Purpose: This paper is second of seven in this volume elaborating different approaches to quality improvement in education. It delineates a methodology called design-based implementation research (DBIR). The approach used in this paper is aimed at iteratively improving the quality of classroom teaching and learning practices in defined problem areas through collaborations among researchers, practitioners and other education stakeholders. Design/methodology/approach: The paper describes the origins of the approach in US education, along with its foundations, core principles and a case application of DBIR in practice. The case focuses on the specific problem of teaching science and genetics in primary and secondary schools in a district. Findings: The guiding principles of DBIR are: a focus on persistent problems of classroom educational practice; iterative and collaborative design and testing of innovations through partnerships between researchers and practitioners, involving multiple stakeholders' perspectives; a concern with developing theory related to both implementation processes and classroom learning outcomes, using systematic inquiry; and development of the capacity of both researchers and practitioners to sustain changes in educational systems. Originality/value: Few theoretical treatments and demonstration cases are currently available in US education that examine common models of quality improvement, particularly DBIR. By engaging practitioners with researchers in designing, testing and implementing reforms meaningfully, DBIR shows promise in offering significant on-the-ground benefits. This paper adds value by allowing readers to compare the DBIR method with the other improvement approaches explicated in this volume.