ELSEVIER zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Agriculrural Sy srems. Vol. 57, No. 4, pp. M--556. 1998 ‘c 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain PII: SO308-521X(98)00008-0 0308-521X/98 $19.00+0.00 Modelling Farmer Behaviour: a Personal Construct Theory Interpretation of Hierarchical Decision Models Roy Murray-Prior Muresk Institute of Agriculture, Curtin University of Technology, Northam, WA 6401, Australia. (Received 24 June 1997; accepted 30 January 1998) ABSTRACT Hierarchical decision models based on a two-stage decision process (Gladwin, 1977, 1989) have proved useful for describing and predicting individual decisions with a minimum of theoretical bias. The model has two weaknesses. First, it does not incorporate an explanation of the underlying motivation for behaviour. Second, the empirical techniques for building the decision models are based on ethnographic approaches that are time consuming and d@cult to implement when multiple visits are a problem. In this article it is shown personal construct psychology (Kelly, 19.5.5), which assumes people behave as ‘scientists’, can overcome these weaknesses. The combination provides a theory and an empirical model of behaviour that explains the motivation and reasons for behaviour, allows for learning, and describes and predicts individual decisions. In addition, it has asso- ciated interviewing techniques that complement the ethnographic techniques and overcome some of their limitations. 0 1998 Elsevier Science Ltd. All rights reserved INTRODUCTION A hierarchical decision model proposed by Gladwin (1989) involves an assumption of a two-stage decision process. Another key assumption of the hierarchical decision model is that each alternative consists of a set of aspects (Gladwin, 1980). Decisions are decomposed into the sequential comparison of the various alternatives using a number of characteristics or aspects. When tested on groups of people, models based on this framework have 541