In this paper, using research in the field of inclusive education as our context, we examine problems with snippets. By this, we mean the selection of phrases, sentences and other units of narrative from our raw data that we extract and use to explain, exemplify, critique and generally take forward our own accounts. Much research in the field of inclusive education, as most educational research, assumes a qualitative approach, generally involving the collection/creation of data comprised of words. Our own research is no exception, as we have frequently explored in depth the perspectives of others - through an ethnographic perspective (Benjamin, 2002) and the use of narrative approaches (Lawson et al, 2006). Coffey and Atkinson (1996: 52) argue that to 'chop' the long and complicated accounts given by research participants into separate coded segments is to lose sight of 'fact that they are often couched in terms of stories - as narratives - or that they have other formal properties in terms of their discourse structures'. Narrative approaches, on the other hand, draw our attention to the representation of social reality as text, in which narratives are social constructions located in the discursive networks through which they are produced and of which they are productive.But it would be wrong to claim that, as researchers seeking to use narrative approaches, we are simply mediating the stories of others. To the contrary, the production of our research narratives involves us in making constant and complex decisions about how to chop our data into fragments that will enable us to produce and sustain our own (persuasive, we hope) storylines: after all, a coherent narrative is easier to read than a fragmented one, and the pressures to make up a consistent narrative out of competing ones are many (Czarniawska, 2004). As Armstrong (2003) notes, texts are still authored, and the representation of (differentiated and stratified) 'voices' does not remove the power of authorship.In this paper we offer an analytic examination of this issue of selectivity of/from data within the field of inclusive education. Using what criteria do we make the decisions to select one utterance in its entirety, to include edited highlights from another, and to exclude others altogether? These are decisions we make routinely, often 'intuitively' and seldom with an explicit rationale beyond the framing of our own narratives which may direct us towards selecting that which is exotic, out-of-the-ordinary or otherwise helpful in enabling us to tell a good story.We argue that this is a particular problem in the field of inclusive education, where we often find ourselves studying those who have been marginalised and Othered: instead of drawing attention to the power relations and practices through which particular voices tend to go unheard, might we instead be guilty of doing recognitional injustice (Fraser, 1997) to those whose interests we claim to represent?Armstrong, D. (2003) Experiences of Special Education: Re-evaluating policy and practice through life stories. London, RoutledgeFalmer Benjamin, S. (2002) The Micropolitics of Inclusive Education: an ethnography. Buckingham, Open University PressCzarniawska, B. (2004) Narratives in Social Science Research. London, Sage. Fraser. N. (1997) Justice Interruptus: Critical reflections on the 'postsocialist' condition. London, Routledge Lawson, H., Parker, M. and Sikes, P. (2006) Seeking stories: reflections on a narrative approach to researching understandings of inclusion European Journal of Special Needs Education 21, 1, pp. 55-68