The Millon Clinical Multiaxial Inventory-II (MCMI-II) scores of 35 schizophrenic inpatients were compared with scores from a matched group of 35 psychiatric inpatients with no thought disorder. The operating characteristics of the Thought Disorder (SS) scale in overlapping and nonoverlapping form were determined using various cutoff points defined by Base Rate (BR) scores or the numbers of prototypic items endorsed by the subject. The Thought Disorder scale was found to perform poorly--no better than chance (Prevalence = 50%)--at correctly classifying patients as schizophrenic or nonschizophrenic, regardless of the type of cutoff lines used. A profile analysis was performed on six schizophrenia-relevant scales in search of a distinguishing schizophrenia profile. Results indicated that the schizophrenic group produced no distinct profile. Results are discussed in terms of their relevance for clinicians and further research.