A meta-analysis of the studies on the effect of computer-assisted instruction on learning English vocabulary

A meta-analysis of the computer-assisted instruction and English language vocabulary learning literature until 2015 was conducted. The present study brings together evidence from a diverse field of methods for comparing computer-assisted and traditional vocabulary learning. In total, 57 studies containing 96 independent learning outcomes were analyzed. The Comprehensive Meta-Analysis (version 2.0) software was used to perform the statistical analysis. After sensitivity analysis, 45 effect sizes were omitted, and the remaining 51 effect sizes were analyzed. Overall, the results indicated a moderate positive effect of computer-assisted instruction on English language vocabulary development (d=.69) under the random-effects model since significant heterogeneity was observed in the obtained effect sizes. The findings also indicated that the duration of instruction, proficiency level of the participants, learning style, and the instructional method moderated the results, meaning that students who received computer-assisted language instruction for a month learned vocabulary (d=.97) better than those who received the instruction for more than a month (d=.46), high school students (d=.81) learned better than elementary and guidance school students (d=.38), cooperative vocabulary learning groups (d=.89) did better than non-cooperative vocabulary learning groups (d=.57), and that student-centered vocabulary learning (d=.78) was better than teacher-centered vocabulary learning (d=.59). Finally, dividing achievement outcomes into synchronous (d=.62) and asynchronous (d=.71) forms of instruction yielded no significant difference. Implications of the findings for the current educational system were discussed.