Exploring Tests for L2 Implicit and Explicit Knowledge: In the Case of English Unreal Conditional

DOI Web Site

この論文をさがす

抄録

Whether or not explicit knowledge can be converted into implicit knowledge has been a controversial topic in the field of second language (L2) acquisition. The non-interface position denies such a possibility whereas the strong and weak interface positions support the opposite. In order to settle this debate, tests to measure L2 implicit knowledge and explicit knowledge separately are indispensable. Nevertheless, the majority of previous studies utilized only tests in which explicit knowledge is favorable. Responding to this drawback, recently, some researchers started to use tests for implicit knowledge; however, the validity of these measurements is usually ignored. To address this gap, Ellis and his colleagues used factor analyses to validate several measurements that they used in order to measure implicit knowledge and explicit knowledge of L2 learners separately. The present study investigated the four tests used in Ellis (2005, 2009b): elicited oral imitation test, timed grammaticality judgment test, untimed grammaticality judgment test, and metalinguistic knowledge test, using Pearson's correlation coefficient and a principal component factor analysis. The target structure was English unreal conditional. Overall, 72 ESL learners participated. The results showed that the present study failed to replicate Ellis' results; the factor analysis in the current study differentiated three factors according to test types, but not knowledge types. The main reason this study failed to replicate Ellis' results was that the participants did not possess enough implicit or explicit knowledge of the target structure to allow for a difference to be found between the two types of knowledge. Furthermore, the grammatical items and the ungrammatical items in the untimed grammaticality judgment test functioned similarly. The correlation coefficient between the score of the metalinguistic knowledge test and that of the untimed grammaticality judgment test was the highest over all, but it did not reach the level of significance.

収録刊行物

詳細情報 詳細情報について

問題の指摘

ページトップへ