The Bruininks-Oseretsky Test of Motor Proficiency -- Second Edition (BOT-2) is a commonly used assessment of children's skills. It is important that assessments have validity evidence reported about them. The objective of the study was to investigate the structural validity of the BOT-2's eight subscales and four composite scales. A sample of 117 healthy children (65 males and 52 females; M = 10 years, 2 months, SD = 1 year, 4 months) completed the BOT-2. The Rasch Measurement Model was used to assess the dimensionality, hierarchical ordering, differential item functioning (DIF), and reliability of the eight BOT-2 subscales and four composite scales. The Fine Motor Precision (FMP), Manual Dexterity (MD), Running Speed & Agility (RSA), Upper Limb Coordination (ULC), and Strength (S) subscales exhibited reasonable dimensionality while the Fine Motor Integration (FMI), Bilateral Coordination (BC), and Balance (B) subscales did not. The Motor Coordination Composite (MCC) and Strength & Agility Composite (SAC) composite scales demonstrated good dimensionality while the Fine Motor Manual Control Composite (FMMCC) and Body Coordination (BCC) scales did not. The FMP, MD, BC, and RSA subscales and SAC composite scales did not exhibit adequate hierarchical ordering. The FMI, B, RSA, and ULC subscales and the MCC and BCC composite scales exhibited DIF based on gender. The eight subscales and four composite scales all had excellent item reliability with all coefficients being >.95. A number of the BOT-2 subscales and composite scales appear problematic with two or more issues in relation to dimensionality, hierarchical ordering, DIF and/or reliability: FMI, BC, B, RSA, and BCC. Therefore, only some of the BOT-2 subscales and composite scales can be used with confidence.