Bilingual Home Literacy Experiences and Early Biliteracy Development among Chinese–Canadian First Graders

Abstract

This study was designed to examine the role of early bilingual home literacy experiences (HLE) (including parent–child shared reading, parents’ direct teaching in Chinese and English, the availability of books in both languages, and children’s access to digital devices for bilingual learning) in the biliteracy development of 66 Chinese–Canadian first graders during the COVID-19 pandemic. Descriptive analyses reveal that overall, parents report higher engagement in English than in Chinese across the four HLE measures. Parent’s engagement in bilingual HLE differs by gender, SES, and immigration status. Pearson correlational analyses of English reading, decoding, and bilingual oral receptive vocabulary reveal that the four dimensions of HLE are not strongly related to English early literacy skills but are positively related to Chinese receptive vocabulary. Finally, hierarchical regression analyses indicate that the availability of books in Chinese and parent–child shared reading in Chinese are key factors associated with Chinese receptive vocabulary score variance; the amount of time using digital devices is found to be significantly related to English reading comprehension, but not Chinese vocabulary; and parents’ direct teaching is not significant with either English early literacy skills or Chinese receptive vocabulary. These findings have important implications for parental engagement in early bilingual home literacy activities and early literacy instruction in school.

1. Introduction

The number of children who are exposed to more than one language has increased concomitantly with increasing world immigration rates. Even though there was a record decrease in 2020 due to the COVID-19 pandemic, immigration to OECD countries increased by 22% or more since 2021 [1]. The United States and Canada remained the largest recipients of permanent immigrants. From 2016 to 2021, 1.3 million people moved to Canada, making it the country with the largest increase in immigrants over the previous ten years [2]. As a result, there was an increase in language diversity in Canada with 4.6 million (12.7%) speaking a language other than French or English (the official languages). The growth in language diversity necessitates increasing attention to immigrant children’s (including those who came as immigrants and those born to immigrant parents) bi/multilingual and literacy development in their official and heritage languages.

Children’s early home literacy experiences (HLE) have been recognized as a crucial factor associated with literacy skills development in first language or L1 [3,4,5,6,7]. Sénéchal and LeFevre [7,8] focused on two types of home literacy experiences for English monolingual children in Canada: formal and informal home literacy activities. Formal home literacy experiences include parents’ direct teaching of words or print-related concepts such as alphabetic knowledge. Informal literacy experiences are those parent–child shared reading activities that provide informal literacy exposure. These researchers found that formal literacy activities were linked to early literacy skills including reading and decoding skills, while informal literacy activities were related to receptive lexical skills. Other researchers have underscored the importance of informal literacy experiences such as parent–child shared storybook reading (e.g., [6,7,9]), singing rhythms or playing games with children [5], and parents and caregivers’ literacy instruction (e.g., [6,10,11]) in early literacy development.

Past studies also explored the relationship between other types of early HLE in L1 such as access to print with early literacy development, for example, experiences such as the number of books children have (e.g., [3,4,5]), household magazine/newspaper subscriptions (e.g., [4,12]), and library visits (e.g., [13,14]). In a longitudinal study by Georgiou et al. [15], following 172 children from grade one to grade three, the researchers observed that the association between home literacy activities and certain emerging literacy skills changed over time. The number of books at home was correlated with phonological awareness and vocabulary skills in grade one and with reading comprehension in grade three.

An emerging HLE factor drawing increased attention from researchers is digital device usage or media-based literacy activities at home (e.g., [16,17,18]). This is especially the case after the outbreak of COVID-19, which was seen by some researchers as a catalyst for the expansion of learning from classrooms to home, promoting online learning as a supplement to traditional in-person classes [19,20]. Although research recognizing the active role of watching TV programs (e.g., [21,22,23]) has been conducted, less attention has been paid to the usage of other digital devices, such as laptops, smartphones, or tablets. Digital device usage (guided or unguided by parents or guardians) has become a major component of home literacy as families encourage from infancy children’s increasing usage of digital devices [24,25], and a growing number of mobile apps have been designed for early literacy development [26].

Research, however, has not shown conclusively whether and how digital device usage at home affects bilingual children’s early literacy development. While some researchers report no strong correlations (e.g., [16,26]), others, such as Wong [27], argue that digital technologies engage children in various literacy activities and, further, provide independent exploration and creation in literacy learning. Some researchers such as He et al. [28] have found that how and for what purposes children engage in digital device use matters as time playing video games was found to be negatively associated with literacy learning, and may increase the risk of dyslexia.

