Observational Play Assessment in Recreational Therapy (OPART): Applications for the Validation of Assessment in Recreational Therapy

Abstract

The need for standardized assessments in healthcare is rapidly growing as new discoveries are made. In the field of Recreational Therapy (RT), there are primarily agency specific developed assessments that have not yet been analyzed and proven reliable or valid1. As a growing occupation, it is critical that RT assessments are evaluated to provide evidence of the effectiveness of treatment. This study analyzed the reliability and validity of a new RT Assessment, the Observational Play Assessment in Recreational Therapy (OPART). The assessment was designed to provide recreational therapists the opportunity to assess levels of physical, cognitive, emotional, and social skills of children with disabilities while observed in play settings. Reliability and validity evidence was generated through the comparison of the OPART to another psychometrically sound assessment used in RT, the Comprehensive Evaluation in Recreation Therapy (CERT). Results suggested that the OPART had some evidence of interrater reliability (overall % agreement = .73; section agreement ranging from .60-.87%) with higher agreements when measuring functional skills with physical attributes (e.g., gross motor skills, fine motor skills, endurance/weakness). The OPART had some evidence of convergent validity with overall (rs= .611; p=.016) and adjusted (rs = .738; p=.002) scores when compared with the CERT. Further analysis of individual sub-sections of the OPART with the CERT suggested limited evidence of convergent validity as only one of the four sections approached statistical significance. The OPART also demonstrated some evidence of predictive validity in its’ ability to accurately classify the functional level of the participants when compared to those assigned by staff familiar with individuals observed. The OPART was accurate in 8/15 (53.3%) cases while the standardized CERT was slightly more accurate at classifying participants in 9/15 (60.0%) cases. Overall, the OPART had some evidence of convergent validity with the CERT in terms of adjusted scores (rs=.738; p= .002) and total scores (rs=.611; p= .016) with total scores. While this initial step to validate a new instrument was promising, it is clear that significant development and validation is indicated. Recommendations and future considerations are provided for the validation of agency specific RT assessments.