chair: Leadership As Vice Chair: Leadership, Neil is responsible for co-ordination of the regional volunteer network embedded within policing. He is also the Regional Co-ordinator for the South West Region. T/ACC Kate Anderson Midlands Co-ordinator Kate was appointed temporary Assistant Chief Constable in September 2025. Kate has worked in Lincolnshire Police since 2021, having previously worked in Cambridgeshire Constabulary and for Northamptonshire Police. Dr Jennifer Norman South East Co-ordinator Jen is Head of Policing at the Open University and co-lead for our South East region. Supt. Mat Burbeck South East Co-ordinator Mat is a local policing Superintendent at Kent Police. Supt. James Sutherland East Co-ordinator James is Head of Organisational Development at Cambridgeshire Constabulary Millie Garrod East Co-ordinator Millie is a Higher Analyst at Norfolk and Suffolk Constabularies Julia Hall North West Co-ordinator Julia is Head of Design for Security at Greater Manchester Police Anna Baker Wales Co-ordinator Anna is Strategic Health and Justice Lead for the Police and Crime Commissioner in North Wales. C/Insp. Peter Bodley London Co-ordinator Peter is a Chief Inspector for the Metropolitan Police Service in London. Insp. Ross Greenwood Yorkshire Co-ordinator Ross leads the EVOLVE team for South Yorkshire Police Dr Ben Havers BTP Co-ordinator Dr Havers is Police Strategic Research Manager for British Transport Police DI Richard Simpson PSNI Co-ordinator Supt. Jon Harris Scotland Co-ordinator Jane Sanderson Midlands Co-ordinator

Understanding evidence The scientific process The scientific process is the foundation of evidence-based policing (EBP). It provides a systematic way to investigate questions, evaluate interventions, and make informed decisions. While experimentation is at the heart of EBP, not all evidence involves conducting experiments—some evidence is observational, and both types play a crucial role in understanding what works. The Steps of the Scientific Process The scientific process typically involves: Observation : Identifying a problem or phenomenon, such as an increase in burglaries in a specific area. Hypothesis Formation: Proposing an explanation or solution, for example, that increasing visible patrols will deter burglaries. Testing: Gathering data through experimentation or observation to test the hypothesis. Analysis: Evaluating the results to determine if the evidence supports the hypothesis. Replication: Repeating the process to confirm the findings. This process ensures that conclusions are based on reliable evidence rather than assumptions or anecdotes. Experimental vs. Observational Evidence Experimental Evidence: Experiments, such as randomised controlled trials (RCTs), are considered the gold standard for determining causation. They test whether a specific intervention, like hotspot policing, directly causes an outcome, such as reduced crime, by comparing results between a treatment group and a control group. Observational Evidence: Observational studies, such as analysing crime trends over time or interviewing victims about their experiences, involve collecting data without manipulating the environment. While these studies can identify correlations and patterns (e.g., increased patrols are linked with reduced crime), they cannot definitively prove cause and effect. One is not better than the other. Both types of evidence are essential: Experiments provide strong evidence of causation. We did this, and that happened as a result. Observational studies help identify patterns and generate hypotheses about how those patterns work, especially when experiments are impractical or unethical. Why This Matters for EBP In EBP, understanding the difference between correlation (things that happen together) and causation (one thing directly causes another) is critical. For example: Correlation: A study might observe that areas with more CCTV cameras have lower crime rates. But does the presence of CCTV reduce crime, or are cameras simply more common in safer areas? Causation: An experiment that installs CCTV in some areas but not others can test whether the cameras directly lead to crime reduction. Policing decisions must consider this distinction: Observational evidence can inform where to focus efforts or identify trends. Experimental evidence helps determine whether a specific action is effective and replicable. The scientific process ensures that policing practices are grounded in robust evidence. By combining observational and experimental approaches, practitioners can make better decisions about what works, for whom, and in what context. Grades of evidence Not all evidence is equal. Fortunately, there are various tools to help us differentiate. The Maryland Scientific Methods Scale (SMS) along with frameworks like EMMIE (which is used by the What Works for Crime Reduction toolkit) and the Youth Endowment Fund (YEF) toolkit, helps us evaluate the strength of research evidence. These tools differentiate studies based on their ability to establish causal relationships—how confidently we can say that A causes B. It’s important to emphasise that these frameworks are not about dismissing lower-level studies. Observational studies, for