top of page

Producing evidence

Defining a problem

Defining the problem is the critical first stage of any research process. Without a clear and precise understanding of the problem, the research risks being unfocused, unproductive, or irrelevant. A well-defined problem lays the foundation for a successful study, guiding your objectives, methods, and interpretation of results.

 

Why Defining a Problem Matters

 

  • Focus: A clearly defined problem keeps your research on track, ensuring that every step—data collection, analysis, and conclusions—addresses the key issue.

  • Relevance: It ensures that the research is meaningful and addresses a real-world need.

  • Efficiency: Poorly defined problems can waste time and resources, leading to inconclusive or unusable findings.

  • Impact: A well-defined problem connects directly to actionable solutions, increasing the likelihood that your research will inform decisions and make a difference.

 

What Happens When a Problem is Poorly Defined?

 

A poorly defined problem is vague, overly broad, or lacks specificity. This leads to confusion about what the research is trying to achieve and makes it harder to design effective methods or interpret findings.

 

Example of a Poorly Defined Problem:

"Why isn’t public trust in the police improving?"

 

  • Issues:

  • The problem is too broad—trust involves many factors, from officer behaviour to communication strategies.

  • There’s no clear focus or measurable outcome. Are we talking about trust in specific areas, communities, or interactions?

  • It doesn’t specify whether the concern is about declining trust, stagnant trust, or variations in trust.

 

What Does a Well-Defined Problem Look Like?

 

A well-defined problem is focused, specific, and measurable. It identifies what needs to be investigated and provides a clear starting point for the research.

 

Example of a Well-Defined Problem:

"Why has trust in police decreased by 15% in Neighbourhood A over the past three years?"

 

  • Strengths:

  • The problem is specific, focusing on a particular area (Neighbourhood A) and timeframe (three years).

  • It includes a measurable change (a 15% decrease in trust).

  • It’s actionable, allowing the researcher to investigate specific causes, such as changes in policing strategies, local incidents, or demographic shifts.

 

How to Define a Problem

 

Identify the Issue:

 

  • What specifically is not working or needs to be understood?

  • Speak to stakeholders, review data, and look for patterns.

 

Focus the Scope:

 

  • Narrow down broad concerns into specific, researchable questions.

  • Example: Instead of “Why is crime increasing?” ask, “Why have burglary rates in District B risen by 10% over the past year?”

 

Make it Measurable:

 

  • Ensure there are clear ways to measure or assess the problem.

  • Example: Define trust using survey scores or complaint levels.

 

Contextualise the Problem:

 

  • Understand the local context, past interventions, and external factors.

  • Example: Has there been a recent policy change, economic shift, or cultural event affecting the problem?

 

Align with Research Objectives:

​

  • Your problem definition should guide what you want to achieve through the research.

  • Example: “We want to identify factors contributing to rising burglary rates and recommend targeted interventions.”

 

Why Good Problem Definition Leads to Better Evidence

 

  • Clarity: Well-defined problems translate directly into clear research objectives and hypotheses.

  • Methods Fit: The problem helps determine the most appropriate research design, whether qualitative, quantitative, or mixed-methods.

  • Actionability: A focused problem ensures your findings lead to practical recommendations.

 

Defining a problem well is like setting the foundation of a building. Without a strong foundation, everything else risks collapse. By taking the time to refine and focus your problem statement, you set yourself up for a successful and impactful research project.

What is already known?

Before starting a study, it’s essential to understand what is already known about your topic. Reviewing existing evidence ensures that your research builds on previous knowledge, avoids unnecessary duplication, and is informed by the successes and challenges of earlier work.

 

For practitioners, this step isn’t about conducting a formal literature review as academics do. Instead, it’s about gathering and interpreting the most relevant evidence to guide your research.

 

Why This Step Matters

 

  • Contextualising the Problem: Existing evidence helps you understand the broader context of the issue you’re studying. For example, has a similar problem been researched before? What interventions have worked elsewhere?

  • Shaping Your Study Design: By reviewing what’s already known, you can identify effective methods, potential pitfalls, and gaps in the evidence that your research can address.

