Self-reported depression is taxonic

Using the Sample-specific simulation method proposed by J. Ruscio, A. M. Ruscio, and T. M. Keane (2004), the S. R. H. Beach and N. Amir (2003) data are re-analyzed. The results for the MAXEIG and MAMBAC procedures are compared across simulated dimensional, simulated taxonic, and original distributions. The comparison indicates that depression is taxonic in the S. R. H. Beach and N. Amir (2003) sample but has a lower base (.06) than previously reported. This result indicates the need to use a relatively high cut point on total Beck Depression Inventory (i.e., BDI>21; A. T. Beck, 1978) when creating groups using the BDI. That is, the data support the presence of a discontinuity between "depressed" and "nondepressed" undergraduates at a relatively high level of depressive symptomatology. At the same time, it is noted that researchers should remain cautious about the interpretation of taxometric procedures used with low base rate taxa until they have been further investigated using Monte Carlo simulations. © 2006 Springer Science+Business Media, Inc.