EFFECTS OF VISUALS IN PEER DISCUSSION USING SPLIT INFORMATION AND SHARED INFORMATION TASKS

Abstract

The study examines the effectiveness of certain task types and types of input in generating talk and peer questions. The role of different kinds of task features in studies of peer interaction without a focus on the influence of input in interaction. Yet task types and input are both important considerations. This study highlights the importance of visuals as input for interaction in split information tasks. The study aimed to examine the effectiveness of visuals as input for interaction in two task arrangements. The first task arrangement involved split information where learners work with individual information for a common goal while the second task grouping involved shared information. Two types of input, visuals and text consisting of key words, were used in both the split information and shared information tasks within and across task types. The study focused on three aspects of peer interaction; quantity of talk, quality of talk and peer questions. A mixed ability, multi-cultural class of 32 students from a Singaporean secondary school participated in the study. The students were randomly assigned to groups of four, bringing the number of groups to eight. The teacher and class records were consulted to ensure a mixed grouping. Within each task type four groups worked with visuals and the other four worked with texts consisting of key words. All eight groups participated in both split information and shared information tasks. Peer discussions of all groups were audio-taped and transcribed to yield data which was analysed for quantity of talk, type of vocabulary items used and type of questions asked during peer interaction. Statistical analyses of the results showed that split information tasks were effective in generating quality talk and a range of peer questions. More importantly, the results indicated that visuals as input when used in split information tasks had potential for generating quality talk and content-oriented questions. The study also looked into types of peer questions and found that more questions were asked in split information tasks compared to shared information tasks. Two major categories of questions were identified from the transcript analysis as either procedural questions or content-oriented questions. Also of interest to the study was the number of content-oriented questions asked during interaction when visuals were provided as input in both task types. However when the two task types were compared, split information tasks with visuals as input for interaction generated more content-oriented questions than shared information tasks with visuals. Overall the results of the study have highlighted the importance of the type of input used in peer interaction. The study also clarified the importance of choosing a task type that would ensure participation of all in group based activities in linking input and the task type as influences in fostering talk. Results show that visuals as input for interaction work well in split information tasks for generating talk and peer questions.