instance, serve different purposes than experimental ones. Instead, the goal is to understand the relative strength of evidence when making decisions about what works. Maryland Scientific Methods Scale The Maryland SMS ranks studies on a five-level scale based on the rigour of their methodology, particularly their ability to assess causality: Level 1: Correlation Studies Studies that describe a pattern such as an association between A and B but cannot determine if A causes B. Example: Crime is observed to drop in areas with CCTV, but the study doesn’t consider other factors that might influence crime rates. Purpose: Useful for identifying patterns and generating hypotheses. Level 2: Before-and-After Studies Without a Comparison Group Measures outcomes before and after an intervention but lacks a control group to isolate the intervention’s effect. Example: Assessing crime rates before and after introducing street lighting in one neighbourhood. Limitation: Cannot rule out other influences. Level 3: Before-and-After Studies with a Comparison Group Introduces a comparison group not exposed to the intervention, helping isolate its effects. Example: Comparing crime rates in one area with new street lighting to another area without it. Strength: Stronger indication of causality but still open to biases because we cannot be sure the comparison area is like-for-like Level 4: Quasi-Experimental Studies Uses techniques like matched comparisons to minimise bias, though not fully randomised. Example: Pairing similar areas (one with CCTV, one without) to measure differences in crime. Strength: Provides robust evidence when RCTs aren’t feasible. Level 5: Randomised Controlled Trials (RCTs) Participants are randomly assigned to intervention and control groups, eliminating most biases. Example: Randomly deploying extra patrols in some areas and not others to measure crime reduction. Gold Standard: The most reliable method for establishing causality. EMMIE: What Works for Crime Reduction Toolkit The EMMIE framework evaluates interventions across five dimensions, not just effectiveness: Effectiveness: Does it work? How well? Mechanism: How does it work? What is the underlying process? Moderators: Under what conditions does it work best? Implementation: How easy or practical is it to deliver? Economic Cost: Is it cost-effective? This approach ensures a broader evaluation, recognising that effectiveness alone doesn’t tell the whole story. YEF Toolkit Scale The Youth Endowment Fund (YEF) toolkit categorises evidence into three levels: Strong: High-quality studies, typically RCTs or systematic reviews, showing clear, consistent results. Promising: Studies with some methodological limitations but still providing useful insights. Mixed or Weak: Evidence with significant limitations or inconsistent findings. The YEF emphasises transparency, helping users understand where confidence in the evidence is strongest. Why Grades of Evidence Matter Understanding these scales helps differentiate the strength of evidence: Strong Evidence: Provides confidence that an intervention causes an effect. Weaker Evidence: Useful for generating ideas, identifying trends, or guiding further research, but less reliable for determining causality. The point is not to create a hierarchy or dismiss certain studies but to illustrate that not all research is equally robust. Even published studies can vary widely in quality. When using evidence, it’s essential to: Critically Assess: Consider the study’s methodology, sample size, and potential biases. Combine Evidence: Use a mix of observational and experimental studies to build a fuller picture. Apply Judgement: Understand the strengths and limitations of the evidence before acting. By appreciating the differences in evidence quality, practitioners can make more informed, effective decisions in their policing strategies. Research methods Evidence in evidence-based policing (EBP) is the product of research designs, which are structured approaches to answering questions and solving problems. Different research designs serve different purposes, and each has its own strengths and limitations. Understanding these designs helps practitioners evaluate the evidence and apply it effectively. Observational vs. Experimental Designs Research designs can broadly be grouped into two categories: Observational Designs: These involve collecting and analysing data without directly intervening. They are ideal for identifying patterns, correlations, or trends but cannot establish causation (i.e., whether A causes B). Experimental Designs: These involve actively testing an intervention by manipulating variables (e.g., introducing a policy change) and observing the outcomes. They are better suited for determining causality. Both types of designs are important in EBP, and the choice depends on the question being asked. Qualitative and Quantitative Research Research methods can also be divided into qualitative and quantitative approaches, which differ in their focus and outputs: Qualitative Research: Focuses on exploring experiences, perceptions, and contexts in depth. Produces rich, descriptive data (e.g., interviews, focus groups, observations). Example: Interviews with community