A Rasch-based validation of the Hooper Visual Organization Test in Chinese-speaking children

Rasch analysis of data from typically developing children showed that twelve items were candidates for deletion due to poor fit to the Rasch model, violation of normality and age-related item bias. Removing these items resulted in a shortened version with 18 items that forms a reliable and strong unidimensional, hierarchical scale. The items were well targeted to the ability level of the children tested. Area under the curve for HVOT-18 was 0.84, indicating very good ability to identify visual integration deficit in children with Down syndrome. The 18-item HVOT can be summed to produce an overall index of visual synthetic ability. Subsequent work is needed to validate its use in other childhood disabilities.