How we provide evidence to deliver results for children
The evaluation policy and practice that foster UNICEF’s evidence-based decision-making.
The context in which UNICEF operates is complex and fast-changing. This is why we continuously innovate our evaluation methods, complementing traditional approaches with those that deliver timely, relevant and actionable recommendations.
We advance the value of our work by reaching the right stakeholders at the right moment with the right evidence. This is how we amplify our findings and recommendations to ensure UNICEF continues to help children survive, thrive and fulfil their potential.
UNICEF's evaluation policy outlines evaluation principles and procedures, and sets out key accountabilities for the function and use of evaluations.
Guided by the norms and standards of the United Nations Evaluation Group (UNEG) and international good practices – including for the evaluation of humanitarian assistance – the policy ensures that evaluations are useful, credible, independent and impartial, and that evaluation processes are transparent and child-centered.
UNICEF works in more than 190 countries and territories to reach the most disadvantaged children and young people. Each year, UNICEF evaluation staff manage dozens of evaluations around the globe.
Decentralization is key to our approach. We conduct evaluations at the country, regional and global levels. Because our evaluations must respond to shifting needs in complex settings, we tailor our approach to deliver timely, actionable findings that support fast-paced and context-specific decision-making.
To improve credibility and institutional learning, we developed the Global Evaluation Reports Oversight System (GEROS), which provides feedback to offices to ensure the evaluations they manage meet the highest standards.
Community Rapid Assessment (CRA) on COVID-19
COVID-19 Social and Behavioral Trends
UNICEF and Harvard Humanitarian Initiative CRA Dashboard
The Community Rapid Assessment (CRA) was a joint pilot led by the Evaluation Office and Social and Behavior Change Section. The initiative took place in two (2) regions – Eastern and Southern Africa, and Southern Asia – with 14 participating countries. In an effort to address the COVID-19 pandemic, the Evaluation Office conceptualized a remote data collection using high frequency mobile telephony data to survey populations and examine trust and behavior changes during the COVID-19 pandemic. The short modules also examined how households coped during the pandemic, as well as tracked their emerging needs and as the pandemic evolves. The sampling frames applies weight models to provide more reliable and representative data and was able to disaggregate the data by sex, age, urban/rural setting, and education. The primary objective of this work is to provide rapid, community-sourced data in real-time in order to strengthen the evidence base and inform country-level programming in response to the COVID-19 pandemic.
The findings also demonstrated the value of deploying rapid analytic systems for emergencies like COVID-19 and highlighted the importance behavioral change data to inform decision-making and program design.
This work is also done in collaboration with our external partners, the Harvard Humanitarian Initiative and Statistics Without Borders. Advanced inferential analysis were published in the Chance Journal, the American Statistical Journal with lessons learned from the exercise. The Harvard Humanitarian Initiative published real time indicators between each round of data collection for public awareness.
Visit for more details on the Community Rapid Assessment: Time Series Social Data for Assessing Behavioral Change During COVID-19