摘要：The aim of this study was to find out if Grammatical Cohesive Devices (GCDs) are affected uniformly by explicit, implicit, and incidental teaching methods and if there is any interaction between time and method. Since there were three teaching conditions (independent variables) and four GCDs (dependent variables), a three by four factorial design was used with MANOVA and Mixed Design ANOVA as the main statistical tests. The study used information from 60 intermediate EFL learners randomly assigned to the three teaching conditions. The explicit group was exposed to conscious learning. In the implicit group the learners largely achieved their knowledge from noticing and using GCDs rather than explicit instruction. And finally, in the incidental group, learning took place through reading the selected passages and figuring out the meaning of the texts without any conscious attention to GCDs. After the treatments, the learners took a posttest and the findings revealed that explicit teaching had been the most effective method of all. The interaction effect between time and method was also significant only for ellipsis and substitution.
其他摘要：The aim of this study was to find out if Grammatical Cohesive Devices (GCDs) are affected uniformly by explicit, implicit, and incidental teaching methods and if there is any interaction between time and method. Since there were three teaching conditions (independent variables) and four GCDs (dependent variables), a three by four factorial design was used with MANOVA and Mixed Design ANOVA as the main statistical tests. The study used information from 60 intermediate EFL learners randomly assigned to the three teaching conditions. The explicit group was exposed to conscious learning. In the implicit group the learners largely achieved their knowledge from noticing and using GCDs rather than explicit instruction. And finally, in the incidental group, learning took place through reading the selected passages and figuring out the meaning of the texts without any conscious attention to GCDs. After the treatments, the learners took a posttest and the findings revealed that explicit teaching had been the most effective method of all. The interaction effect between time and method was also significant only for ellipsis and substitution.
