Process and product: An investigation of the writing of non-native speakers of English on a computer-based academic English writing test

Academic writing ability has been particularly recognized as one of the most crucial aspects of language ability for successful academic achievement. In addition, computer-based writing assessments are gaining prominence as the use of the computer technology in academic contexts has increased. Although much research has been conducted concerning either processes or products of academic writing, no study thus far has examined both processes and products involved in computer-based academic writing tests. The present study aimed to investigate the components on which ESL test takers focused, and the processes and strategies that they employed during a computer-based academic writing test.; The data of this study were collected from 87 students enrolled in ESL courses at University of California, Los Angeles. The students took a group-administered computer-based writing test, using the Track Changes function of the word processor. The Track Changes function identified changes that the test takers made during proofreading. In addition, the writing test was administered to 19 students individually in order to closely examine their writing processes. The individual test takers also participated in stimulus recall. All test takers' essays were scored based on holistic and analytic rating scales, focusing on language, organization, and content. Regression analyses were conducted to determine which subscales were significant predictors of good academic writing. Qualitative analyses regarding textual changes and verbal protocol were also conducted to explore students' strategies during the test.; The results showed that topic development, rather than organization or language usage, was the most significant predictor of good academic writing. A closer look at textual changes indicated that ESL students focused on content development during the test. Despite the time constraint of the testing situation, it was found that ESL students were engaged in a non-linear and recursive writing process. However, this non-linear process occurred in a compressed, rather than in an exhaustive, manner. This study also includes various findings related to the computer-specific strategies that were employed by the ESL students. Based upon the current findings, practical implications related to assessment and ESL instruction were also discussed.