  • Strengthening Your Interpretation: Knowing the broader evidence base helps you make sense of your findings. Are they consistent with other studies? If not, what might explain the differences?

  • Saving Time and Resources: If a robust answer to your research question already exists, you can focus your efforts on applying or adapting those findings instead of starting from scratch.

 

How to Approach This Step

 

You don’t need to read hundreds of papers or get bogged down in academic jargon. Instead, take a practical, targeted approach:

 

Start with Key Questions:

 

  • What has already been tried in similar situations?

  • What are the main findings or patterns from previous research?

  • Are there gaps in the evidence that your study can address?

 

Search for Relevant Evidence:

 

  • Refer back to the Where to Find Evidence section for tips on searching for toolkits, databases, and systematic reviews.

  • Focus on high-quality evidence, such as systematic reviews, randomised controlled trials, or trusted sources like the College of Policing’s What Works Toolkit.

 

Extract Key Points:

 

  • What methods were used, and what were the outcomes?

  • Were there any challenges or limitations?

  • Are there specific recommendations or lessons that apply to your context?

 

Summarise the Findings:

 

  • Create a concise summary of the most relevant evidence. This will help you justify your research decisions and explain the importance of your study to others.

 

Practical Example:

Imagine you’re researching the impact of body-worn cameras (BWCs) on complaints against officers. You might find:

 

  • What’s Known: Multiple studies, including randomised controlled trials, show that BWCs reduce complaints. The evidence is consistent across urban police forces.

  • Gaps in the Evidence: There is limited research on how BWCs affect small rural forces, where community interactions differ significantly.

  • How This Informs Your Study: You decide to focus on a rural force and tailor your study to explore whether the context affects the outcomes seen elsewhere.

 

Using Existing Evidence to Design and Interpret Your Study

 

Design:

​

  • If previous studies show that BWCs reduce complaints in urban areas, you might adopt similar methods (e.g., comparing outcomes between officers with and without cameras).

  • If past research highlights challenges, such as officer compliance, you can build strategies to address these in your study.

 

Interpretation:

 

  • If your findings align with earlier studies, they reinforce the robustness of the evidence.

  • If your results differ, understanding what’s already known can help you explore why—perhaps your context is unique, or there were methodological differences.

 

How This Step Adds Value

 

By reviewing what’s already known, you ensure that your research is:

 

  • Grounded in Evidence: Building on existing knowledge makes your work more credible and actionable.

  • Focused on Gaps: Addressing unanswered questions ensures your study contributes something new.

  • Easier to Interpret: Comparing your findings with previous research helps identify patterns and anomalies.

 

Understanding what is already known is a vital step in producing meaningful, evidence-based research. By taking the time to review existing evidence, you not only improve your study design but also ensure your findings have real-world relevance and impact.

Random controlled trials

Randomised Controlled Trials (RCTs) are often considered the highest standard for evaluating whether an intervention works. They are designed to test causation—whether one thing (the intervention) directly causes another thing (the outcome). RCTs are commonly used in evidence-based policing to test the effectiveness of strategies, policies, or practices.

 

What is an RCT?

 

An RCT is a research method where participants are randomly assigned to either:

 

  • An Intervention Group: Receives the new approach being tested (e.g., hotspot policing).

  • A Control Group: Does not receive the intervention, allowing researchers to compare outcomes.

 

The key to an RCT is randomisation, which ensures that participants in both groups are similar at the start of the trial. This minimises bias and makes it more likely that any differences between the groups at the end are due to the intervention and not other factors.

 

Example: What Questions Are RCTs Good For?

RCTs are ideal for questions about effectiveness, such as:

 

  • “Does body-worn camera use reduce complaints against officers?”

  • In this example, officers might be randomly assigned to use cameras on some shifts (intervention group) and not on others (control group). Researchers could then compare the number of complaints received on the two kinds of shifts over a period of time.

