Analyzing User Expectations and Usability Alignment in Augmented Reality Using Principal Component Analysis and Regression

Abstract

Understanding visitor expectations is crucial for developing user-centered Augmented Reality (AR) applications for cultural heritage. This study investigates the relationship between visitor expectations and perceived usability of the Mooi Indie AR painting application using Principal Component Analysis (PCA) and regression analysis. Initial data were collected from 78 participants, and PCA reduced eight expectation variables into three principal components, which explained 65.73% of the total variance. PCA reduced these into three main components: 1) Perceived Experience Value, 2) Engagement and Cultural Familiarity, and 3) Technology Adoption Intention. Regression analysis was then conducted as an exploratory approach to examine the relationship between the extracted expectation dimensions and perceived usability. Results indicate that the AR application generally met or exceeded visitors' initial expectations, demonstrating its potential to enhance user engagement in cultural heritage exhibitions. The proposed analytical framework provides practical insights for identifying key expectation dimensions to support the design of more effective, user-centered AR applications.