Options
Reactivity of concurrent verbal reporting in second language writing
Author
Yang, Chengsong
Supervisor
Hu, Guangwei
Abstract
Concurrent verbal reporting has long been used as a data elicitation tool in second language acquisition (SLA) to explore learners’ cognitive processes. However, it is only recently that concerns have been raised over the issue of whether such a tool will cause reactivity, i.e., whether it will alter the very cognitive processes it is assumed to accurately reflect. This paper reports an empirical study designed to explore this possible effect in the process of second language writing, an area that has rarely been investigated previously. Ninety-five participants were randomly assigned to three writing conditions: no thinking aloud (NTA), non-metacognitive thinking aloud (NMTA) and metacognitive thinking aloud (MTA). All participants first wrote an argumentative essay silently to provide baseline data which would be used to assess between-group homogeneity and to statistically control prior between-group differences. Then, in the main task, the NTA group completed a similar essay silently, while both think-aloud groups were asked to think aloud while they were writing about the same essay topic. The NMTA group were asked to report what they were thinking, while the MTA group were told to think aloud what was going on in their minds and additionally justify what they wrote. There was no time constraint on either the baseline or the main task. Participants’ essays were analyzed in terms of linguistic fluency, complexity and accuracy to see if there would be any significant between-group differences. Such differences were taken as indicative of reactivity. Results indicated that the metacognitive type of verbalization significantly increased time on task, caused participants’ rate of production to slow down, incurred dysfluencies, and affected participants’ willingness to use different verb forms, compared with the non-metacognitive type of verbalization, which was reactive only to the number of different verb forms. Kellog’s (1996) model of writing processes in relation to Baddeley’s (1986) working memory is used to explain the reactivity caused by the two types of thinking aloud and cautious use of these two tools is suggested.
Date Issued
2011
Call Number
P118.2 Yan
Date Submitted
2011