Effects of EMI-CLIL on secondary-level students’ English learning: A multilevel meta-analysis

This meta-analysis synthesized the effects of the English medium instruction and content and language integrated learning (EMI-CLIL) approach on secondary-level students’ English learning. The dataset included 44 samples (N = 7,434) from 38 primary studies. The results revealed EMI-CLIL’s overall effectiveness for the development of English competence compared to the mainstream condition in the short term (d = 0.73, SE = 0.06, 95% CI [0.61, 0.86]) and longer term (d = 1.01, SE = 0.06, 95% CI [0.88, 1.15]). Additionally, we found that EMI-CLIL’s overall effectiveness was influenced by several moderator variables. Its effectiveness was significantly: (1) higher for learners whose first language (L1) was linguistically related to English; (2) lower for primary studies which confirmed the homogeneity of the EMI-CLIL and comparison groups; (3) lower when studies targeted the productive (rather than receptive or overall) dimension of English learning; and (4) higher when outcome measures focused on vocabulary. Implications for pedagogy and future research are discussed.