Chinese Students Learning English as a Second Language

Abstract

The way in which different cities teach children to read in Chinese may have an impact on the skills they later utilize to acquire English word reading skills. This study examined the relative contributions of several cognitive–linguistic measures to English word reading for Chinese students learning English as a second language in two Chinese cities, one whose school system teaches Pinyin (Beijing) and one whose school system does not teach Pinyin (Hong Kong). Students in grades 2–3 completed measures on Chinese morphological awareness (MA), Chinese phonological awareness (PA), Pinyin writing, and English word reading. In the Beijing group, it was found that PA (β = 0.334, p < 0.01) and Pinyin (β = 0.257, p < 0.05) were significant predictors of English word reading. In contrast, in the Hong Kong group, only MA (β = 0.263, p < 0.05) was found to be a significant predictor of English word reading. The difference in predictors could be due to the availability of a phonological tool (Pinyin) for the Beijing students when learning Chinese, while the Hong Kong group may have relied more heavily on learning using MA and rote memory techniques. Overall, the results from this study provide data supporting the benefits of having a phonological tool like Pinyin for Chinese children when learning to read in English.

1. Introduction

The purpose of this study was to examine the phonological and morphological transfer from Chinese to English word reading in two cities with different approaches to Chinese reading instruction. Our measures of phonological skill included both phonological awareness in Chinese and Pinyin knowledge. Pinyin is the official Chinese phonetic alphabet. This Romanized system is used as a teaching tool to aid in the learning of Mandarin. In Mainland China, Pinyin allows children to make the connection between their oral language and Chinese characters via a consistent phonological code [1].

There are differences in approaches to reading training in Hong Kong as compared with Beijing. In Hong Kong, children typically learn to speak and read words in Cantonese (one Chinese language). There has been a trend toward a rote memorization approach to learning to read Chinese words in Cantonese, a so-called “look and say” method adopted both by teachers and parents in their focus on Chinese (e.g., [2]). Pinyin is not typically used as an aid for reading intervention in Cantonese. Moreover, Pinyin represents Mandarin, rather than Cantonese, so it is a coding system that is used as a second language (L2) tool in Hong Kong if it is used at all. In contrast, Pinyin is a common tool for Chinese character reading in Beijing.

Pinyin instruction has been shown to be beneficial in the longitudinal development of Chinese word reading [3,4]. Pinyin symbols are also based on the Roman alphabet, and there are some similarities between the sounds the letters make across Chinese and English. Given this commonality, Pinyin may not only be associated with Chinese literacy learning but with English word reading. Indeed, in a study of second-grade students enrolled in a Chinese immersion program, Pinyin instruction was found to be correlated with both English word and pseudoword decoding measures [1]. In another study of Chinese sixth graders [5], Pinyin was also uniquely associated with both Chinese and English word reading, with other cognitive–linguistic skills statistically controlled. Thus far, such studies have focused primarily on Mandarin-speaking children in Mainland China. One recent study [6] examined Pinyin in children learning Cantonese as both a first and a second language in relation to spelling and showed some overlaps between Pinyin spelling and English word spelling. In the present study, we explored the association of Pinyin with English word reading in Chinese children from Beijing, where Pinyin is commonly used, and in Chinese children in Hong Kong, where Pinyin is less commonly used. Importantly, Pinyin maps onto Mandarin, the official language of Mainland China. In Hong Kong, Cantonese is used instead, but Mandarin is taught in school in addition to Cantonese. Thus, Mandarin is learned as an additional language, and Pinyin reflects this language in print.

1.1. Similarities between Pinyin Spelling and English Spelling

As a Romanized, alphabetic system, Pinyin shares a number of commonalities with English. The Pinyin writing system uses all 26 letters of the Latin alphabet, the same letters found in the English alphabet. As cited in Xiao et al. 2020, the Institute of Linguistics, Chinese Academy of Social Sciences describes the Pinyin writing system as using 21 onsets, 39 rimes, and four lexical tones (five including the neutral tone) [7]. Zhou and McBride highlighted a number of similarities between the Pinyin writing system and English spelling [6]. For example, many of the Pinyin onsets resemble the consonants found in the English alphabet. Further, a number of the onsets used in Pinyin also have sound representations that are similar to those in English. Additionally, Pinyin employs the use of rimes, using some single vowel phonemes that are also used in the English alphabet [6]. Thus, Pinyin spelling appears to share a number of features present in English spelling.

1.2. Phonological Awareness in Chinese vs. English

Phonological awareness refers to “the awareness of and access to the sound structure of a language” (e.g., [8]). Phonological awareness plays an important role in the development of both English [9] and Chinese [10] word reading. It has been theorized that phonological activation in recognizing printed words may be shared across all writing systems [11], i.e., that there may be a “universal phonological principle”. However, the level to which phonological awareness plays a role in reading development across different languages appears to vary, with some research suggesting that it may be more important in English than in Chinese due to differences in orthography and language [12,13].

Nevertheless, research has indicated that phonological awareness may have a transfer effect across alphabetic languages [14,15], with some research [6,16] indicating that this cross-linguistic transfer may also occur between non-alphabetic and alphabetic languages, namely, Chinese and English. In their study examining the transfer effects of phonological skills in English, Mandarin, and Cantonese among Hong Kong children, Zhou and McBride [6] found that, in both groups, increased phonological skills in oral Cantonese were associated with increased phonological skills in both Mandarin and English. Further, the measures examining phonological sensitivity in Mandarin, Cantonese, and English were all significantly correlated, indicating that there could be a common phonological mechanism across both similar non-alphabetic languages and distinct alphabetic and non-alphabetic languages [6]. In the present study, we tested phonological awareness in Mandarin in both Beijing and Hong Kong.

1.3. Morphological Awareness in Chinese vs. English

Morphological awareness refers to the awareness and understanding of the meaning structure of a given language [8]. Previous research has indicated that morphological awareness is related to reading development in both Chinese and English; however, it may be more relevant in learning to read in Chinese [12]. Additionally, evidence from a four-year longitudinal study examining the roles of morphological and phonological awareness in English word reading for children in grades 2–5 showed that, while morphological awareness contributed significantly to pseudoword reading and reading comprehension, it was rarely a significant contributor to single-word reading [17]. Identifying Chinese morphemes is often more straightforward than in English given that a phonological syllable, a morpheme, and a character almost always represent the same linguistic unit in Chinese [18]. (This is not the case in English, where morphemes can be represented by single phonemes, syllables, or even multisyllable words.)

Nevertheless, research has suggested that there may be a transfer of morphological awareness skills across languages [19,20,21,22]. However, the process of this transfer appears to be influenced by various factors such as the morphological complexity of the languages, language proficiency in both the L1 and L2, and morphological similarities between the L1 and L2 [23]. In the present study, we tested the extent to which L1 Chinese morphological awareness is associated with L2 word reading. Given the complexity of morphology across languages, we were not sure whether L1 Chinese morphological awareness is uniquely associated with L2 English word reading.

1.4. Chinese Morphology and English Word Reading

Although morphological awareness skills appear to have some transfer effect across languages, less is known about how L1 Chinese morphological awareness may impact L2 English word reading skills. One study observed this type of transfer effect in bilingual Chinese–English children from Hong Kong [24]. In their study, researchers found that Chinese morphological awareness could explain 4% of the variance in English word reading in grade 2 bilingual Chinese–English children. All participants from this study were typically developing and did not have any reported learning difficulties [24]. In the present study, we examined L1 Chinese children from both Hong Kong and Beijing to examine whether Chinese morphological awareness is uniquely associated with L2 English word reading in both children with and without dyslexia.

1.5. Dyslexia in Chinese Children

Dyslexia is diagnosed differently in Beijing and Hong Kong. The present study did not focus on dyslexia per se, but some of the children from each city had been diagnosed with dyslexia. By definition, these children had lower Chinese character and word reading scores. We included these groups because they contributed to increased variability in literacy scores.

1.6. Education Systems in Beijing and Hong Kong

The education systems in Beijing and Hong Kong each teach both Chinese and English in primary school. In Beijing, Mandarin is the medium of instruction in schools and English is presented formally in grade 3, as directed by the Ministry of Education of the People’s Republic of China [25]. However, some schools implement English instruction as early as grade 1 [26]. In Hong Kong, the medium of instruction has traditionally been Cantonese, with Mandarin taught as an additional subject, like English [27]. In Hong Kong, English is typically taught in kindergarten, around age four or five, with the Education Bureau of the Government of the Hong Kong Special Administrative Region stressing the idea of “biliteracy and trilingualism” [28].

1.7. Present Study

The present study aimed to assess whether primary school children from two Chinese cities, one which teaches Pinyin routinely (Beijing) and one that often does not use Pinyin for literacy instruction (Hong Kong), showed differences in the skills that contribute to English word reading. Phonological awareness, an oral metalinguistic skill, was also tested in both places. In addition, we tested whether morphological awareness in Chinese uniquely predicted English word reading in each city. We hypothesized that Pinyin and Chinese phonological awareness would be stronger correlates of English word reading in children from Beijing, where Pinyin is used, than in children from Hong Kong, where Pinyin is less frequently used. Further, we hypothesized that morphological awareness would contribute more to English word reading in children from Hong Kong than in children from Beijing due to the lack of a phonological tool, namely, Pinyin.

2. Materials and Methods

2.1. Participants

As part of a larger study, children were recruited across primary schools. The participants represented in the present study were in grades 2–3 from Beijing and Hong Kong. Some of the children in each group had been diagnosed as dyslexic. Participants from Hong Kong who were diagnosed with dyslexia had been diagnosed by either educational or clinical psychologists using a standardized test for diagnosis of dyslexia in Hong Kong, the Hong Kong Test of Specific Learning Difficulties in Reading and Writing for Primary School Students—Third Edition (HKT-P(III)) [29]. Participants from Beijing with dyslexia had been diagnosed using methods similar to those used by Meng et al. [30].

In Hong Kong, 91 Cantonese-speaking children (49 boys) in grades 2–3, with a mean age of 8.03 years old (SD = 0.67, range = 7.00–10.00), were recruited via local primary schools. Thirty-three children in the sample were reported to have dyslexia. Data collection for participants in Hong Kong took place between two recruitment periods, November 2019 to January 2020 and November 2020 to April 2021.

In Beijing, 64 Mandarin-speaking children (40 boys) in grades 2–3, with a mean age of 8.38 years old (SD = 0.49, range = 7.47–9.19), were recruited via local primary schools. Thirty-nine children in the sample were reported to have dyslexia. Data collection for participants in Beijing took place between October and November of 2019.

2.2. Procedure

Once parental consent was obtained, children were individually assessed with a battery of educational tests. Assessments of morphological awareness were conducted in Mandarin in Beijing and Cantonese in Hong Kong. Assessments of phonological awareness were administered in Mandarin for participants in both Hong Kong and Beijing, with instruction given in Cantonese for participants in Hong Kong. Our Pinyin Writing Task was administered in Mandarin for both groups of participants, with a prepared recording being played for individuals in Hong Kong. The English word reading tasks were administered via a paper-and-pencil format in Beijing, with instructions explained in Mandarin, while in Hong Kong the tasks were performed via the use of a web platform prepared by the team, with instructions explained in Cantonese. Participants in Beijing completed assessments in person at school, while participants in Hong Kong completed the assessments online via Zoom due to the onset of the COVID-19 pandemic.

