Recent school science curriculum developments in many countries emphasise that scientists derive evidence for their claims through different approaches; that such practices are bound up with disciplinary knowledge; and that the quality of data should be appreciated. This position paper presents an understanding of the validity of data as a set of conceptual relationships, illustrating the application of the network of ideas and their inter-relationships necessary for the "thinking behind the doing" with examples from practice. We explore ways in which this understanding of data is inherently related to underpinning disciplinary ideas. We suggest how the recognition of a conceptual basis for understanding the quality of data represents an ontological shift with respect to widespread characterisations of scientific practices which addresses some long-standing issues in science education research, policy, curricula and practice.