on and sistencies in the classificatiThere are many incon classification ood behaviors. The establishedassessment of childhlike the IDEA system ldren are either nonspecific, systems used for chiystem. Finally, the empirical data, like the DSM sor not derived from en use the absent ms currently in use for childrclassification systeildrens behavior at t allowing for the study of chvs. present model noamines an severity. The current study exdiffering levels of ldrens behavior in pology of 6 to 11-year-old chiempirically based tys the potential to termines if this typology holdthe classroom and deor problems in cation system for child behavibe a useful classified, but few have onal systems have been developschool. Many dimensi, thus many have not standards in their developmentadhered to rigorous ght to externally results. The current study souproduced replicable mative sample. gy found by the BASC-TRS-C norreplicate the typoloample of children was conducted on a referred sExternal replicationhat the problem6 and 11. The results showed tbetween the ages of rred sample. The ntially replicated in the refeclusters were substaternal correlates of re then compared on several exobtained clusters weand overall havior, academic achievement, cognition, school beo significant t. Contrary to expectations, nfunctional impairmenial reasons for he clusters were found. Potentdifference between t future tests of the nsidered with implications forthese results are co analytic typology discussed. utility of a cluster