 

Strengths of RCTs

 

High Internal Validity: Randomisation ensures that differences in outcomes are likely caused by the intervention, and not just a coincidence. Randomisation is the secret sauce of RCTs that make this design one of the most reliable methods for proving causation. You may hear or read it called ‘the gold standard’ – that’s probably going too far. There isn’t really a ‘gold standard’ because every evaluation requires compromises and has limitations.

 

Clarity of Results: Comparing the outcomes of intervention and control groups provides clear evidence about whether the intervention works.

 

Reproducibility: RCTs are (normally) designed in a way that others can replicate the study to confirm the results.

 

Limitations of RCTs

 

Complexity and Cost:RCTs can be resource-intensive, requiring careful planning, monitoring, and funding.

 

Ethical Considerations: Randomly assigning participants means some may not receive an intervention that could benefit them, which can be ethically challenging.

 

Limited External Validity: Results may not always apply to other contexts or populations, especially if the study area is very specific.

 

Logistical Challenges: Ensuring randomisation, consistent implementation, and robust data collection can be difficult in the real world.

 

Key Considerations When Planning an RCT

 

To successfully run an RCT, careful planning and transparency are essential. Here are some key steps:

 

Are you sure?

 

  • RCTs are not for the ‘faint-hearted’. They take lots of time, effort, planning, sometimes lots of money and always a part of your soul.

 

 

Define Your Question and Hypotheses Clearly:

 

  • Your question should focus on whether a specific intervention causes a specific outcome. 

  • It should not be ‘pie in the sky’ – there should be some theory behind why your intervention causes the outcome.

  • Your theory should be framed as a hypothesis. You might be as specific as stating to what sort of difference there will be in the outcome.

 

Create a Trial Protocol:

 

  • A protocol is a detailed plan created before anything else happens that explains:

 

  1. The research question and hypotheses.

  2. How participants will be randomly assigned.

  3. What the intervention and control groups will do.

  4. How data will be collected and analysed.

 

  • The protocol ensures the trial is consistent, transparent, and ethically sound.

 

  • To do an RCT by the letter of the law, you should not deviate from the protocol. It’s normal to need to though because what works on paper doesn’t always work in practice. 

 

  • Consider running a pilot or rehearsal of your trial to identify potential problems before you start proper.

​

RCT Checklist

 

  1. Ensure Randomisation:

    • Use a method like a random number generator to assign participants to groups, ensuring there’s no bias in the selection process.

  2. Ethical Approval:

    • Submit your trial protocol to an ethics board to ensure participants’ rights and well-being are protected.

  3. Sample Size Matters:

    • Ensure your trial includes enough participants to detect meaningful differences between groups. Too few participants may lead to inconclusive results.

  4. Monitor Implementation:

    • Make sure the intervention is delivered as planned. If officers in the intervention group don’t use the body-worn cameras consistently, for example, the trial’s results may be unreliable.

  5. Analyse Data Transparently:

    • Stick to the analysis plan set out in your protocol. Avoid changing your methods midway to achieve more favourable results—this undermines the credibility of the trial.

 

Final Thoughts

​

RCTs are a powerful tool for understanding what works, but they require careful design, resources, and ethical consideration. They’re not suitable for every situation, but when done well, they provide some of the strongest evidence available for cause-and-effect relationships.

 

If you’re considering an RCT, remember:

 

  • Plan meticulously.

  • Be transparent about your methods.

  • Embrace the results, even if they don’t show the intervention worked—this is still valuable learning for evidence-based policing.

Quasi-experiments

Quasi-experiments are a research method used to evaluate the effectiveness of interventions when randomisation—the hallmark of randomised controlled trials (RCTs)—is not feasible. While they don’t offer the same level of control as RCTs, quasi-experiments still provide valuable insights, especially in real-world settings where randomisation may be impractical or unethical.

 

How Are Quasi-Experiments Different from RCTs?

​

The key difference between quasi-experiments and RCTs is randomisation:

​

  • In RCTs, participants are randomly assigned to intervention and control groups.

  • In quasi-experiments, groups are pre-existing (e.g., different police districts, time periods) or assigned based on non-random criteria (e.g., who can access the intervention).

​​

This lack of randomisation means that quasi-experiments can be more vulnerable to bias, as differences between groups may influence the results rather than the intervention itself.

 

Types of Quasi-Experiments

​

There are several common types of quasi-experiments, each suited to different research questions:

​

Before-and-After Studies:

  • These compare outcomes before and after an intervention in the same group.

  • Example: Crime rates in a neighbourhood before and after installing CCTV cameras.

  • Limitation: It’s hard to rule out other factors (e.g., seasonal trends, unrelated changes) that might explain the results.

 

Interrupted Time Series:

  • Measures outcomes at multiple time points before and after an intervention to identify patterns.

  • Example: Examining monthly crime rates over several years to see if a policy change had a long-term effect.

  • Strength: Helps distinguish the intervention’s impact from natural fluctuations.

  • Limitation: Requires consistent, high-quality data over time.

 

Non-Equivalent Control Group Designs:

  • Compares outcomes between an intervention group and a similar but non-randomised control group.

  • Example: Comparing two neighbourhoods, one with increased patrols and one without.

  • Strength: Provides a comparison to strengthen causal claims.

  • Limitation: Differences between the groups may bias the results.

 

Regression Discontinuity:

  • Assigns participants to groups based on a threshold (e.g., age, score) and compares those just above and below the cutoff.

  • Example: Evaluating a programme offered only to offenders under 18 by comparing outcomes for those just younger and older than 18.

  • Strength: Mimics randomisation in certain situations.

  • Limitation: Only works well when the cutoff is strictly applied.

 

What Are Quasi-Experiments Used For?

​

Quasi-experiments are particularly useful for:

​

  • Evaluating Real-World Interventions: When randomisation is impractical, such as testing the effect of new policies or large-scale programmes.

  • Examining Complex Systems: In contexts like policing, where interventions can’t always be controlled tightly.

  • Exploring Ethical Constraints: When withholding an intervention from certain groups (as in RCTs) would be unfair or unethical.

 

Strengths of Quasi-Experiments

​

  1. Practicality:

    • They allow research in real-world settings without the logistical challenges of randomisation.

  2. Flexibility:

    • Can evaluate interventions already in place or those implemented on a large scale.

  3. Ethical Viability:

    • Suitable when withholding an intervention for randomisation would be unethical.

  4. Cost-Effectiveness:

    • Often less expensive than RCTs because they work with existing groups and systems.

 

Limitations of Quasi-Experiments

​

  1. Lower Internal Validity:

    • Without randomisation, it’s harder to rule out other factors (confounders) that might explain the results.

  2. Selection Bias:

    • Pre-existing differences between groups can distort findings.

  3. Data Demands:

    • Interrupted time series and similar designs require high-quality, consistent data over time.

  4. Complex Analysis:

    • Adjusting for confounders often requires advanced statistical techniques.

 

Key Considerations When Planning a Quasi-Experiment

​

  1. Define the Research Question Clearly:

    • Focus on causal questions: “Does X intervention lead to Y outcome?”

  2. Choose the Right Design:

    • Select the design that best matches your context and available data.

  3. Identify and Account for Confounders:

    • Confounders are variables that could influence the outcome. For example, if you’re evaluating CCTV’s impact on crime, population density could also affect crime rates.

  4. Plan for High-Quality Data:

    • Ensure consistent data collection and consider how missing data will be handled.

  5. Be Transparent About Limitations:

    • Clearly state what your design can and cannot show. For example, explain that differences between groups might be due to pre-existing factors rather than the intervention.

 

Final Thoughts

Quasi-experiments are a valuable alternative to RCTs when randomisation isn’t possible. While they come with challenges, careful design and transparent reporting can provide meaningful insights into the effectiveness of policing interventions. They are particularly well-suited to the real-world complexities of policing, where perfect conditions for research are rare.

Observational studies

Observational studies are a research method used to understand patterns, relationships, and trends by observing the world as it is, without manipulating variables or introducing interventions. They are particularly useful in policing, where practical or ethical constraints may make experimental methods unfeasible.