2.3. Measures

Morphological Awareness. The 46-item measure included 27 morphological construction test items and 19 compounding production test items. Both item subsets included two practice items to start. The items were presented orally. For the morphological construction items, children were asked to create a new object or concept based on a description of a scenario of an object or concept for an object. For example, “用花生嚟做嘅油，我哋會叫佢做花生油。咁用冬菇嚟做嘅油，我哋會點叫佢呀?” (The oil that is made of peanuts, we call peanut oil. What would we call the oil that is made of mushrooms?) The correct answer for this item would be, “冬菇油” (mushroom oil). For the compounding production items, children were presented with an open-ended question and were tasked with coming up with the correct compound word. For example, “我地叫專門吃鐵既怪獸做咩”? (What should we call monsters that only eat iron?) The correct answer for this item would be, “吃鐵怪” (iron-eating monsters). This measure was adapted from Liu and McBride [31] and McBride et al. [32] to ensure the items presented were suitable for the participants in this study; 1 point was given for each correct item on the morphological awareness task, with a maximum possible score of 46 for the full measure. In the current study, the obtained Cronbach’s alpha was 0.71 for the Hong Kong group and 0.91 for the Beijing group.

Phonological Awareness. The 13-item measure included two sections, onset detection and rime detection. Both sections included two practice items to start. The onset detection section included seven test items and the rime detection section included six test items. In both sections, children were orally presented with three different syllables and asked to select the one that had a different onset or rime from the other two. This measure was adapted from a previous study [33]. The measure had a maximum possible score of 13, with each correct item being allotted 1 point. In the current study, the obtained Cronbach’s alpha was 0.51 for the Hong Kong group and 0.74 for the Beijing group.

Pinyin Writing. The 22-item measure was adapted from the invented Pinyin writing task used in previous research [34]. The measure includes 11 single-syllable items and 11 two-syllable items. Among the single-syllable items, there were eight real characters and three pseudocharacters. Among the two-syllable items, there were five real words and six pseudowords. Participants were asked to write out the Pinyin that matched each syllable presented. Scoring was determined by four linguistic features, onsets, rimes, tones, and order [35], with each syllable having a maximum possible score of 12. Onset and rime were scored on a 5-point scale (0–4), lexical tone was scored on a 4-point scale (0–3), and order of onset and rime was scored on a 2-point scale (0–1). The measure had a maximum possible score of 396. In the current study, the obtained Cronbach’s alpha was 0.98 for the Hong Kong group and 0.87 for the Beijing group.

English Word Reading. The two-part measure consisted of an individual silent English word reading task and an English silent contextual reading task [36]. In both parts of the measures, participants were given two practice trials before beginning. The strings of English words for both tasks were presented without spaces in between and in order of progressive difficulty. The total measure included 45 items. The individual silent English word reading task consisted of 25 chains of unrelated words. The English silent contextual reading task consisted of 20 sentences in a word chain. Each chain/sentence had between 10 and 20 words. Participants were instructed to draw a line between each distinct word in both tasks within a four-minute timeframe; 1 point was given for each correctly identified word, with a maximum possible score of 522 for the full measure. In the current study, the obtained Cronbach’s alpha was 0.98 for the Hong Kong group and 0.87 for the Beijing group.

2.4. Data Analyses

Statistical analyses were performed in SPSS version 29. Two (enter method) multiple linear regression models for the Beijing and Hong Kong group were conducted. Assumptions of absence of multicollinearity and outliers, linearity, homoscedasticity, and multivariate normality were checked prior to data analyses. To evaluate differences in the Beijing and Hong Kong groups on cognitive–linguistic measures, an independent samples t-test was performed.

3. Results

Table 1. Descriptive Statistics and Correlations for Cognitive–Linguistic Measures.

