ITEM RESPONSE THEORY SCALING OF AN ACADEMIC MEDICAL CENTER STUDENT SATISFACTION SURVEY

Abstract

Item response theory scaling has been well established in achievement testing, but the practice has seen limited use in student development research. The purpose of this study was to explore the reliability and validity of a graded response model scaling of an existing college student satisfaction survey. The existing survey was found to measure three aspects of student satisfaction: community, diversity, and substance abuse. The three scales were found to have acceptable internal consistency. Parameter estimates using the grade response model were able to be obtained for the community and diversity but not the substance abuse subscale. A reliable community satisfaction score scale was created for the community subscale. Validity evidence was accumulated based on established research about student academic program, student involvement, and mode of survey. Differences in community satisfaction scores between involved and not-involved students were present, differences in community satisfaction scores between online- and paper-based surveys were present, and differences between community satisfaction scores and student academic program were not present.