 

What Are Observational Studies?

​

Unlike randomised controlled trials (RCTs) or quasi-experiments, observational studies do not involve assigning participants or groups to receive an intervention. Instead, they observe and analyse what happens naturally.

​

For example:

  • A study might examine crime rates in areas with different levels of police visibility without actively changing or controlling patrol patterns.

​​

Observational studies focus on correlation (how things are related) rather than causation (whether one thing causes another). While they can’t definitively prove cause and effect, they provide valuable insights into real-world phenomena.

 

Types of Observational Studies

​

There are several types of observational studies, each suited to different kinds of research questions:

​

Cross-Sectional Studies:

  • Collect data at a single point in time to understand the current state of a problem.

  • Example: Measuring public confidence in policing through a one-off survey.

  • Strength: Quick and cost-effective.

  • Limitation: Can’t show how things change over time or what caused the observed relationships.

 

Longitudinal Studies:

  • Collect data over time to track changes and trends.

  • Example: Analysing crime rates in a neighbourhood over five years.

  • Strength: Helps identify patterns and trends over time.

  • Limitation: Requires consistent data collection, which can be resource-intensive.

 

Case Studies:

  • Focus on an in-depth analysis of a single event, individual, or location.

  • Example: Studying the impact of a high-profile incident on community trust in policing.

  • Strength: Provides rich, detailed insights.

  • Limitation: Results are often not generalisable.

 

Ecological Studies:

  • Look at data aggregated by groups, such as neighbourhoods or cities, rather than individuals.

  • Example: Comparing crime rates between cities with and without community policing initiatives.

  • Strength: Useful for studying large-scale patterns.

  • Limitation: Can miss individual-level factors that contribute to the results.

 

What Are Observational Studies Used For?

​

Observational studies are ideal for:

​

  • Understanding Context: Exploring how factors like socioeconomic conditions or public trust influence crime rates.

  • Generating Hypotheses: Identifying patterns or correlations to test further in experimental studies.

  • Monitoring Trends: Tracking changes over time, such as the effect of policy shifts on crime reporting.

  • Ethically Sensitive Areas: Studying topics where interventions might not be appropriate, such as the relationship between officer demographics and use-of-force incidents.

 

Strengths of Observational Studies

​

  1. Real-World Relevance:

    • Observational studies provide insights into natural settings without altering conditions.

  2. Broad Applicability:

    • They can be applied to diverse topics, from crime trends to public attitudes.

  3. Ethical Viability:

    • Useful for studying sensitive topics where intervention could be harmful or impractical.

  4. Cost-Effective:

    • Often less expensive than experimental methods, as they rely on existing data or routine observations.

 

Limitations of Observational Studies

​

  1. Cannot Prove Causation:

    • Observational studies can only show correlations, not whether one thing causes another.

    • Example: If areas with high officer visibility have lower crime rates, it’s unclear whether visibility reduces crime or if officers are more visible in already low-crime areas.

  2. Confounding Variables:

    • Other factors (confounders) might influence the results.

    • Example: A study might link higher education levels to lower crime rates, but income or social services could also play a role.

  3. Bias Risk:

    • Selection bias (choosing non-representative samples) or reporting bias (incomplete data) can affect results.

 

Key Considerations When Planning an Observational Study

​

  1. Define Your Question:

    • Observational studies are best suited to descriptive or exploratory questions, such as “What are the trends in public complaints against police over time?”

  2. Choose the Right Design:

    • Match the type of observational study (e.g., cross-sectional, longitudinal) to your research needs.

  3. Account for Confounders:

    • Use statistical techniques or careful planning to minimise the influence of other variables.

  4. Ensure Data Quality:

    • Observational studies rely heavily on existing data, so ensure it is accurate, consistent, and relevant.

  5. Be Transparent About Limitations:

    • Clearly state that the study identifies correlations, not causation, and discuss potential confounding factors.

