This paper critically reviews the opportunities and challenges in designing and conducting actionable research on the learning and development of children in conflict- and crisis-affected countries. We approached our review through two perspectives championed by Edward Zigler: (a) child development and social policy and (b) developmental psychopathology in context. The aim of the work was to answer the following questions: What works to enhance children's learning and development in such contexts? By what mechanisms? For whom? Under what conditions? How do experiences and conditions of crisis affect the basic processes of children's typical development? The review is based on a research–practice partnership started in the Democratic Republic of the Congo in 2010 and expanded to research in Niger and Lebanon in 2016. The focus of the research is on the impact of Healing Classrooms (a set of classroom practices) and Healing Classrooms Plus (an additional set of targeted social and emotional learning activities), developed by the International Rescue Committee, on children's academic outcomes and social and emotional learning. We sought to extract lessons from this decade of research for building a global developmental science for action. Special attention is paid to the importance of research–practice partnerships, conceptual frameworks, measurement and methodology. We conclude by highlighting several essential features of a global developmental science for action.

With Sustainable Development Goal 4, the global education community promises to improve equitable access to quality education for all children. This promise can only be upheld through investments by multiple stakeholders in the most educationally marginalized children: the 75 million children who are currently out of school—and the millions more who are in school but not learning—in crisis contexts. For researchers working in partnership with practitioners and governments, investments are needed to develop and adapt measurement tools that provide accurate and timely data about critical dimensions of program implementation (PI) and children’s holistic learning and development (CHILD) in crisis contexts. Such tools can be used in both research efforts and embedded in practitioner monitoring and evaluation systems, resulting in data that can: a) improve the ability of researchers to detect impacts and interpret results of randomized trials of education programs; and b) guide practitioners in data-driven quality improvement efforts. In Paper 1 of this dissertation, I apply theories and evidence from the prevention and intervention, developmental, and implementation sciences to outline how research-practice-policy partnerships can make strategic and coordinated progress towards developing and adapting CHILD and PI measurement tools. In Papers 2 and 3, I provide empirical evidence on the psychometric properties of two such measurement tools: one developed in Western countries and used widely in humanitarian contexts to assess a key dimension of CHILD (social-emotional well-being) and one developed in a crisis context to assess a key dimension of PI (classroom implementation quality). Specifically, in Paper 2 I provide some of the first evidence of the reliability, validity, and cross-contextual comparability of the teacher-report version of the Strengths and Difficulties Questionnaire (Goodman, 1997) within and across three humanitarian contexts. In Paper 3, I provide initial evidence of the content and predictive validity of an observation tool developed by staff at the International Rescue Committee (IRC) to assess the quality of classroom implementation of Learning in a Healing Classroom, the IRC’s teacher professional development and curricular program. Implications of the findings of each of the studies for research-practice-policy partnerships are discussed within each paper and cumulatively.