The repertory grid: a critical appraisal

George Kelly claims a scientific status for Personal Construct Theory, but I argue that it is more appropriately characterised among the interpretive human sciences. Examination of the theory from the point of view of the grid user discloses a number of weaknesses, the most directly relevant being Kelly's assumption of the dichotomous nature of constructs. Even when this assumption is weakened by allowing grading between oppositional poles, the grid matrix retains a positivism that appears at variance with the main thrust of Kelly's theorising. The central chapters appraise technical aspects of grid methodology, dealing sequentially with elements, constructs, bipolarity, the completion of a grid matrix, analysis, and the stability of grid data. Analysis of underlying assumptions, reflection upon the 'grid literature', and some empirical studies indicate that grid methodology is often flawed in both conceptualisation and practice. While some improvements may be made regarding technique, element X construct interactions radically undermine the grid as a research instrument, as does Kelly's later claim for the importance of events. I further argue that short verbal labels are inadequate to bear the load of meaning that respondents wish to convey, and that grid methodology excludes the richness of figurative language: developments based upon fuzzy set theory are unlikely to improve matters. If, as I suggest, communication of meaning is a prime requirement of construct theory research, then alternative approaches to the elicitation of constructs are necessary. I conclude by sketching a possible response to the criticisms that have been advanced, and argue for a 'personal construct hermeneutics' in which theory and method are brought into a closer alignment. Indications are given of how this might be operationalised in terms of 'accounts methodology' and of some of the implications for the conduct of research in the human sciences