 

Final Thoughts

​

Observational studies are a vital tool for evidence-based policing, providing valuable insights when experiments aren’t feasible. While they can’t prove cause and effect, they help identify patterns, generate hypotheses, and deepen our understanding of complex issues. Careful design and transparency about limitations ensure these studies contribute meaningfully to the evidence base.

Exploratory studies

Exploratory studies are used to investigate new or poorly understood problems where there is little existing knowledge. They are designed to ask open-ended questions, generate ideas, and provide insights that can guide further research. While they don’t aim to test specific hypotheses, exploratory studies play a crucial role in shaping the evidence base and defining future directions.

 

What Are Exploratory Studies?

​

Exploratory studies are flexible and open-ended by nature. They are not focused on testing cause-and-effect relationships or producing definitive answers but instead aim to:

  • Identify Patterns: Highlight trends or issues that need closer investigation.

  • Generate Hypotheses: Provide ideas for more targeted research.

  • Understand Context: Gather detailed insights into a complex problem.

​​

Example: If a police force notices rising levels of youth violence but doesn’t know why, an exploratory study might involve interviews, focus groups, or reviewing available data to identify potential causes and inform future interventions.

 

Types of Exploratory Studies

​

Exploratory studies can take many forms, depending on the nature of the problem:

​

  1. Qualitative Methods:

    • Focus on depth and detail, such as interviews, focus groups, or ethnographic observation.

    • Example: Observing youth interactions in high-crime areas to understand risk factors for violence.

  2. Case Studies:

    • Provide in-depth analysis of a single case or incident to uncover broader insights.

    • Example: Analysing a community’s response to a controversial policing policy.

  3. Preliminary Data Analysis:

    • Use existing data to identify patterns or gaps in knowledge.

    • Example: Examining complaint records to explore trends in public dissatisfaction with policing.

 

What Are Exploratory Studies Used For?

​

Exploratory studies are particularly useful for:

​

  • Defining Problems: When the scope or cause of an issue isn’t well understood, exploratory studies can clarify what’s happening.

  • Developing Research Questions: They provide the foundation for more structured studies, like experiments or longitudinal research.

  • Understanding Context: They help capture the complexity of social issues, such as community-police relationships.

 

Strengths of Exploratory Studies

​

  1. Flexibility:

    • Exploratory studies adapt to evolving questions and discoveries during the research process.

  2. Rich Insights:

    • They provide deep, detailed understanding, especially when using qualitative methods.

  3. Foundation for Future Research:

    • They generate ideas and hypotheses for more targeted studies.

  4. Real-World Relevance:

    • By focusing on context and detail, exploratory studies provide insights that directly inform practice.

 

Limitations of Exploratory Studies

​

  1. Lack of Definitive Answers:

    • These studies don’t test hypotheses or establish causation, so their findings are often more suggestive than conclusive.

  2. Potential Bias:

    • Qualitative methods, in particular, rely on subjective interpretation, which can introduce bias.

  3. Limited Generalisability:

    • Results may apply only to the specific context studied and may not be widely applicable.

  4. Resource Intensity:

    • Collecting and analysing rich, detailed data can be time-consuming and resource-heavy.

 

Key Considerations When Planning an Exploratory Study

​

  1. Define Your Broad Objective:

    • What do you want to learn or understand? Keep the scope open but focused enough to guide your approach.

  2. Choose the Right Methods:

    • Select methods that suit the complexity of the problem, such as interviews for understanding attitudes or data analysis for identifying trends.

  3. Be Transparent About Limitations:

    • Clearly state that the findings are exploratory and may not provide definitive answers or generalisable results.

  4. Consider Ethical Implications:

    • Ensure that the study respects participants’ privacy, especially when using methods like interviews or observation.

  5. Prepare for Further Research:

    • Use the findings to shape more targeted, hypothesis-driven studies that can test specific interventions or solutions.

 

An Example of an Exploratory Study

​

Imagine a police force notices increasing community complaints about officer conduct but lacks specifics about what’s driving dissatisfaction:

  • Approach: Conduct interviews with community members and officers, review complaint data, and hold focus groups to explore perceptions.

