ABSTRACT
There is a consensus in the literature on narrative research that it should not be judged by the same criteria as those that are applied to more traditional and broadly accepted qualitative and quantitative research methods. (See, for instance, Polkinghorne (1988), Riessman (1993), Huberman (1995), Amsterdam and Bruner (2000) and Geelan (2003).) Traditional approaches to reVHDUFKWHQGWREHEDVHGRQVFLHQWL¿FPHWKRGVIDFWVDQGSURFHVVHV1DUUDWLYH LQTXLU\DQGVWRU\WHOOLQJUHVHDUFKDVGH¿QHGLQRWKHUFKDSWHUVLQWKLVERRN seeks to elaborate and investigate individual interpretations and worldviews of complex and human-centred events. It is more concerned with individual truths than identifying generalisable and repeatable events. We see narrative DVDXQLTXHKROLVWLFUHVHDUFKPHWKRGWKDWUHGH¿QHVPHDVXUHVWKDWDUHXVHG WRUDWHDQGFRPSDUHUHVHDUFK7KHUHIRUHDWWKHYHU\OHDVWWKHGH¿QLWLRQVRI reliability and validity, commonly used in traditional research, require a reWKLQNLQJDQGUHGH¿QLQJIRUQDUUDWLYHUHVHDUFK7KLVFKDSWHUVHHNVWRGRWKDW
To state matters succinctly, reliability in narrative research usually refers to the dependability of the data, while validity typically refers to the strength of the analysis of data, the trustworthiness of the data and ease of access to that data (Polkinghorne, 1988). As noted by Riessman (1993), concepts of YHUL¿FDWLRQDQGSURFHGXUHVIRUHVWDEOLVKLQJYDOLGLW\IURPWKHH[SHULPHQWDO model) rely on measurable and objectivist assumptions that are largely irrelevant to narrative studies. A personal narrative is not meant to be read as an exact record of what happened, nor is it a mirror of the world ‘out there’. This can be, of course, an upsetting view to those researchers who may be troubled by anything that does not neatly conform to the divides of traditional quantitative and qualitative views of research methods.
