127 - Comparing Assessment Validity of Three Visual Bell-ringer Modalities in Anatomy Testing

Poster Board Number: 127
There are separate poster presentation times for odd and even posters.
Odd poster #s – first hour
Even poster #s – second hour

Co-authors:

Yiming Zhang - McMaster University; Nicole Fu - McMaster University; Kaitlin Marshall - University of Toronto; Joshua Mitchell - McMaster University; Ilana Bayer - McMaster University; Judi Laprade - University of Toronto; Bruce Wainman - McMaster University; Joshua Nederveen - McMaster University; Peter Helli - McMaster University; Irena Rebalka - McMaster University

Abstract Body : Introduction: Anatomy is an inherently visual field that entails spatial recognition of structures in the human body. This is especially important in bellringer (BR) testing, which requires students to identify structures on cadaveric specimens, diagrams, or images. While traditional forms of BR testing using the visualization of cadaveric specimens have long remained the standard, three-dimensional (3D) viewing in virtual reality platforms (VR) or the use of two-dimensional (2D) images on paper are becoming more common. There is, however, a paucity of literature comparing the validity of evaluation in these newly popular modalities to the historic standard of presenting questions with cadaveric specimens, which this study seeks to investigate. Research Question: Compared to cadaveric specimens and 2D images, how do VR models influence BR question validity and testing performance among undergraduate students? Methods: Beginning in December 2023, approximately 150 undergraduate students in the Anatomy and Physiology I/II courses at McMaster University will participate in a BR exam containing 12 stations with questions provided in all three visual formats: specimens, 2D images on paper, and 3D images in VR headsets. Responses will be marked as correct or incorrect to determine item difficulty (# correct / study population) for each question. Additionally, points biserial (pBis) will be correlated with item difficulty to gauge question accuracy in representing student performance. This study will compare pBis values and item difficulties across all three modalities; informing whether disparate visual qualities impact test performance. Hypothesis: VR will yield the lowest item difficulties and highest pBis values, followed by cadaveric specimens, and images. This would validate VR as an accurate modality for representing anatomical structures, providing students with an informed foundation for participating in credible BR testing via alternate formats.