Policy Library
Policy evidence from official statistics, legislation, consultations, government, NGO, and intergovernmental sources.
Learning or accountability: what should policy evaluation prioritize?. Break down the recurring debate over whether evaluation should mainly support learning or accountability. Show why the practical answer is usually a balanced design that protects independence while keeping findings usable for improvement.
Policy evaluation keeps running into the same fight: should it mainly prove performance for accountability, or help people learn and improve policy? WHO and OECD say that is a false choice; the goal is both[[cite:1]][[cite:2]]. WHO says evaluation should be conducted as systematically and impartiall...
ViewFive tests for whether government evaluation systems are usable. Create five cards covering institutional mandates, evaluation quality, impact on decisions, data access, and skills for evidence use. Include the 79 percent central evaluation guidelines finding as one concrete anchor where appropriate.
Evaluation systems are useful but uneven across government. Clear institutional mandates are needed so evaluations are conducted and used systematically across government. On quality, the OECD says 79% have adopted central evaluation guidelines, but that passage does not prove those guidelines speci...
ViewThe hidden side effects of intergovernmental grants. Summarize the main design risks: fly-paper effects, crowding-out of own taxes, procyclicality, accountability problems, and overloading a single grant instrument. Close with the practical lesson that grant design should match objectives rather than pursue every goal at once.
Intergovernmental grants can do real good, but the OECD warns they also carry hidden side effects. One common risk is the fly paper effect, where recipients increase spending by more than they would after an equal rise in local income. Grants can also crowd out own source taxes, because local govern...
ViewFive data standards that shape policy evidence. Create five cards that show how apparently technical data choices determine what policy questions can be answered. Each card should pair one data infrastructure concept with its practical consequence for researchers or decision makers.
Representative sampling turns displacement data into population evidence, not just project monitoring. Core EGRISS identification questions make refugees and IDPs classifiable and comparable across countries. Rich questionnaire metadata lets analysts map evidence gaps by theme, such as labor, financ...
ViewHow can parliaments make evidence-based lawmaking routine?. Lay out the governance loop from regulatory impact assessment and government-provided evidence to committee scrutiny, consultation transparency, and post-legislative review. Use the Western Balkans findings as the institutional case and connect them to the broader OECD lesson on fit-for-purpose capacity under complexity.
Can parliaments make evidence-based lawmaking routine, not exceptional? OECD says yes, but only if the whole loop works: evidence in drafting, timely access for scrutiny, and follow-up after laws pass[[cite:1]]. In the Western Balkans, that loop is often broken. Not all preparation evidence, includi...
ViewWhen should governments use randomized trials in policy design?. Build a scenario-based knowledge check on randomized trials, counterfactuals, quasi-experimental methods, cost-effectiveness, feasibility, ethics, data access, and skills. Keep the emphasis on choosing the right evaluation method rather than promoting trials as universally superior.
Q1. A ministry wants to know which of two housing support models is most cost-effective for improving the same outcome. What is the best evaluation choice? - Run a randomized trial only and ignore costs - Use an economic evaluation that combines a robust impact estimate with programme costs - Choose...
View- 1(current)