Effect of Task Conditions on Writing From Sources

Abstract

The present study, motivated by the need to understand how students interpret and accomplish reading-to-write tasks for different purposes, examines the thinking and performance outcomes of undergraduate students who are writing from sources in an EAP course. The particular tasks investigated were those of summary, advocacy, and planning. The setting for the study was a Malaysian teachers' college from which a cohort of 135 student-teachers were selected and randomly assigned to one of three task conditions and reading orders. Two participants in each of the three task groups were selected as case study students to provide think-aloud protocols. In the design of the experiment, the factors of task and order of reading the source texts were between subject factors, while the dependent variables (task interpretation, origin of information, conceptual transformation, and rhetorical constituents) represented repeated measurements on the same student-teachers. Comparisons were made between the ways in which the three task groups interpreted the assigned tasks, as well as how they selected, transformed, and organised source information in their written essays. The quality of the observed trends in the data was assessed using both multivariate and univariate analysis of variance. The results of analysing the students' essays showed that the task groups differed considerably in their interpretation of the three tasks and in their selection, conceptual transformation and organisation of source ideas in their essays. A summary task was interpreted as a highly text-dependent task with written outcomes characterised by extensive use of source ideas which were usually organised within a topical collection, view-elaboration framework. On the other hand, a planning task, and to a lesser extent an advocacy task, were seen as highly prior knowledge-dependent and concept-driven with written outcomes characterised by the writer's ideas and by the extensive synthesis of source ideas. The results of analysing the case students' verbal reports while reading source materials revealed: (1) similar reading-to-write behaviours (task interpretation, selecting, transforming, and organising of source ideas) to those inferred from the essays; (2) attention to task specifications during the second reading of source materials; and (3) the features of a good reader of multiple texts, namely: adopting an interpretive reading stance; prioritising task-related goals; detailed and efficient analysis of task requirements; efficient exploitation and coordination of available resources; and intertextual linking of source ideas. The results suggest the need to re-evaluate aspects of current reading-to-write programmes in the light of the study of task conditions and their effects, including (1) the particular language skills required for reading across texts and writing from sources, (2) the operations of interpreting a task, and selecting, transforming, and organising source information, (3) use of available resources, and (4) the opportunity for enhancing expertise of particular tactics or cognitive operations. Finally, a model of reading-to-write task effects is proposed in which the questions of how knowledge about tasks comes to be represented in students' minds and how this representation becomes available as a basis for their performance are seen to be important pedagogical issues.