Knowledge Bank

Over the past decade, hundreds of studies have introduced genomics and bioinformatics (GB) curricula and laboratory activities at the undergraduate level. While these publications have facilitated the teaching and learning of cutting-edge content, one key aspect of evidence-based practice has been left behind: the development of assessment tools capable of generating valid and reliable inferences about student learning. Content validity is a core facet of construct validity, and must be used to guide instrument and item development. Based on previous work which reported on the correspondence of content validity evidence gathered from independent sources, our current work details: (1) the process of item development using this evidence and (2) the results from a pilot administration of the assessment. By including only the subtopics that were shown to have robust support across our content validity sources, 22 GB subtopics were established for inclusion in our assessment. An expert panel subsequently developed, evaluated, and revised two multiple-choice items to align with each subtopic, producing a final item pool of 44 items. These items were piloted with student samples of varying content exposure levels. We report on Classical Test Theory (CTT) and Rasch analyses of individual items and overall instrument quality.