Evaluating metaverse-based L2 vocabulary learning effectiveness using a proposed metric of vocabulary forgetting percentage

Introduction Vocabulary gain and retention are widely recognized as essential metrics in second language (L2) vocabulary learning. However, these traditional measures often fail to reflect the proportional loss of learned vocabulary knowledge over time, thus limiting their practical, diagnostic, and comparative value across different instructional contexts. Methods To address this gap, the present study proposes a percentage-based metric: Vocabulary Forgetting Percentage (VFP). To evaluate metaverse-based vocabulary learning (VL) effectiveness and also to empirically validate the VFP, a quasi-experiment was conducted, involving 50 Chinese middle school EFL learners who were assigned to either a metaverse-based group (MG) or a slides-assisted control group (SG). Over three learning sessions, participants learned equivalent vocabulary content and completed pretests, immediate post-tests, and delayed post-tests. Quantitative data were analyzed using independent-samples t-tests to compare vocabulary gains, retentions, and VFPs across groups. Results The MG significantly outperformed the SG in vocabulary gain and retention in each session and in mean scores. However, VFP results showed a different pattern: the MG’s third-session and mean VFPs were significantly lower than those of the SG, while differences in the first and second sessions were not significant. Discussion The MG’s late-emerging VFP difference from the SG’s suggests that extended exposure to immersive environments may be required before full benefits appear. Findings also confirm the pedagogical potential of the metaverse for vocabulary learning and empirically validate VFP as a complementary metric. By proportionally quantifying vocabulary loss, VFP offers researchers and educators a more nuanced tool for evaluating learning efficiency and retention sustainability in varied L2 contexts.