Brief ReportThe role of metalinguistic awareness and character properties in early Chinese reading
Introduction
Learning to read characters is the first step toward literacy for Chinese children. In previous studies, factors at the participant level (e.g., morphological awareness) and the item level (i.e., character properties) were found to be important in learning to read Chinese characters (e.g., Liu, Li, & Wong, 2017; Pan et al., 2016). Less known are the interactions between the two levels of factors on character reading. In this study, we examined the effects of these factors and their interactions on character reading in early Chinese readers.
At the participant level, morphological awareness (MA) and phonological awareness (PA) are two important aspects of metalinguistic awareness in learning to read Chinese. MA includes the knowledge of word structure and the ability to use the structural knowledge in word analysis and formation (Carlisle, 1995). It helps children to understand word meaning through the knowledge of morphemes and word structures, which contributes to their reading ability (e.g., McBride-Chang, Wat, Shu, Zhou, & Wagner, 2003). PA is the ability to manipulate phonological components of words (e.g., syllables) (Oakhill & Kyle, 2000). In Chinese, one character corresponds to one syllable, and a character is generally considered as a morpheme. Such prominence of the syllable in this language might be one of the reasons why the ability to differentiate and manipulate syllables (i.e., syllable awareness) is important in both language development and literacy learning (e.g., Shu, Peng, & McBride-Chang, 2008).
Factors at the item level (i.e., character properties) also exert influence on character learning. One of these is character family size, that is, the number of words that contain a given character. For example, the character 海 (ocean) is in many words such as 海盗 (ocean robber; pirate) and海军 (ocean army; navy), and the number of all words that contain the character is the family size of 海 (ocean). Because characters are mainly embedded in words in Chinese, they absorb the contextual information from the family words (e.g., word meaning and meanings of the neighboring morphemes), which provide additional cues for character learning. For example, when learning the character 海 (ocean), its family words 海盗 (pirate) and 海军 (navy) can be used as memory cues for children to solidify the meaning of the character. Because the family size of the character indicates the extent of this information, it has been found to predict character reading in Chinese children (Wang & Liu, 2020). Another important item-level factor is character frequency. Children start literacy learning with high-frequency characters, and they encounter more high-frequency characters in daily life, so the lexical quality of these characters is higher (Perfetti, 2007), which in turn makes it easier to recognize them.
When a child is learning a new character, the characteristics of the child and the character will interact with each other. For example, Wang and Liu (2020) found that as family size decreased, the level of MA had a larger influence on character reading in lower-grade students from primary school. This pattern suggests that learning to read characters with small family sizes might rely heavily on MA to use the limited semantic information, thereby making the level of MA more relevant. On the other hand, characters with large family sizes are rich in semantic information from family words, which could make them easier to acquire irrespective of a child’s MA level (Wang & Liu, 2020). In sum, when the character property is less desirable (i.e., small family size), children’s levels of MA become more important in learning to read characters.
The participants in Wang and Liu (2020) study were lower-grade students in primary school. It was still unknown whether the same mode of interaction effect exists in younger children. Because younger children have lower levels of MA and less language and literacy experience than older children, they might not be able to use MA to extract the limited contextual information for small family size characters or might need extra help from metalinguistic awareness even when reading characters with favorable properties (e.g., high frequencies). If this is true, in contrast to the compensatory interaction observed in older children (i.e., MA exerts a larger effect on the characters with small family sizes), the interaction in younger children could be nonsignificant or synergistic (e.g., MA works better when the character properties are also favorable). Examining the interactions between the two levels of variables in different developmental phases allows us to observe and compare these relationship patterns so as to clarify the developmental mechanisms of character learning.
MA and PA were chosen as the participant-level factors in this study. MA and PA (in the form of syllable awareness) pertain to the ability to manipulate morphemes and syllables, which are the two types of units that map directly to a character in Chinese and thus are important to the processing of the character (Pan et al., 2016). A lexical compounding awareness task and a syllable awareness task were used to tap MA and PA, respectively, fitting the development of our participants (Liu, McBride-Chang, Wong, Shu, & Wong, 2013; Shu et al., 2008).
For item-level variables, character family size and character frequency were included. A dialogue-based corpus, SUBTLEX-CH (Cai & Brysbaert, 2010), was chosen to extract the family size and the character frequency because it was representative of the language and literacy environment for our participants (mostly short, simple sentences).
Kindergarten children from Beijing in Mainland China (Mainland) and one city in Hong Kong (HK) were tracked at two time points. Children from HK received formal literacy instruction in kindergarten, whereas children from Mainland did not, which was found to cause the heightened literacy ability in HK early readers compared with their counterparts from Mainland (Li, Corrie, & Wong, 2008). The longitudinal design with the two samples allowed us to observe possibly different patterns of the learning process due to the variation in development and literacy environment. Two research questions were raised:
- 1.
How will participant-level variables and item-level variables contribute to character reading in early Chinese readers?
- 2.
Will interactions between participant-level and item-level variables be observed in early Chinese readers? If so, what will the pattern be (e.g., compensatory, synergistic)?
Section snippets
Participants and procedure
In Beijing, the original sample was 118 children (Mage = 62 months, SD = 3.49; 60 girls) in the second year of kindergarten (K2) from a mainstream kindergarten. Of these, 99 children (Mage = 74 months, SD = 3.50; 50 girls) participated again 12 months later when they were in the third year of kindergarten (K3). All were native speakers of Mandarin Chinese. The two waves of the tests took place around June each year.
In HK, the original sample was 88 K2 children (Mage = 52 months, SD = 3.29; 51
Results
The descriptive statistics and correlations of all measures for the two samples are shown in Table 1. All measures showed good internal consistency. PA, MA, and character reading were mostly correlated moderately with each other.
The estimated fixed effects in this study are shown in Table 2. For Mainland children, PA predicted character reading when they were in K2 (β = .12, SE = .04, p < .01), whereas MA predicted character reading when they were in K3 and longitudinally (βs = .08, SE
Discussion
To investigate how participant characteristics and character properties interact on character reading among early Chinese readers, kindergarten children from Mainland and HK were tested at K2 and K3. The main effects of participant-level factors and item-level factors on character reading were found in both samples. Specifically, PA predicted reading at K2, whereas MA predicted reading longitudinally and at K3 (for children from Mainland). In addition, synergistic interaction patterns were
Acknowledgment
This research was partially supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (845812), to Dan Lin.
References (22)
- et al.
Bidirectional relationship between visual–spatial skills and Chinese character reading ability in Chinese kindergartners: A cross-lagged analysis
Contemporary Educational Psychology
(2016) - et al.
The relation between phonological awareness and working memory
Journal of Experimental Child Psychology
(2000) - et al.
Package ‘lme4’
Convergence
(2015) - et al.
SUBTLEX-CH: Chinese word and character frequencies based on film subtitles
PLoS One
(2010) Morphological awareness and early reading achievement
Stages of reading development
(1983)- et al.
Early teaching of Chinese literacy skills and later literacy outcomes
Early Child Development and Care
(2008) - et al.
Chinese children’s character recognition: Visuo-orthographic, phonological processing and morphological skills
Journal of Research in Reading
(2012) - et al.
The anatomy of the role of morphological awareness in Chinese character learning: The mediation of vocabulary and semantic radical knowledge and the moderation of morpheme family size
Scientific Studies of Reading
(2017) - et al.
Morphological awareness in Chinese: Unique associations of homophone awareness and lexical compounding to word reading and vocabulary knowledge in Chinese children
Applied Psycholinguistics
(2013)