Second Language Learners’ Orthographic Learning in English

Second Language Learners’ Orthographic Learning in English

First Author: Anne E. Cunningham -- University of California - Berkeley
Additional authors/chairs: 
Yi-Jui Chen University of California, Berkeley
Keywords: Orthographic Knowledge, Self-teaching, E2L, Word Learning
Abstract / Summary: 

Two questions were answered in this study: (a) Do second language learners in Taiwan acquire orthographic representations via the self-teaching mechanism that Share (1995) has proposed? (b) Will imperfect decoding of a target word still lead to orthographic learning?

Thirty elementary school students (Grade 3-6) were assessed using the Elision and Blending subtests of the CTOPP, the Word Attack and Letter Word Identification subtests of the WJ-III, the receptive vocabulary subtest of the PPVT, and a self-designed computer-based orthographic processing assessment. Then, the participants engaged in self-teaching experiments modeled on Cunningham’s (2006) design.

A majority of the students were more likely to choose the correct spelling in the orthographic choice task, both immediately and two days after the experiment, demonstrating orthographic learning. However, accuracy in spelling target words varied widely from 13% to 41%, depending on the target word. The effect of the two-day delay was evident for both the orthographic choice task and the spelling task.

Multiple regression analysis demonstrated that decoding ability predicted students’ orthographic learning at the 5% significance level after controlling for prior orthographic knowledge, phonological processing, orthographic processing, receptive vocabulary, and letter and word identification. Interestingly, item-based analyses demonstrated that the imperfect decoding of a target word may nonetheless still lead to correctly selecting or spelling target words.

Results indicated that, although students lacked sufficient English receptive vocabulary and decoding experience, decoding was still a mechanism by which they acquired orthographic representations, consistent with Bowey and Miller (2005), Cunningham (2006), and Share (1999). Moreover, an important and unique element of this study is that imperfect decoding can still lead to orthographic learning (as found in Nation, Angell and Castle (2007).