members to understand their trust in the police. Strengths: Helps define problems, explore complex issues, and generate hypotheses for further testing. Limitations: Findings are not easily generalisable or measurable. Quantitative Research: Focuses on measuring and analysing numerical data. Uses structured methods (e.g., surveys, experiments, statistical analysis) to test hypotheses. Example: Measuring changes in crime rates before and after an intervention. Strengths: Provides measurable, comparable results and is essential for evaluating effectiveness. Limitations: May oversimplify complex issues by focusing solely on numbers. Mixed-Methods Research: Combines qualitative and quantitative approaches to leverage the strengths of both. Example: A study might use surveys to measure public confidence (quantitative) and interviews to explore why trust is increasing or decreasing (qualitative). Strengths: Provides a fuller understanding of issues by integrating numerical data with contextual insights. Examples of Research Methods Here are some common methods used in EBP and their purposes: Observational Methods Case Studies: Detailed exploration of a single example or incident. Useful for understanding unique or complex issues. Cross-Sectional Surveys: Collect data at a single point in time to identify patterns or relationships. Longitudinal Studies: Observe changes over time, such as tracking crime rates over several years. Experimental Methods Randomised Controlled Trials (RCTs): Participants are randomly assigned to intervention or control groups to test causality. Quasi-Experiments: Similar to RCTs but without full randomisation, often used when random assignment isn’t possible. Natural Experiments: Observing the impact of real-world changes or policies without direct manipulation. Mixed-Methods Approaches: Combining surveys with interviews to evaluate both the measurable and contextual impact of an intervention. Why EBP Focuses on Quantitative Research Because EBP is primarily concerned with effectiveness—understanding what works and why—it tends to rely more heavily on studies with quantitative measurements. Quantitative methods are better suited to: Measuring changes in outcomes (e.g., crime rates, arrest numbers). Comparing interventions (e.g., whether one strategy reduces harm more than another). However, qualitative research plays a critical role in EBP by: Defining Problems: Providing insights into the lived experiences of communities or officers. Establishing Hypotheses: Highlighting issues or mechanisms to be tested in quantitative studies. Complementing Quantitative Data: Adding depth and context to numerical findings, helping to explain the “why” behind the “what.” For example, if a quantitative study finds that visible patrols reduce fear of crime, qualitative interviews can explore why people feel reassured by seeing officers. Choosing the Right Method The best research design depends on the question being asked: Exploratory Questions: Use qualitative or mixed-methods approaches to understand context or generate hypotheses. Causal Questions: Use experimental or quasi-experimental methods to determine what works. Monitoring Trends: Use observational methods to track changes over time or across groups. Each method has its place in the evidence base and combining them provides the strongest foundation for decision-making. Understanding research methods allows practitioners to critically assess evidence, ensuring that the decisions they make are based on reliable, relevant, and robust findings. Measurement Measurement is at the heart of any study in evidence-based policing (EBP). Whether it’s tracking crime rates, gauging public confidence, or evaluating the effectiveness of an intervention, the tools and methods used to measure outcomes determine the quality and trustworthiness of the evidence. Why Measurement Matters Measurement is more than just collecting data—it’s about ensuring that the data accurately represents what you’re trying to study. Without good measurement, even the most sophisticated study can produce misleading or unreliable results. For example: If a study measures "crime reduction" by counting arrests, it may not truly capture whether crime has decreased, as arrests don’t always equate to fewer offences. If a survey asks unclear or biased questions about trust in police, the results might not reflect actual public sentiment. This is where two key concepts come in: validity and reliability. Measurement Validity Validity refers to whether a measurement instrument accurately measures what it claims to. In plain terms: Are you measuring the right thing? Example 1: A study aims to measure fear of crime but only counts the number of reported burglaries. While burglary rates might be related, they don’t directly measure people’s fear, so the validity of this approach is low. Example 2: A public confidence survey uses clear and unbiased questions to gauge trust in the police. This increases the validity because the survey directly measures what it’s supposed to. Measurement Reliability Reliability refers to whether a measurement instrument produces consistent results. In other words: Can you trust the data to be the same if measured