Published online by Cambridge University Press: 29 January 2018
Torgerson (1968) has suggested that the most appropriate form of classification of functional mental illnesses may prove to be partly categorical and partly dimensional. Everitt, Gourlay and Kendell (1971) have tended to agree with this conclusion since their attempt at validating existing diagnostic categories by means of cluster analysis proved to be only partially successful. In the present paper the dimensional approach is considered in some detail and it is shown that it has limitations which are due in part to the distributional properties of the data which research psychiatrists record for their patients, and in part to the fact that some key symptoms occur only rarely. In the final paragraphs of the paper an attempt is made to clarify the dimensional and categorical roles in the description of functional illnesses. In doing so it becomes clear that some difficulties remain unresolved.
eLetters
No eLetters have been published for this article.