Abstrakt
In this work, we present overview of validation studies of admission test to Czech and foreign colleges, mostly consisting of analysis of predictive validity. Further, on example of validation of conceptual assessments in physiology, we demonstrate that validation encounters besides gathering proofs of predictive validity also providing proofs of content and construct validity, estimation of test-retest reliability and internal consistency, as well as providing detailed item analysis assuring range of item difficulties, sufficient item discrimination and checks for item fairness towards gender and minority groups.
Last, we apply these methods to provide complex validation of admission test to medical school. We demonstrate, how freely available statistical software R and its web application framework Shiny can provide complex but still user-friendly analysis toolbox.