Abstract
As the world becomes a global village, parents who want their children to be more competitive in the future will send them to learn English, an international language. This stimulates the birth of bilingual schools. Most of the bilingual programs in Taiwan, according to Amato’s (1996) and Brown’s (1994) definition and classification, adopt the immersion type. Similar to many examples of students’ incomplete acquisition in Canadian immersion programs (Day & Shapson, 1987; Swain & Lapkin, 1986), the questionnaire results done in an immersion program in Taiwan (Li, 2002) indicated that the necessity of grammar instruction was affirmed, but which method the teachers should adopt remained uncertain. In fact, whether and how to include ‘grammar’ in L2 immersion programs is controversial. From many examples of incomplete acquisition in input-rich environments such as immersion programs (Harley & Swain, 1984; Day & Shapson, 1987; Lyster, 1987), it was confirmed that learning an L2 implicitly from exposure to comprehensible language input is not sufficient. Realizing that learners need to pay some degree of attention to language forms, Long and Robinson (1991) proposed a new approach – ‘focus on form,’ which draws learners’ attention precisely to a linguistic feature without distracting them from their original communicative intent. This project examined the effects of the ‘focus on form’ approach on EFL learning in an immersion program in Taiwan, and explored the potential for adopting it in Taiwan in the future. This study was conducted from October, 2002 to January, 2003. A total of 79 second-year senior high school students from two intact classes in an immersion program participated in this study. With a pretest, immediate posttest, and delayed posttest design, performance of the experimental group who received the instruction with the ‘focus on form’ approach was compared with that of the control group who did not receive the experimental treatment during an average 8 hours of an eight-week period. Students’ performance on the target structure - the English conditional- was assessed by the written measures, including a fill-in-the-blank task and a sentence-completion task on 79 students, and the oral measures, which asked students to make wishes, and complete sentences according to the given situations on nine students of each group. The accuracy rate of students’ correct and appropriate use of the English conditional on the four tasks of the three tests was compared by using the T-test procedure. Findings indicated that at the end of eight weeks, the experimental group showed a significant advantage over the control group on the written tasks on the immediate posttest measure. Although the experimental group also had higher gains on the oral tasks than the control group in the immediate posttest, the differences between them were recognized as statistically nonsignificant. One month later, both groups had worse performance in the delayed posttest. The experimental group even had significantly more regression on the written tasks than the control group. The findings confirmed that there were some immediate advantages in speaking and writing for students who were exposed to an experimental treatment designed to reflect the ‘focus on form’ approach to use the specific grammatical structure. Nevertheless, the positive effect was not maintained over 4 weeks. These findings are not conclusive, and further research is needed for an integral understanding of the effect of the ‘focus on form’ approach.