Due to their logographic nature, Chinese characters are traditionally learned with a whole-word approach. Starting from kindergarten, students in Hong Kong learn Chinese and English words and attain bilingual literacy in this way. However, this type of learning results in low levels of both Cantonese (L1) and English (L2) phonological awareness (PA). Phonological awareness is not only significant for emergent literacy in an alphabetic language but also for decoding unfamiliar words. For this reason, Hong Kong students have difficulty learning to read English (especially sounding out and spelling unfamiliar words). Meanwhile, Chinese readers in mainland China and other places who learn to read Chinese with a phonetic script are found to have higher phonological awareness of Chinese (Bertelson, Chen & de Gelder, 1997; H. Cheung et al., 2001; Holm & Dodd, 1996; Huang & Hanley, 1995; Read et al., 1986).
In light of the advantage brought about by a native phonetic script, the present research aims to investigate the effect of Cantonese phonological awareness on English learning by comparing the performances of an Experimental Group with phonological awareness training and a Control Group without training, both before and after training.
Two experiments were conducted from 2003 to 2004. In the pilot experiment, the Experimental Group (n = 8) which had received 10 hours of Cantonese Romanisation instruction had no immediate advantage over the control group (n = 9) concerning their post-training English phonological awareness test. However, after the two groups had received 10 hours of English phonics instruction, the Experimental Group a made significantly bigger mean progress than the Control Group (p = 0.035) in their post-training English phonological awareness test. In the main experiment, the Experimental Group (n = 19) which had received 20 hours of Cantonese Romanisation instruction made a bigger mean progress than Control Group 1 (n = 20) and Control Group