Objective:: The primary aim of the study was to identify a classification scheme, for determining the predominant type of maltreatment in a child's history that best predicts differences in developmental outcomes. Method:: Three different predominant type classification schemes were examined in a sample of 519 children with a history of alleged maltreatment. Cases were classified into predominant maltreatment types according to three different schemes:1.Hierarchical Type (HT)-based on a hierarchy of types that prioritizes active forms of abuse over passive abuse; 2.Severity/Frequency Type (SFT)-sorts cases into the type of maltreatment associated with the highest severity or frequency rating; 3.Expanded Hierarchical Type (EHT)-differentiates multiple maltreatment type combinations from ''pure'' or single sub-types. Hierarchical regression analyses examined whether the HT, SFT, and EHT type classifications contributed to prediction of child behavior problems, trauma symptoms and adaptive functioning. Results:: After controlling for demographic factors, the HT definitions predicted four outcomes, while the SFT definitions predicted three, and the EHT classifications contributed to the prediction of five child outcomes. The co-occurrence of multiple types of maltreatment was robustly related to outcomes. However, the HT and SFT classifications predicted outcomes even after accounting for the co-occurrence of multiple maltreatment subtypes. Conclusion:: A classification scheme that differentiates between type combinations and single maltreatment types may have the greatest predictive validity. Over and above knowing about co-occurrence of maltreatment sub-types, it is important to understand what type, or constellation of types, of maltreatment have been alleged in a child's history.