What counts as ‘evidence’?

This policy brief argues for the adoption of evidence-based policing in South Africa, defining 'evidence' as knowledge derived from the scientific method rather than the law of evidence or forensic science. It advocates for a collaborative relationship between law enforcement and academic researchers to replace traditional reactive policing with a cycle of targeting, testing, and tracking.

• Evidence-based policing is defined as the systematic application of empirical research to police decision-making. In this specific context, 'evidence' does not refer to forensic science or the law of evidence, but to knowledge derived from the scientific method, which involves the "systematic accumulation of ideas that have been empirically tested and haven’t yet been proven wrong."
• The document proposes replacing the traditional policing model—characterized by the 'three Rs' of random patrol, rapid response, and reactive investigations—with the 'three Ts': targeting, testing, and tracking. Targeting focuses on specific problems or places, testing involves empirical experimentation via the scientific method, and tracking involves documenting outcomes.
• Evidence exists on a spectrum of rigour determined by research design. The 'gold standard' is the randomised control trial, which is the most effective way to eliminate alternative explanations. Other designs include meta-analysis/systematic reviews at the top, followed by non-randomised trials, correlational research, and qualitative studies, with unsystematic observations at the bottom.
• Successful integration of evidence-based practice requires a mutual relationship and a shared vocabulary between 'policing professionals' (law enforcement officials) and 'knowledge professionals' (academics or researchers). This collaboration can produce 'pracademics' or 'evidence cops' who ensure practice is based on the best available knowledge.
• The EMMIE evidence rating framework is highlighted as a tool for decision-making. It evaluates policing practices based on five metrics: Effect (impact on crime), Mechanism (how it works), Moderators (where it works), Implementation (how to do it), and Economic cost (cost effectiveness).
• The author concludes that South Africa should prioritise activities that reduce crime and build trust in police to maximise limited resources. This requires an organisational culture of critical thinking, curiosity, and a willingness to be proven wrong, moving away from decisions based on assumption.

APA

Kriegler, A. (2019). What counts as ‘evidence’? Institute for Security Studies. https://issafrica.s3.amazonaws.com/site/uploads/pb134.pdf

Chicago

Kriegler, Anine. What counts as ‘evidence’? Institute for Security Studies, 2019. https://issafrica.s3.amazonaws.com/site/uploads/pb134.pdf.

Wikipedia

{{cite report |last1=Kriegler |first1=Anine |title=What counts as ‘evidence’? |publisher=Institute for Security Studies |date=2019 |url=https://issafrica.s3.amazonaws.com/site/uploads/pb134.pdf |access-date=17 August 2026 |via=Climate Insights Directory}}

BibTeX

@techreport{kriegler2019what, author = {Kriegler, Anine}, title = {{What counts as ‘evidence’?}}, institution = {Institute for Security Studies}, year = {2019}, url = {https://issafrica.s3.amazonaws.com/site/uploads/pb134.pdf}, urldate = {2026-08-17}, note = {Indexed by Climate Insights Directory} }