Measurement error and detectable change for the modified Fresno Test in first-year entry-level physical therapy students.

Abstract

Teaching evidence-based practice (EBP) skills is a core component in the education of health care professionals. Methods to assess individual student development of these skills are not well studied. The purpose of this study was to estimate the standard error of measurement (SEM) and minimal detectable change (MDC) for the modified Fresno Test (MFT) of Competence in EBP in first-year physical therapy students. Using a test-retest design, the MFT was administered two times to 35 participating first-year physical therapy students. Tests were scored by two trained physical therapist educators. Mean test scores clustered near the middle of the 232 point scoring range, 107 points (SD 14.9) and 103 points (SD 18.9). Inter-rater reliability [ICC (2, 1)] for scorers was 0.83 (95%CI 0.74-0.96). Intra-rater reliability was 0.85 (95%CI 0.60-0.97) and 0.94 (95%CI 0.86-0.99). Test-retest reliability [ICC (2, 1)] was 0.46 (95%CI 0.16-0.69), with a calculated SEM of 11 points, a confidence in a single measurement of 18.2 points, and MDC90 (90% confidence) of 25.7 points. Knowledge about estimates of SEM and MDC for specific student populations is important to assess change in individual student performance on the modified FT.