The effects of error treatments and students' language abilities on the usage of English tenses through the use of computer-assisted language learning

This study investigated the effects of different error treatments-overt correction (OC) and self-correction (SC)-on the usage of English tenses of Thai undergraduate students. Both treatments were given through CALL. Language abilities of the students-high, moderate, and low achievers-served as a moderator variable. The study was conducted with 210 first-year undergraduates of Huachiew Chalermprakiet University in a 2x3 factorial design. At the commencement of the study, all subjects were pre-tested by using the Test of English Tenses Usage (TETU) developed by the researcher. After they completed 9 CALL lessons in 3 weeks, the posttest was conducted. Two-way analysis of Variance (ANOVA) and descriptive statistics were used to analyze and to explain the data. Six weeks after the posttest, the delayed test was administered to investigate the retention of both treatments. The data was then analyzed by two-way repeated measures ANOVA. Results from the immediate posttest showed that the error treatment factor did not have a significant effect on the usage of tenses, while the language ability factor did. An interaction effect between the error treatments and the language abilities was not found. Results from the self-rating scales revealed that students in both treatment groups rated that they knew about the tenses much more than they did before practicing with the program. However, comparison of their ratings across treatment groups did not yield a significant result. Regarding the differences of English tenses used by different ability groups, it was found the two prominent tenses that the low group could perform were present continuous and simple present; the moderate ability group, in addition to these two tenses, could do the future tenses. The high ability group could master almost all tenses; the two tenses that were found to be the most difficult for this group were past perfect and present perfect. Results from the delayed test revealed that the self-correction group had significantly higher retention than the overt correction group did. Additional findings from the opinionnaire also confirmed that the SC group rated the materials significantly higher than the OC group in terms of the usefulness of the feedback.