Expert consensus on designing a metaverse supported blended EFL module in chinese higher education: A Fuzzy Delphi method

Abstract

Metaverse technologies can provide immersive, interaction-rich experiences for English as a Foreign Language (EFL) learning, yet curriculum-level design principles and implementation guidance remain limited. This study developed, through expert consensus, a ranked blueprint for a metaverse-supported blended learning module for undergraduate EFL learners in Chinese higher education. Using an exploratory design, Phase 1 elicited candidate elements through semi-structured interviews with EFL instructors (n = 5), and Phase 2 applied the Fuzzy Delphi Method (FDM) with a 12-expert panel using a five-point Likert scale. Items were retained only if they met all prespecified criteria (Agreement ≥ 75%, interquartile range [IQR] ≤ 1.0, and fuzzy distance [d] ≤ 0.20). Retained items were prioritized by defuzzified value (DV), with ties resolved by IQR, then d, then Agreement. Actionable consensus was reached across six domains: learning objectives, learning content, instructional strategies, learning activities, assessment methods, and learning resources. Resampling-based stability checks (leave-one-out and bootstrap) supported the robustness of the induced priority ordering. Kendall’s W indicated limited overall concordance across heterogeneous items; accordingly, item inclusion relied on the prespecified thresholds. The study contributes a replicable, ranked blueprint that embeds constructive alignment in immersive EFL contexts and provides an implementation-ready specification to support staged adoption and subsequent validation in higher education.