| Sumario: | Introduction: Ecologically valid assessment linking cognitive strategies to observed functional performance has been strongly advocated. This study examines the construct validity of one such assessment: the Perceive, Recall, Plan and Perform (PRPP) System of Task Analysis for the evaluation of adults with brain injury. Method: Eighteen occupational therapists observed videoed patient performance (n = 16) of personal and instrumental activities of daily living, then used the PRPP System of Task Analysis to identify task performance errors and to attribute these errors to underlying cognitive strategy deficits. Multifaceted Rasch analysis was used to generate a hierarchy of test items and to propose a linear continuum, along which the difficulty of test items, raters, patients and tasks could be measured simultaneously. The generated hierarchy was examined for congruency with theories of information processing and neurorehabilitation. Results: Test items, raters, patients and tasks demonstrated 'fit' with the Rasch model. Construct validity was well supported by strong parallels between the Rasch-generated hierarchy of PRPP items and conceptual models of information processing and occupational performance. Conclusion: The PRPP System of Task Analysis is a valid measure of cognitive strategy use during the occupational performance of adults with brain injury, demonstrating high levels of fit with Rasch modelling for test items, raters and patients.
|
|---|