Evaluation Guidelines

This document offers practical guidance on how to apply Machine Learning (ML) analytical approaches in evaluations. It presents an operational framework that was developed after a thorough literature review and UNICEF Evaluation Office’s (EO) experiments with application of ML methods. Additionally, it discusses potential use cases of these methods, along with associated risks and limitations.

This guidance note was developed to guide the organization on the progress toward achieving the UNDIS Indicator 10 standards. The note presents tailored UNICEF considerations, good practices, and specific key considerations for action for disability-inclusive evaluations.

This guidance note is intended for UNICEF headquarters, regional, and country office staff for planning, managing, and participating in a Real-Time Evaluation Plus of UNICEF’s Humanitarian Responses.

This guidance is intended as a “how-to” for integrating a gender lens in UNICEF evaluations.

The objective of this guidance note is to introduce the concept of evaluability by providing a set of concise, practical, and user-friendly tools to support UNICEF, HQ, Regional, and Country Offices in the conduct of Evaluability Assessments.

This guidance note was created in response to that challenge, to assist UNICEF staff in determining what form of Adolescent Participation in Monitoring and Evaluation can best add value; how adolescents can be meaningfully, safely, and ethically involved in different contexts; what practical issues should be considered when designing and implementing participatory M&E activities; and what tools and resources can be used to support such a process.