again? Example 1: A radar gun used to measure vehicle speeds gives the same readings under the same conditions, making it highly reliable. Example 2: An officer’s handwritten notes about incidents vary greatly between shifts. This inconsistency reduces reliability. A reliable instrument isn’t necessarily valid. For example, a flawed survey question might consistently produce the same wrong result, making it reliable but not valid. Interpreting Evidence Through Measurement Understanding validity and reliability is crucial when interpreting evidence. Before trusting the results of a study, ask: What was measured? Does the measurement align with the study’s goals? For example, measuring "officer workload" by counting hours worked might miss key aspects like case complexity. How was it measured? Was the measurement tool robust? For instance, was a standardised survey used, or was it created without testing? Is the data consistent? Were the same methods applied across all participants or settings? Inconsistent methods can weaken the findings. Why Measurement is Important for EBP Good measurement ensures that: Evidence is accurate: Policymakers and practitioners can trust the findings. Comparisons are meaningful: Data can be compared across studies, locations, or time periods. Interventions are properly evaluated: Success or failure is judged based on reliable and valid criteria. For example: A hotspot policing study might use the crime harm index to measure reductions in the severity of crimes in targeted areas. The index is somewhat valid because it reflects a perspective of harm, not just the number of incidents. A community engagement survey might use validated scales to measure trust and confidence, ensuring the findings are both reliable and accurate. What Practitioners Should Consider When evaluating evidence, pay attention to how measurements were done: Were the instruments tested for validity and reliability? Were clear and consistent methods used across participants or sites? Are the results presented transparently, with limitations acknowledged? These considerations help you critically assess the strength of the evidence and make better decisions based on the findings. Measurement is the foundation of good evidence. By understanding and applying the principles of validity and reliability, practitioners can ensure that the data they use is fit for purpose and leads to effective, evidence-based decisions. Why most EBP studies are quantitative When considering “what works” in policing, evidence-based policing (EBP) often leans towards quantitative evaluation. This is because many of the key questions EBP seeks to answer—such as whether an intervention reduces crime, improves efficiency, or enhances accountability—require measurable, numerical data. Quantitative studies are well-suited for evaluating these outcomes in a clear and standardised way. The Role of Quantitative Studies Quantitative research focuses on numbers and measurements. It allows practitioners to: Evaluate Effectiveness: For example, comparing crime rates before and after deploying hotspot policing. Demonstrate Accountability: Police forces can use quantitative results to justify decisions to the public or policymakers. Track Performance: Quantitative metrics, such as response times or conviction rates, are critical for assessing performance over time. Support Resource Allocation: Data-driven decisions ensure resources are targeted where they have the most impact. In the real-world context of policing, where budgets are tight and results are scrutinised, the ability to quantify impact is invaluable. For instance, demonstrating a reduction burglary rates after implementing a new strategy provides clear evidence of success. Where Qualitative Studies Fit In Quantitative studies are essential, but qualitative research plays a crucial role in complementing and enhancing EBP. Qualitative studies explore the lived experiences of individuals and communities, providing depth and context to numerical findings. They are particularly valuable for: Defining Problems: Understanding the root causes of issues or identifying emerging trends. Example: Interviews with community members might reveal concerns about a lack of visible policing that wouldn’t appear in crime data alone. Developing Interventions: Ethnographic research (observing and analysing behaviours within a community) can inspire innovative approaches to policing challenges. Example: Observations in a neighbourhood could highlight informal networks that influence local crime dynamics. Contextualising Results: Qualitative research adds meaning to quantitative findings, explaining not just what happened but why and how. Example: If hotspot policing reduces crime, interviews with officers and residents can uncover whether the reduction was due to deterrence, displacement, or increased community trust. EBP does not view qualitative research as “lesser.” Instead, it recognises that qualitative and quantitative methods serve different but complementary purposes. An Example Imagine a police force wants to address rising levels of youth violence: Defining the Problem with Qualitative Research: Conduct interviews with youth workers, teachers, and community