Stages in Chinese children's reading of English words.

Abstract

This study examined the developmental stages in Chinese children's reading of English words in 118 English as a second language (ESL) children from Grade 2, 4, and 6 (M age = 7:6, 10:5, and 12:5 years, respectively) from a working-class elementary school in Tian Jin, China. Six ability tests measuring children's level of proficiency in Chinese and English and four experimental tasks exploring children's strategy in reading English words and the ability of making orthographic analogy in both languages were administered. Our hypothesized model was partially supported, and a revised model based on the initial one fit the data nicely. Evidence showed that (a) Chinese children follow similar stages in alphabetic reading development to those in Ehri and McCormick's (1998) phase theory depicting native English-speaking children: the Pre-Alphabetic stage, the Partial-Alphabetic stage, and the Full-Alphabetic stage; (b) The use of orthographic analogy does not form a separate stage independent of the alphabetic decoding stages; rather, it is a concurrent option available to Chinese children from an early age; (c) Similar to Goswami's (1993) interactive analogy model depicting native English-speaking children, Chinese children refine and improve their orthographic analogy-making ability over time, starting from making analogies at the onset-rime level to making analogies across the onset-rime boundary (e.g. coda analogy and middle vowel analogy); (d) Different from Goswami's (1993) findings, Chinese children made better 'onset + vowel' analogies than rime analogies, instead of vise versa. This might be explained by the dominance of open-vowel structures (C+V or C+VV) and rarity of rime structures (V+C or V+CC) in Chinese syllables; (e) Chinese children's English proficiency was highly predictable from Chinese proficiency measures.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 30 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.