Augmented reality technology in language learning: Ameta‐analysis

Abstract Background As a recently emerging innovative technology, augmented reality (AR) has become a popular tool for language learning. However, to date, very few meta‐analytical studies has been conducted on AR in this field to understand its effectiveness on language learning. Objectives This meta‐analysis was conducted to systematically synthesize the findings from primary studies published between 2008 and 2020 to establish the effects of AR on language learning gains and students' motivation. Methods The 21 studies met all the inclusion criteria were included in the meta‐analysis to extract effect size statistics. The robust variance estimation (RVE) technique using the “Robumeta” R‐package was adopted to estimate the pooled effect size. Given the heterogeneity of the effect sizes, a mixed‐effects meta‐regression model was estimated to examine any association between the effectiveness of AR technologies and moderator variables. Results and conclusion The pooled effect‐size estimate was 0.93 for language gains and 0.42 for motivation, which indicates that AR applications have a large effect on learners' language gains and a small to medium effect on learners' motivation. The moderator analysis results suggested that learners' educational levels and intervention durations are significant moderating factors that impact the effect of AR on learners' motivation. In particular, elementary school students in this meta‐analysis study experienced a large measurable effect in terms of both language gains and motivation. Additionally, exposure to AR applications for up to 1 week is especially effective for enhancing learners' motivation. Major takeaways The findings of this meta‐analysis study demonstrated how AR applications can be utilized in language teaching and learning contexts, and how language educators could adopt AR technologies in classrooms to promote learners' language gains and motivation.