  • Outcome: Identify recurring issues, such as concerns about officer visibility or communication style, and develop hypotheses to test in future studies.

 

Final Thoughts

​

Exploratory studies are a vital tool in the early stages of evidence-based policing. They help practitioners and researchers understand complex problems, identify new areas of focus, and lay the groundwork for more structured research. While they may not provide all the answers, they play a crucial role in shaping meaningful and actionable evidence.

Documentation

Proper documentation is essential for any type of study, including exploratory research. Good records ensure transparency, reproducibility, and credibility. They also make it easier to share findings and build on the research in the future. For exploratory studies, where flexibility is a strength, documentation helps track decisions and maintain rigour.

 

What to Document

​

Exploratory studies involve varied and flexible methods, but certain key elements should always be recorded:

​

  1. Study Purpose and Objectives:

    • Clearly define the broad aims of the study and what you hope to learn.

    • Example: “Understand public perceptions of police presence in high-crime neighbourhoods.”

  2. Research Questions:

    • Even exploratory studies should have guiding questions to keep the research focused.

    • Example: “What concerns do community members have about police interactions?”

  3. Methodology:

    • Document the methods used, including:

      • Data collection techniques (e.g., interviews, focus groups, observations).

      • Sampling strategy (e.g., who was included and why).

      • Tools used (e.g., survey templates, interview guides).

  4. Participant Information:

    • Record details about participants while ensuring anonymity and confidentiality.

    • Example: Demographics, geographic location, or other relevant characteristics.

  5. Data Collected:

    • Keep accurate and secure records of the raw data, including:

      • Transcripts from interviews or focus groups.

      • Field notes from observations.

      • Any quantitative data (e.g., survey results).

  6. Decisions and Changes:

    • Exploratory studies often evolve as new insights emerge. Document:

      • Changes to research questions or methods.

      • Reasons for those changes (e.g., unexpected findings, logistical challenges).

  7. Analysis and Interpretation:

    • Record how the data was analysed, including any coding frameworks or statistical techniques used.

    • Example: “Interview responses were thematically analysed using NVivo software.”

  8. Findings and Insights:

    • Summarise key findings in clear, accessible language.

    • Note any patterns, themes, or surprises that emerged.

  9. Limitations:

    • Acknowledge the study’s constraints, such as:

      • Limited sample size.

      • Potential biases in data collection or interpretation.

  10. Ethical Considerations:

    • Document consent forms, ethics approvals, and how participant rights were protected.

 

Templates and Standard Formats

​

Using templates can ensure consistency and thoroughness in your documentation. Some widely used tools and formats include:

​

  1. Research Protocol Templates:

    • These can be adapted for exploratory studies and include sections for objectives, methodology, and ethical considerations.

    • Example: Templates such as the Crim-PORT (Criminological Protocol for Randomised Trials) and CONSORT statement checklist (give it a Google) are worth your time.

  2. Field Note Templates:

    • Structured formats for recording observations, including date, location, and key observations.

  3. Interview and Focus Group Guides:

    • Pre-prepared question lists or prompts to guide discussions.

  4. Analysis Logs:

    • Spreadsheets or software logs to track how data was coded, categorised, or summarised. If you get in to publishing your research, open criminology practices really value publication of data alongside reports (data protection allowing).

  5. Report Templates:

    • Formats for presenting findings, such as those used by policing bodies or academic institutions.

 

Why Documentation Matters

​

Good documentation:

​

  • Ensures Transparency: Others can understand how you conducted the study and trust your findings.

  • Supports Reproducibility: Future researchers can replicate or build on your work.

  • Helps with Reflection: Detailed records allow you to evaluate what worked well and what didn’t.

  • Facilitates Sharing: Comprehensive documentation makes it easier to communicate findings to stakeholders.

 

Final Thoughts

​

Even in exploratory studies, where flexibility is key, rigorous documentation is essential. By keeping clear records of your process, decisions, and findings, you ensure that your research is credible, transparent, and valuable for evidence-based policing.

 BETA   Contact us to help improve this site
bottom of page