A study of the effectiveness of CAI to support beginning level EFL grammar instruction in Taiwan (China)

The purpose of this study was to explore whether the addition of CAI to traditional instruction would have an effect on reducing Taiwanese beginning EFL students' written error rates, as well as to explore the relationship between student in-house TOEFL scores and their written error rates. A quasi-experimental research design was conducted at a private college located in southern Taiwan. The subjects were divided into two groups: the control group (traditional English grammar instruction) and the experimental group (CAI as a supplement to English grammar instruction). A post-writing assessment was administered for both control group and experimental group after the treatment. Data were analyzed through a one-way ANOVA on overall error rates as well as a series of one-way ANOVA on error rates within individual error categories. The Pearson correlation coefficient was used as well to examine the relationship between students' TOEFL scores and their error rates.; The major finding on overall error rates demonstrates that there was no statistical difference between the control group and the experimental group. However, findings on individual error categories suggest that CAI as an instructional aid may have produced a significant effect on reducing students' error rates in the error category of nouns and prepositions. Due to some factors beyond the scope of this study, more research is needed to examine this effect.; The findings in terms of the relationship between students' TOEFL scores and error rates indicate that there was a significant relationship (weakly negative) between TOEFL scores and error rates. That is, those students who had higher TOEFL scores tended to have lower error rates. Likewise, those students who had lower TOEFL scores tended to have higher error rates.; Although the supplemental CAI program in this study did not produce statistically significant effects on reducing beginning EFL learners' overall written error rates, no evidence was found suggesting that CAI had a negative effect on learning. The research concluded that CAI as a supplement to traditional instruction learning could be at least as good as traditional instruction.