4. General Discussion

This study is one of only a few studies that examined the relationship of compound awareness with both word reading and vocabulary development among monolingual Chinese children. We found that Chinese children improved on both compound awareness measures from grade one to grade two. At the beginning of second grade, the percentage of correct was 78% on the compound structure task, and 88% on the compound analogy task. These results demonstrate that compound awareness develops relatively early among Chinese children. More importantly, our results point to the central role of compound awareness in Chinese children’s vocabulary development. Each aspect of compound awareness explained unique variance in expressive vocabulary. In combination the two measures explained 23% of the total variance, far exceeding the contribution of the phonological measures.

Results of this study also reveal that compound awareness is important for Chinese character reading. The compound analogy task was significantly associated with Chinese character reading after controlling for the effects of age, RAN, and phonological awareness. As will be discussed in more detail below, this finding indicates that a level of “deep” compound awareness that requires access to every morpheme facilitates the mapping of morphemes from oral language onto print. Taken together, our study highlights the importance of compound awareness in literacy development among monolingual Chinese children. The results also suggest that, as in other alphabetic languages (Fowler & Liberman, 1995; Nagy et al., 2006), compound awareness is a construct distinct from phonological awareness in Chinese.

4.1 Compound Awareness and Vocabulary Knowledge

We found that the compound structure task, which measured children’s ability to identify the head of a compound noun, was significantly related to Chinese expressive vocabulary. According to Clark et al. (1985), English-speaking children learn the head noun first because it is the label of a basic category. Having acquired the head noun, children can add a modifier to the basic level term to label a subcategory. This method may be particularly effective in Chinese given the fact that Chinese morphemes are highly productive. Once children have acquired the meaning of a head morpheme and understand its role in compound words, they can use this knowledge to comprehend and produce many different compound words. For instance, a child that knows the head morpheme 杯bei (glass, cup) will quickly learn words such as 酒杯jiubei (wine glass), 水杯shuibei (water cup: cup), 茶杯chabei (tea cup), and so on.

Our findings concerning the compound structure task were different from those reported by Wang et al. (2006), where a similar task was not significantly correlated with oral vocabulary. One reason for the difference may be that children in our study were younger than in Wang et al. (2006). As our results showed, the ability to identify the head morpheme develops relatively early and children in the Wang et al. study may have reached ceiling on this task. Another possibility is that the contribution of the compound structure task is affected by children’s linguistic background. Our participants were monolingual children who received Chinese instruction in public schools on a daily basis. Thus, at least among monolingual Chinese-speaking children, the ability to identify the head morpheme is significantly associated with vocabulary development in early grades.

The compound analogy task was also significantly related to Chinese vocabulary development. This task taps the ability to understand the structure and meaning of a compound word as a combination of two or more constituent morphemes. This aspect of morphological awareness is important for Chinese vocabulary development for several reasons. First, a large proportion of Chinese morphemes are highly frequent. When children encounter a new compound word, they are likely to be familiar with some or all of the constituent morphemes. Second, most Chinese compound words are semantically transparent. The meaning of each constituent morpheme is directly related to the meaning of a compound. Therefore, knowledge about compounding rules and about morpheme meaning enables children to produce a large number of compound words with a limited number of familiar morphemes.

The contribution of compound awareness to Chinese vocabulary far exceeded that of phonological awareness. Neither syllable awareness nor rime awareness was a significant predictor of vocabulary when entered into the regression equation before the compound awareness measures, though syllable awareness was moderately correlated with vocabulary. A possible interpretation for lack of unique contribution of syllable awareness is that Chinese has a large number of homophones. One syllable often corresponds to a number of different morphemes and characters. For example, the syllable /xi/ can mean west, stream, rest, hope, sunset, knee, scarce, breathe, and is represented by 西, 溪, 息, 希, 夕, 膝, 稀,吸, respectively. Thus, the role of syllable awareness in Chinese vocabulary acquisition is limited.

