Psychometric Analysis of an Integrated Clinical Education Tool for Physical Therapists.

Abstract

Integrated clinical education (ICE) courses require opportunities for practice, assessment of performance, and specific feedback. The purposes of this study were to 1) analyze the internal consistency of a tool for evaluating students during ICE courses, 2) examine the responsiveness of the tool between midterm and final assessments, and 3) develop a model to predict the final score from midterm assessments and explore relationships among the 6 domains. Several clinical education assessment tools have been developed for terminal clinical experiences, but few have focused on the needs of learners during the ICE. Eighty-five student assessments were collected from 2 consecutive cohorts of physical therapist students in a first full-time ICE course. The tool contained 29 items within 6 domains. Items were rated on a 5-point scale from dependent to indirect supervision. Cronbach's alpha was used to analyze the internal consistency of the tool, whereas responsiveness was examined with paired t-test and Cohen's d. A best subsets regression model was used to determine the best combination of midterm variables that predicted the final total scores. Coefficients of determination (R2) were calculated to explore the relationships among domains. The tool was found to have high internal consistency at midterm and final assessment (α = 0.97 and 0.98, respectively). Mean scores increased over time for each domain score and for the total score (P < .001; d = 1.5). Scores in 3 midterm domains predicted more than 57% of the variance in the final total score. Results support the use of this tool to measure student performance and growth in a first full-time ICE course. Targeted measurement of students' abilities in ICE courses assists with differentiating formative and summative learning needed to achieve academic success.