Measure	n	M	SD	Correlations
				1	2	3	4
Beijing
1. Average of English Silent Word Reading	64	37.19	23.62	-	-	-	-
2. Sum of Chinese Morphological Awareness	64	30.59	5.99	0.212 *	-	-	-
3. Sum of Chinese Phonological Awareness	64	9.38	2.83	0.431 **	0.203	-	-
4. Sum of Pinyin Writing	64	343.17	52.03	0.365 **	0.250 *	0.234 *	-
Hong Kong
1. Average of English Silent Word Reading	83	43.98	26.94	-	-	-	-
2. Sum of Chinese Morphological Awareness	83	22.29	6.48	0.362 **	-	-	-
3. Sum of Chinese Phonological Awareness	83	6.34	2.67	0.322 **	0.423 **	-	-
4. Sum of Pinyin Writing	83	64.39	93.94	0.206 *	0.431 **	0.443 **	-

* p < 0.05; ** p < 0.01.

shows the descriptive statistics and correlations for the cognitive–linguistic measures for both Hong Kong and Beijing in the areas of Chinese morphological awareness, Chinese phonological awareness, Pinyin writing, and English word reading.

Table 2. Differences between Beijing and Hong Kong on Cognitive–Linguistic Measures.

Measure	Beijing M (SD)	Hong Kong M (SD)	t	df	p	d
Average of English Silent Word Reading	37.19 (23.62)	43.67 (26.91)	1.54	149	0.126	0.25
Sum of Chinese Morphological Awareness	30.59 (5.99)	22.31 (6.39)	−8.140	152	<0.001	1.33
Sum of Chinese Phonological Awareness	9.38 (2.83)	6.34 (2.66)	−6.73	148	<0.001	1.11
Sum of Pinyin Writing	343.17 (52.03)	63.94 (91.91)	−23.65	141	<0.001	3.60

presents the results from the independent samples t-test that was used to evaluate differences between the Beijing and Hong Kong groups in the areas of Chinese morphological awareness, Chinese phonological awareness, Pinyin writing, and English word reading. The children in Beijing scored significantly higher on morphological awareness, phonological awareness, and Pinyin knowledge. There were large effect sizes for Chinese morphological awareness, Chinese phonological awareness, and Pinyin knowledge and a small effect size for English silent word reading. However, the two groups did not differ in English word reading scores.

Multiple linear regression was used to assess whether various cognitive–linguistic assessments and the presence of dyslexia significantly predicted English word reading in second and third graders in Hong Kong and Beijing.

Table 3. Regression Coefficients of Chinese Morphological Awareness, Chinese Phonological Awareness, Pinyin Writing, and Dyslexia on English Word Reading.

Variables	B	SE	t	p	95% CI
					LL	UL
Beijing
Constant	−32.84	21.25	−1.55	0.128	−75.36	9.69
Sum of Chinese Morphological Awareness	0.25	0.46	0.54	0.593	−0.67	1.16
Sum of Chinese Phonological Awareness	2.78	0.97	2.86	0.006	0.84	4.73
Sum of Pinyin	0.12	0.05	2.21	0.031	0.01	0.22
Dyslexia	−5.92	5.46	−1.08	0.691	−16.84	5.01
Hong Kong
Constant	7.60	12.42	0.61	0.542	−17.12	32.32
Sum of Chinese Morphological Awareness	1.10	.52	2.12	0.037	0.07	2.12
Sum of Chinese Phonological Awareness	2.10	1.21	1.73	0.088	−0.32	4.51
Sum of Pinyin	−0.01	0.04	−0.13	0.897	−0.08	0.07
Dyslexia	−2.54	6.36	−0.40	0.691	−15.19	10.12

Note: CI: Confidence Interval, LL: Lower limit, UL: Upper limit.

shows the regression coefficients for these variables in Beijing and Hong Kong. Because this was a larger study that compared those with and without dyslexia in other analyses, we included the grouping of dyslexia vs. typically developing in the analyses. The overall model explained 28% of the variation in English word reading in Beijing and 17% of the variation in Hong Kong. The fitted regression model was significant in explaining English word reading for students in both Beijing (R² = 0.279, F(4,59) = 5.71, p < 0.001) and Hong Kong (R² = 0.167, F(4, 78) = 3.92, p = 0.006).

The individual predictors were examined further and indicated that, in Hong Kong, Chinese morphological awareness was the only significant predictor of English word reading (β = 0.263, p = 0.037). Chinese phonological awareness (β = 0.208, p = 0.088), Pinyin writing (β = −0.016, p = 0.897), and dyslexia (β = −0.046, p = 0.691) were not significant predictors of English word reading.

