Dimensions Versus Categories--The Empirical Distinction
An Important Distinction Empirically?
Once one accepts that the dimensions versus categories distinction matters conceptually, there remain two empirical questions: first is the question of whether the distinction matters empirically; second is the question of whether the distinction can be empirically detected. In psychology, the best work on the detection of latent classes has been done by Meehl (1995). Indeed, he has coined his own word for the numerical aspects of category detection: taxometrics. Because it brings the importance of the empirical questions answered by taxometric research clearly into focus, I wish to quote at length from a recent piece of such research:
"When typological models are valid they provide information that cannot be obtained from dimensional models, such as group membership probabilities, latent base rates of types, and indicator specificity and sensitivity rates (Meehl, 1995). When they are not valid, however, unfortunate problems can arise that serve to undermine the research enterprise--such as reductions in statistical power (Cohen, 1988), decreases in scale reliability (Cohen, 1983), the spurious overestimation (Maxwell & Delaney, 1993) and underestimation (Cohen, 1983) of empirical relationships, and the inability to uncover nonlinear relationships with other variables (Tellegen & Lubinski, 1983)" (Fraley and Waller, in press, p.1).
A Detectable Distinction Empirically?
This quote should show the empirical importance of the question of whether psychological disorders in fact come in categories or dimensions. Specific procedures have been developed to determine whether disorders are packaged as traits or types. Two such procedures, called MAMBAC and MAXCOV-HITMAX, have been developed through the persistent efforts of Meehl and colleagues (Meehl, 1995; Meehl and Golden, 1982; Meehl and Yonce, 1994; Waller and Meehl, 1997). While this is not the forum in which to detail the nature of these taxonometric methods, it is the forum in which to call attention to their potential for answering the empirical questions that our conceptual analysis has shown to be so important for those of a realist philosophical persuasion.