Evidence-Based Policy

The Promise and Limits of Evidence

Evidence-based policy (EBP) is the principle that policy decisions should be informed by the best available research and data rather than ideology, anecdote, or tradition alone. The movement gained momentum in the late 1990s when the UK's Blair government made it a central governing philosophy, establishing the Cabinet Office's Strategy Unit to bring rigorous evidence into Whitehall decision-making.

The appeal is obvious: if we know what works, why would we do anything else? But evidence-based policy is more complicated in practice than in theory. Evidence is often incomplete, contested, or context-dependent. A program that worked in Finland may fail in Nigeria — not because the evidence was wrong but because the institutional, cultural, and economic context was different. The policy analyst's job is not to mechanically apply research findings but to interpret evidence wisely within a specific political and institutional context.

The Evidence Hierarchy

Not all evidence is equally reliable. Policy analysts use an evidence hierarchy to assess the strength of available research:

Tier 1: Systematic reviews and meta-analyses. These synthesize findings across multiple studies and are the gold standard. The Campbell Collaboration maintains a library of systematic reviews specifically designed for social policy questions.

Tier 2: Randomized controlled trials (RCTs). The Nobel Prize-winning work of Esther Duflo and Abhijit Banerjee at MIT's J-PAL demonstrated the power of RCTs in development policy. By randomly assigning participants to treatment and control groups, RCTs isolate causal effects with high confidence.

Tier 3: Quasi-experimental studies. When randomization is impossible or unethical, researchers use techniques like difference-in-differences, regression discontinuity, or instrumental variables to approximate causal inference.

Tier 4: Observational studies and case comparisons. Useful for generating hypotheses and understanding mechanisms, but vulnerable to confounding variables.

Tier 5: Expert opinion and stakeholder testimony. Valuable for filling gaps where formal research does not exist, but the weakest form of evidence for causal claims.

Institutionalizing Evidence in Government

Several countries have built permanent institutions to embed evidence in policymaking:

The What Works Network in the UK operates ten independent centers covering areas from education (the Education Endowment Foundation) to crime reduction (the College of Policing What Works Centre). Each center synthesizes research, rates evidence quality, and translates findings into practitioner guidance.

The US Commission on Evidence-Based Policymaking, established in 2016, recommended creating a National Secure Data Service and improving federal agencies' capacity to evaluate programs. The resulting Foundations for Evidence-Based Policymaking Act of 2018 required every federal agency to designate a Chief Evaluation Officer.

These institutions matter because individual analysts come and go, but institutional structures ensure that evidence remains part of the policy process regardless of who is in office. Without such structures, evidence use depends on the personal commitment of individual ministers or officials — a fragile foundation for systematic improvement.