Answers are grounded in this paper and cite their sources.
Background and Purpose. A central issue surrounding the use of a visual analog scale (VAS) of the Physical Therapist Clinical Performance Instrument (CPI) is the level of measurement represented by the method. An often-made assumption is that the VAS is an interval-level measure of the construct of interest. In order for the scale to be valid and for this assumption to be supported, raters must demonstrate the ability to discriminate among 100 levels of student performance. The purpose of this study was to use a measurement model to investigate the validity of the VAS of the CPI when used by clinical instructors to evaluate student performance. Subjects. Final CPI raw scores were collected from 182 subjects from 3 different physical therapist education programs across 4 full-time clinical experiences. Methods. Visual analog scale raw scores for each item were measured (in millimeters) along with rater use of the With Distinction Box. An analysis using the Rasch Measurement Model was performed on the collected data. Step calibration values and average measure values provided by the model were used to analyze rater use of the VAS. Results. A valid rating scale was achieved when the VAS was considered to be a 6-level ordinal scale. Following the collapse of response categories, the 6 groupings of VAS raw scores were as follows: 0-30 mm=l, 31-50 mm=2, 51-70 mm=3, 71-90 mm=4, 91-100 mm=5, and 100 mm + With Distinction Box=6. Discussion and Conclusion. The inability of clinical instructors to discriminate among 100 levels of student performance is most likely contributing to the invalid scale.
No takes yet. Share an insight, caveat, or question.
Straube et al. (2003) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: