This monograph critically examines the emerging discourse on quality associated with Education for All (EFA). It contends that EFA discourse has moved from a welcome and vital commitment to quality to its measurement without adequate consideration of what "quality" entails, particularly in the vital domain of pedagogy. Meanwhile, the demand for quality indicators by governments and international agencies has left important methodological questions unanswered. Citing various international examples, the paper notes a concern with input and context at the expense of the proper elucidation of educational process and outcome, arbitrariness in what is focused upon, excessive use of proxies, overly selective use of international research on learning and teaching, and confusions about the key terms "quality", "indicators" and "measures". The paper proposes criteria for assessing frameworks for evaluating the quality of classroom provision, central to which are evidential breadth, conceptual comprehensiveness, validity, reliability, impact, manageability and appositeness to level and context of use. Arguing that at root the problem is as much conceptual as empirical and procedural, the paper proposes a map of the territory of pedagogy at the levels of ideas and action, together with principles of procedure to guide future work on indicators and measures of quality in the EFA context. Originally prepared for use in India, the paper uses examples from that country by way of illustration. (Contains 2 figures and 4 footnotes.)