Occupational performance in school settings : evaluation and intervention using the school AMPS

Abstract: Background: This thesis is was designed to evaluate aspects of reliability and validity of the School Version of the Assessment of Motor and Process Skills (School AMPS) (Fisher, Bryze, Hume, & Griswold, 2007), an observation-based evaluation of quality of occupational performance when children perform schoolwork tasks in school settings. The long term goal was to contribute to knowledge about children at risk or with mild disabilities who experience difficulties with occupational performance in school settings, and describe how the School AMPS can be used when a true top−down process of planning and implementing school-based occupational therapy services is implemented in a Swedish context. Methods: In Study I, two different split-half methods and were used to estimate reliability of the School AMPS measures. These were cross-validated using Rasch equivalent of Cronbach’s alpha. The standard error of measurement (m) was also calculated. In Studies II and III, many-facet Rasch analyses and/or relevant inferential statistics (e.g., ANOVA, tests) were used to examine for evidence of validity based on (1) internal structure related to differential item functioning (DIF), (2) relations to other variables (sensitivity) in terms of comparing groups (typically-developing children vs. children with mild disabilities), and (3) consequences of testing (benefits of testing) in terms of test fairness. In Study IV, ANOVA and tests were used to examine relations to other variables in terms of sensitivity of the School AMPS measures for detecting change based on repeated School AMPS evaluations pre- and post-interventions. Results: The three methods for estimating reliability of the School AMPS measures yielded high reliability coefficient estimates (≥0.73) and low ms. Minimal DIF was identified, and despite minimal DIF, the School AMPS measures were found to be free of differential test functioning. The School AMPS measures were sensitive enough to detect differences between groups as well as changes following consultative occupational therapy services provided in natural school settings. Conclusions: The results support the reliability and validity of the School AMPS scales and measures when used to evaluate quality of occupational performance in school settings. The results are also of clinical importance as they provide evidence that occupational therapists can have confidence in the School AMPS measures when they are used in the process of making decisions about individual students, planning interventions, and later perform follow-up evaluations to measure the outcomes. We also have objective evidence that children with mild disabilities demonstrate diminished quality of "doing" when performing schoolwork tasks. The potential long term benefits of such evidence may be to support or justify the need for children with mild disabilities to receive occupational therapy services within school settings in Sweden; and through collaboration with teachers, plan and implement better targeted and more effective interventions.