In Beijing, the individual predictors were examined and indicated that both Pinyin writing (β = 0.257, p = 0.031) and Chinese phonological awareness (β = 0.334, p = 0.006) were significant predictors of English word reading. Chinese morphological awareness (β = 0.062, p = 0.593) and dyslexia (β = −0.123, p = 0.283) were not significant predictors of English word reading for students in Beijing.

4. Discussion

In the present study, we compared phonological skills and morphological awareness in relation to English word reading across Beijing and Hong Kong. Because Pinyin is a phonological coding tool focusing children’s attention on phonological properties of reading in Chinese, we hypothesized that the associations among Pinyin, phonological awareness, and English word reading might be stronger in Beijing than in Hong Kong. In this study, the children from Beijing had higher scores on Chinese morphological awareness, Chinese phonological awareness, and Pinyin writing. However, there was no significant difference in English word reading between the children from Beijing and Hong Kong. In addition, we found that both Pinyin and phonological awareness were uniquely associated with English word reading in Beijing; morphological awareness was not. In contrast, in Hong Kong, only morphological awareness was uniquely associated with English word reading. The results likely reflect a stronger reliance on phonological coding in Beijing than in Hong Kong, with some transfer of this coding from L1 Chinese to L2 English word reading. Moreover, morphological awareness has been shown to be an important tool in learning to read in Chinese. Thus, it is possible that in the absence of an additional phonological tool, Hong Kong Chinese children learning to read English rely more on morphological awareness when encountering a new language and script.

The relative contributions of these cognitive–linguistic skills to English word reading suggest that the availability of a phonological tool, such as Pinyin, may benefit Chinese children in learning to read English words. The Beijing–Hong Kong comparison highlights how a phonological coding tool such as Pinyin may influence early word recognition even in a foreign language. Pinyin is a transparent, easy-to-use coding system that might enhance early word recognition in English, a less transparent script.

The presence or absence of dyslexia was not a significant predictor of English word reading in either Beijing or Hong Kong. We included the status of dyslexia in our analyses to increase our sample size and variability in word reading. However, our primary focus was on comparisons of phonological and morphological skills for reading. The evidence for phonological skills as the primary correlate of dyslexia in Chinese has been mixed [37]. It is more likely that dyslexia in Chinese is determined by multiple cognitive–linguistic skills. Given that English word reading in Chinese children represents a complicated task, influenced by multiple factors, the fact that the status of dyslexia was not uniquely associated with English word reading is not surprising [38].

There were clear limitations in this study. For example, our sample sizes were small, and our range of metalinguistic measures included was limited. Also, it may not be “fair” to compare Pinyin in Beijing and Hong Kong given that this measure represents an L1 skill in Beijing and L2 skill in Hong Kong. The percentage of variance explained in English word reading was relatively small in each group. There were also some differences in testing, with children in Hong Kong sometimes being tested remotely. Additionally, the phonological awareness measure had a relatively low reliability in Hong Kong. It is unclear as to why this was the case. Therefore, contributions of the phonological awareness construct to English word reading in Hong Kong should be cautiously interpreted.

Despite these limitations, we view these results as reflecting an interesting phenomenon. Chinese children who are taught to read with more of an emphasis on phonological coding tend to make greater use of this coding even when reading in a second language. It appears that those with less such training may make less use of phonological coding and rely instead on alternative strategies such as morphological awareness transfer. While the correlations of English word reading and phonological and morphological skills across these two groups did not appear to differ from one another statistically, regression analyses reflected a stronger relative association of morphological awareness with English word reading (e.g., r = 0.36 for Hong Kong and r = 0.21 for Beijing) and with the phonological measures in Hong Kong. A smaller percentage of variance in English word reading could be explained in Hong Kong as compared with Beijing as well. Such differences likely reflect the results of different approaches to teaching reading of Chinese and the transfer of such learning to L2 English word reading.