ABSTRACT
In order to compare the Canadian and British systems, and because of the amount and complexity of the data, a path diagram was used. Use of a causal model has been suggested by Asker (1976) as a means of facilitating ‘the clearer statement of hypotheses and the generation of additional insights into the topic at hand’. A causal diagram ‘not only specifies the relationships between the independent variables and the ultimate dependent variable of interests... but also makes explicit the relationships among prior variables’ (p.9). Although the data do not meet all the assumptions for using multiple regression (for example, some of the distributions are not normal), it was decided that a path diagram would be useful in providing a global picture of a complex system. Kerlinger (1986) has pointed out the validity of employing such a model even with the limitations noted above.