members to hypothesise the root causes (e.g., lack of opportunities, gang influence). Conduct ethnographic observations in hotspots to reveal how and where violence tends to escalate. Testing an Intervention with Quantitative Research: Implement a youth mentorship programme and use quantitative methods (e.g., a randomised experiment) to evaluate its effectiveness. Contextualising the Results: If the programme reduces violence, qualitative interviews with participants and mentors could help explain why—perhaps the mentors provided role models, or the programme offered much-needed support structures. By combining these methods, the force can not only measure success but also understand and refine the intervention for future use. Balancing Quantitative and Qualitative Approaches While EBP often focuses on quantitative studies to answer “what works,” it values the insights qualitative research provides. Together, these approaches ensure: Holistic Understanding: Policymakers and practitioners get a full picture of a problem and its solutions. Practical Applicability: Interventions are informed by both hard data and the realities of lived experience. Ethical Awareness: Decisions consider the human impact alongside measurable outcomes. Quantitative studies may dominate EBP because of their ability to demonstrate impact in measurable terms, but qualitative studies are equally critical for shaping, interpreting, and improving policing practices. The most effective EBP draws on both to inform smarter, more empathetic decision-making. Statistics Statistics are essential tools in evidence-based policing (EBP). They help us make sense of data, identify patterns, and evaluate the effectiveness of interventions. While statistics can seem complex, they can be broken down into two main types: descriptive and inferential statistics. The goal here is not to teach you how to do statistics, but to help you understand the key concepts so you can interpret research with confidence. Descriptive vs Inferential Statistics Descriptive Statistics: These summarise data so we can understand it at a glance. Descriptive statistics don’t tell us why something happened—they just describe what the data looks like. Examples: Mean (average): The sum of all values divided by the number of values. For example, the average number of calls to a police station per day. Standard Deviation: This tells us how spread out the data is around the mean. A low standard deviation means most values are close to the average; a high one means there’s more variation. Percentages and Proportions: Used to show how often something occurs, such as the percentage of crimes resolved within a month. Inferential Statistics: These allow us to draw conclusions about a larger group (a population) based on data from a smaller group (a sample). Inferential statistics help answer questions like “Did this intervention reduce crime?” or “Is this difference meaningful or just random?” Core Concepts in Inferential Statistics: Sample vs Population: A population is the entire group you want to understand (e.g., all crimes in the UK). A sample is a smaller group you study to make inferences about the population (e.g., crimes in one city). The sample must be representative to give accurate conclusions. Standard Error: This measures how much the sample's results might differ from the population. Smaller samples tend to have a larger standard error, meaning the results are less precise. P-Values: A p-value tells us whether the results of a study are likely due to chance. Example: If a p-value is less than 0.05, it means there’s less than a 5% chance the result happened randomly. This suggests the result is “statistically significant.” Note: A small p-value doesn’t guarantee the result is meaningful in the real world—it’s just one part of the picture. Confidence Intervals: These give a range of values where we’re fairly confident the true result lies. Example: If a study finds crime reduced by 10% with a 95% confidence interval of 8%–12%, we can be 95% sure the true reduction is between 8% and 12%. Effect Sizes: Effect sizes tell us how big the difference or impact is, not just whether it exists. Example: If an intervention reduces crime by 2% in one area but 20% in another, the effect size tells us which is more impactful. Statistical Modelling and Tests Statistical models are like mathematical tools for understanding relationships between variables. For example, a model might explore whether an increase in patrol hours leads to a reduction in crime. The type of test used depends on the nature of the data: T-Test: Compares averages between two groups (e.g., crime rates in areas with and without CCTV). Chi-Square Test: Examines whether there’s a relationship between two categorical variables (e.g., crime type and location). Regression Analysis: Predicts how changes in one variable (e.g., patrol hours) affect another (e.g., crime rates). Different tests make different assumptions about the data, like whether it’s normally distributed (follows a bell curve). Choosing the wrong test can lead to misleading results, which is why statistics need to be handled with care. Why Statistics Can Be Tricky Statistics are notoriously complex, and