This paper argues that there is merit in the study of students' identity constructions as a means of understanding inequalities in schooling outcomes. However, the effectiveness of such accounts remains limited by the hiatus between the identity theories deployed in research studies and the empirical data used to illuminate such theories. The methodology by which interview protocols are used in relation to identity theories is tacit or unclear. This paper aims to address this issue by developing a methodology of identity construction for students in schools that makes explicit the links between theory, methods and data. In applying this methodology to an analysis of one longitudinal case study example, its utility is illustrated.