The current results indicate that compound awareness is significantly associated with vocabulary development among Chinese monolingual children. This association is likely to be bidirectional in nature (e.g. McBride-Chang, Shu, Ng, Meng & Penney, 2007; Nagy et al., 2003). Vocabulary knowledge provides the basis for developing abstract understanding of morphemes and compounding rules, as early language develops primarily from whole to parts (McBride-Chang et al., 2007). According to Gombert’s (1992) metalinguistic development framework, we hypothesize that compound awareness develops in three stages. Children first begin to demonstrate functional control of the language in rich pragmatic contexts. For example, Clark (1995) observed that even two-year olds were able to create novel compounds to express meaning, e.g. crayoner for a person who uses crayons. This preliminary awareness evolves into actual awareness, where children can think about morphemes and compounding rules in somewhat abstract terms. Eventually, children reach a conscious awareness stage, where they are able to deliberately manipulate morphemes to form compound words. Because the insights into meaning and word formation processes facilitate vocabulary acquisition, the association between compound awareness and vocabulary likely becomes increasingly reciprocal as children’s language and literacy skills are advanced. McBride-Chang, Tardif, Cho, Shu, Fletcher, et al. (2008) demonstrated the reciprocal relationship in a recent study. They found that for Chinese four-year olds, morphological awareness predicted vocabulary knowledge a year later, and vocabulary knowledge also predicted subsequent morphological awareness. Future research needs to examine the relationship among older students.

4.2 Compound awareness and Chinese character reading

Our results showed that the compound analogy task, but not the compound structure task, was a significant predictor of Chinese character reading. The compound analogy task demands conscious access to every morpheme in a word. For example, for the question, 斑马是身上有斑纹的一种马，那么身上有斑纹的牛我们叫什么？[Striped horse (Zebra) is a kind of horse with stripes on the body. What should we call a cow with stripes on the body?], the desired answer is “斑牛” (striped cow). To construct this novel compound, children need to extract the morpheme “斑”ban (stripe) from the word “斑纹”banwen (spot stripe: meaning stripe) and combine it with another morpheme 牛niu (cow). The significant association between the compound analogy task and character reading indicates that children with heightened sensitivity to morphemes in the oral language have a better chance to succeed in learning to read, as they can map morphemes onto characters more effectively. The compound structure task, in contrast, focuses children’s attention on the head morpheme only. It does not necessarily require access to other morphemes in a compound word. Thus, only tasks that measure “deep” compound awareness that require access to every morpheme are significantly related to character reading.

The relation between compound awareness and Chinese character reading is an interesting one. In alphabetic languages, morphological awareness facilitates word reading by providing insights into the structure of morphologically complex words (e.g. Nagy et al., 2003, 2006). Insights into the structure of compound words, however, may only have a limited impact on Chinese character reading as most characters represent single morphemes. This may be the reason why the compound structure task, which tapped one aspect of structure awareness, was not a significant predictor of character reading. Taken together, our findings indicate that the two levels of compound awareness, structure and meaning, contribute differently to character reading and vocabulary development. While both levels predict vocabulary development, only awareness of morpheme meaning is significantly associated with character reading. In this regard it is useful to note that most previous studies conducted in Chinese used a dependent measure that required children both to read single characters and to read (or to produce) two-character compound words (McBride-Chang et al., 2003, 2005; Shu et al., 2006). As a result, those studies did not distinguish the effects of morphological awareness on character reading from its effects on word reading and vocabulary. Our study is the first to demonstrate that there is a specific link between compound awareness and Chinese character reading.

The syllable awareness task was the only phonological awareness measure that significantly predicted Chinese character reading. Syllable awareness is more sensitive to individual differences in reading ability than other aspects of phonological awareness as each Chinese character corresponds to a syllable in the oral language. This finding replicated the findings of previous studies among Hong Kong children regarding the role of syllable awareness in Chinese character reading (McBride-Chang, Bialystok, Chong & Li, 2004; McBride-Chang et al., 2005) and extended it to a population of monolingual Chinese children. The two compound awareness tasks jointly explained 7% of the variance in character reading, and the variance predicted by the two phonological awareness measures was about the same. Taken together, these results reveal that both phonological awareness and compound awareness are important for character reading.