it’s normal to feel overwhelmed when reading about them in research. Even experienced researchers sometimes struggle with advanced analyses or interpreting results correctly. Don’t feel bad if you find it difficult—it’s more important to understand the general concepts and ask questions when you’re unsure. When looking at statistical results: Look Beyond the Numbers: Statistics don’t provide all the answers—they’re tools to help interpret the data. Check Assumptions: Was the right test used for the data? Did the study explain its methodology? Focus on Practical Meaning: Even statistically significant results might not be relevant in real-world policing if the effect size is too small. Understanding these core statistical concepts allows you to interpret evidence more critically and confidently. The more familiar you become with these ideas, the easier it will be to engage with research and apply it to your work. How much can we believe in this evidence? Evidence-based policing (EBP) relies on research to guide decisions, but interpreting evidence isn’t as simple as just trusting what a study says. Practitioners need to think critically, using their judgment and experience to decide how much weight to give to the evidence and whether it applies to their specific situation. Two important concepts can help with this: internal validity and external validity. Internal Validity: How Reliable Are the Results? Internal validity refers to how well a study is designed to ensure its results are accurate and not influenced by other factors. In simple terms: Did the study really show that A caused B, or could something else explain the outcome? High Internal Validity: A study uses rigorous methods, such as randomised controlled trials (RCTs), to minimise bias and isolate the effect of the intervention. Example: An RCT tests whether increased patrols reduce street robberies by randomly assigning extra patrols to certain areas. The randomisation ensures other factors, like differences in neighbourhoods, don’t affect the results. Low Internal Validity: A study lacks controls or fails to account for other factors that might influence the outcome. Example: A study claims crime dropped after more patrols were introduced, but it doesn’t consider that the weather also changed, which might have reduced outdoor activity and crime. Key Takeaway: The higher the internal validity, the more confident you can be that the intervention caused the result but there is no fixed numerical scale for it. External Validity: Will It Work Here and Now? External validity refers to how well the findings of a study apply to other settings or situations. In other words: If it worked there, how likely is it to work here? High External Validity: A study has been tested successfully in different locations, with different populations, and under varying conditions. Example: A strategy like hotspot policing has worked in multiple cities around the world, making it more likely to succeed in your area. Low External Validity: A study was conducted in a very specific context that may not apply elsewhere. Example: A study showing that foot patrols reduced crime in a densely populated city might not apply in rural areas where population density is lower. Key Takeaway: Even if evidence has high internal validity, it’s up to practitioners to decide if the context and conditions are similar enough for it to work in their setting. Balancing Evidence with Judgment No matter how strong the evidence, there’s no guarantee an intervention will work every time or in every place. Practitioners play a critical role in interpreting evidence, using their expertise to assess: The Quality of the Evidence: Is the study well-designed and reliable? Does it have high internal validity? Scales like SMS, EMMIE and the YEF Toolkit help us here. The Relevance to Their Context: Are the conditions in the study similar to those in their area? Does it have external validity? Only you can decide. The Fit with Local Needs: Even if the intervention worked elsewhere, does it align with local resources, priorities, and community needs? Again – this is on you, the practitioner. Example: Imagine a study shows that body-worn cameras reduce complaints against officers in large metropolitan areas. If you work in a similar city with similar issues, you might expect similar results. If you work in a rural area with few complaints to begin with, you might question whether the intervention is relevant or likely to have the same impact. The Practitioner’s Role Interpreting evidence is not just about following what the research says. It’s about applying professional discretion and experience: Consider Probabilities, Not Certainties: If a study has worked multiple times in different places, it doesn’t guarantee success, but it does make it more likely. How much more likely is for you to decide, based on the evidence and your context. Think Critically: Ask questions like: What might make this work—or fail—here? Are there local factors the study didn’t account for? Combine Evidence with Local Knowledge: Use the research as a guide but adapt it to your specific circumstances. Add to the evidence: don’t just try it, test it. Put in place an evaluation framework