Measurement Properties of Fine Motor Scale of Peabody Developmental Motor Scales-Second Edition

Abstract

To investigate the measurement properties (including rating scale performance, unidimensionality, and differential item functioning) of the fine motor scale of the Peabody Developmental Motor Scales-Second Edition in children, by using the Rasch analysis. A total of 419 children (including 342 typically developing children and 77 children with fine motor delays or difficulties) were recruited in Taiwan for this prospective study. Each child was evaluated with the Peabody Developmental Motor Scales-Second Edition that consists of 26-item grasping and 72-item visual-motor integration subtests. Partial credit Rasch analysis was used for all analyses. The Rasch analysis indicated that middle rating category for 19 grasping and 52 visual-motor integration items could be collapsed to allow only dichotomous response categories. Item fit analysis and principal component analysis suggested that the unidimensionality of the grasping and visual-motor integration subtests could be achieved after removal of two grasping and eight visual-motor integration misfitting items. All but 13 items in the composite scale could form a unidimensional construct of overall fine motor ability. Furthermore, only a few items were found to show differential item functioning across sex (ten items) or fine motor status (seven items). However, significant ceiling effects were found in the Peabody Developmental Motor Scales-Second Edition subtests and composite scale when applied to these typically developing children. Our results suggest grounds for the revision of the Peabody Developmental Motor Scales-Second Edition in a subsequent edition. Simplifying the rating scales and reducing the misfitting items in the subtests and composite scales might result in a unidimensional assessment of children's fine motor ability. Clinicians and researchers could use the reduced Peabody Developmental Motor Scales-Second Edition as a criterion-referenced outcome measure to document changes; however, further work is needed to reduce the ceiling effects.