of your own and feed it back to other so they can learn from your experiences. This is the scientific process in action. Believing in evidence isn’t about blind trust. It’s about weighing its strengths and weaknesses, considering its relevance, and making informed decisions that suit your unique situation. Critics of EBP sometimes point to it being a polar opposite to experience and craft-based knowledge. It really isn’t. It’s something that fits in underneath those things. Evidence provides the map, but you decide the route. Reading scientific papers The reality of working in evidence-informed ways will inevitably lead you to have to read scientific papers from time to time. These are a special kind of human invention! Scientific papers can seem daunting, especially if you’re unfamiliar with their structure or technical language. However, you don’t need to read every word to understand the key points. By focusing on specific sections, you can quickly extract the information you need for evidence-based policing. Here’s a straightforward framework to guide you. Understanding the Structure of a Scientific Paper Most scientific papers follow a standard format, often abbreviated as IMRAD: Introduction: Explains the problem and why the study was done. Methods: Details how the study was conducted, including design, data collection, and analysis. Results: Presents the findings, often with tables, graphs, and statistical data. Discussion: Interprets the results, explains their implications, and addresses limitations. Abstract: A short summary of the entire paper, typically found at the start. A Quick-Read Framework You don’t need to read a paper from start to finish. Instead, follow these steps to get the key points efficiently: Start with the Abstract: The abstract gives a snapshot of the study’s purpose, methods, main findings, and conclusions. Ask yourself: Does this study seem relevant to my question or context? Check the Introduction: Focus on the first and last paragraphs, which often explain the problem being addressed and the study’s goals. Ask yourself: Why was this study conducted? Is it tackling a problem I recognise? Skim the Methods: Look for the type of study (e.g., randomised controlled trial, observational study). Note the sample size, location, and key methods. Ask yourself: Is this method robust? Could these findings apply to my situation? Focus on the Results: Check tables and graphs for key outcomes. Look for summaries of the main findings (often highlighted in text or captions). Ask yourself: What were the main results? Are they statistically significant (e.g., p-values) and practically meaningful (e.g., effect sizes)? Read the Discussion and Conclusion: Focus on how the authors interpret their findings and the practical implications they suggest. Pay attention to any limitations they mention, as these could affect how you use the evidence. Ask yourself: What do the results mean in practice? Do the authors identify any risks or caveats? Review the References (Optional): If the study seems highly relevant, check the references for related studies to explore further. What to Look For in Each Section To get the most out of a paper, focus on these elements: The Research Question: What problem is the study addressing? The Study Design: Was it experimental or observational, quantitative, qualitative or mixed? How does this affect its reliability? The Main Findings: What does the evidence say? Are the results clear and consistent? The Implications: How do the findings relate to your policing context? The Limitations: Are there weaknesses in the study (e.g., small sample size, lack of randomisation) that might affect its applicability? Good studies will tell you about these openly. Tips for Practitioners Don’t Be Intimidated by Jargon: Focus on the big picture. You don’t need to understand every technical detail to grasp the key points. Use Your Judgment: Not all studies are equally relevant or robust. Look critically at how the evidence aligns with your needs. Start with Reviews or Summaries: If you’re new to reading scientific papers, begin with systematic reviews or toolkits like EMMIE or the YEF Toolkit. These often summarise multiple studies and provide clear takeaways. Practice Makes Perfect: The more papers you read, the easier it becomes to spot what’s important. By focusing on the key sections and asking the right questions, you can quickly understand what a paper offers and decide how its findings might apply to your work. Remember, reading scientific papers is a skill that improves with practice—and you don’t need to be an expert to make evidence-based decisions. BETA Contact us to help improve this site

In this introductory course, SEBP's Chief Operating Officer, Dr Matt Bland, aims to provide an introduction to the subject and offer a structured approach for further learning about EBP. The program begins with definitions and an overview of the scientific process. It then moves on to how research is put together and teaches tips on how to find and interpret evidence-based papers, the challenges of implementation and how to do EBP in an ethical way. The course is around 8 hours and involves cheat sheet resources, quizzes, and case studies.