Early HLEs appear critical to early literacy development. Most studies, however, have focused on monolingual learners. The effects of these experiences on immigrant children’s bilingual development, especially their mainstream (or official) and heritage languages, have yet to be fully explored. We address this gap by examining the role of bilingual home literacy experiences in children’s early biliteracy development among first graders (N = 66) from Chinese immigrant families in Canada. Building on the findings of previous studies, we examine four broad types of early HLE, namely parent–child shared reading, parents’ direct literacy teaching, access to print materials, and access to digital devices. Although the concepts have evolved to include many dimensions, in this study we adopted a “simple view of reading” (e.g., [29,30,31,32,33,34]) to focus on reading comprehension as well as early decoding skills (i.e., word recognition or knowledge of letter–sound relationships) and language (linguistic) comprehension (i.e., the ability to understand spoken language such as oral receptive vocabulary) skills. Researchers have found that in both monolingual and bilingual children, letter and word recognition skills and language comprehension explain a substantial amount of variance in reading comprehension (e.g., [35,36,37,38]). However, most studies have focused on bilingual children’s English development. The association between HLE and children’s minority language abilities, unfortunately, is under-studied. Therefore, our objectives were to understand immigrant children’s early HLE and their bi-literacy development patterns as they transition to formal mainstream schooling and the relationship between their HLE and bi-literacy development. Specifically, the study was guided by the following research questions:

• What are the patterns of Chinese–Canadian first graders’ early bi-literacy achievements?
• What are the patterns of the children’s early bilingual home literacy experiences (HLE)?
• What role does their bilingual HLE play in their early biliteracy development?

2. Home Literacy Experiences and Early Literacy Development in Bilingual Families

While there are similarities between monolingual and bilingual families, there are features unique to bilingual families. First, there is variation in how parents engage in children’s home literacy activities in different languages (e.g., [23,39,40]). Van Steensel [23], by observing a group of 68 minoritized families with diverse home language backgrounds in the US, found that parents actively engaged in home literacy activities in English learning to support their children’s entry into mainstream culture and education. Hammer et al. [41] reported similar findings in a study of 43 mother–child dyads from Puerto Rican immigrant families in the US. In their study, the preschoolers with an average age of three years and eight months were categorized into two main groups: sequential learners, those who learn English after entering preschool, and simultaneous learners, those who learn English and Spanish simultaneously at home. It was found that mothers of simultaneous learners held higher expectations of their children’s literacy achievement, which resulted in richer home literacy experiences in English and, hence, higher English early reading ability. In a recent study of 31 English monolinguals and 50 bilingual third graders (with varying first language backgrounds) in Canada, Peets et al. [42] noted that bilinguals were different from monolinguals in their home literacy environment in that bilingual parents demonstrated a higher frequency of shared reading in English than in their first language, even though they had fewer English children’s books at home and lower English literacy knowledge.

Second, how early HLE affects bilingual children’s literacy development in each language is also different (e.g., [43,44,45]). Research has found that heritage language development is more related to HLE than mainstream language development, especially after children start formal schooling [46,47,48]. For instance, in their study investigating 96 fifth-grade Spanish–English bilinguals, Duursma et al. [21] found that children’s English proficiency was not correlated with any of the home literacy experiences. In contrast, Spanish proficiency was highly correlated with Spanish use at home and, in addition, with parents’ instructional support. Ryan [48] revealed a similar finding after tracking 42 French–English bilingual early elementary children in the US for eleven months: home literacy experiences were more influential in children’s French than in English, as language exposure and printed materials at home were significantly correlated with French vocabulary, but not English vocabulary. It was also found that the different effects were related to parents’ language proficiency [49,50,51,52,53]. Language exposure at home provided by native speakers is more likely to have a positive effect on children’s proficiency. For immigrant families, parents’ language usage of a heritage language is more effective in heritage language development than in mainstream language development [54,55].

Third, bilingual families are highly diverse in their socio-demographic backgrounds including L1, gender, socioeconomic status (SES), and immigration status, which have been found to affect HLE and early literacy development [39]. While gender and SES are not unique to bilingual children, factors such as L1 and immigration statuses are unique to bilingual families. Previous studies established the important role of family SES (often measured by parental educational level and/or family income level) in mediating early reading development in both first and second language [56,57,58,59,60,61]. However, in recent studies situated in Italy, Bonifacci et al. [62] and Cangelosi et al. [57] found that while being bilingual did not influence vocabulary and text-comprehension skills in the age group being tested (9–11 years), SES played a significant role in early development with low-SES (both monolingual and bilingual children) underperforming in all core literacy skills including decoding, reading comprehension, and language comprehension compared to those of high-SES backgrounds. Further, SES was found to affect bilingual groups differently (i.e., Western European, Eastern European, Asian, Middle Eastern, and African bilinguals in Højen et al. [63]), suggesting the need to further understand the role SES plays in Chinese–Canadian children’s HLE and early literacy development.

While research has shown a significant role of SES in HLE and early literacy, studies on gender and immigration status have been inconclusive. While many studies find gender differences favouring girls (e.g., [64,65,66]), some such as Sabra [67] have observed the opposite, in that boys performed higher than girls in reading comprehension in both first and second language. Similarly, studies on the effects of immigration status on reading development have been inconclusive [68,69,70]. While some studies such as De Feyter et al. [71] found that first-generation immigrant children outperformed second-generation and non-immigrant children on most literacy measures including reading, others such as Palacios et al. [69] found that first-generation immigrant children maintained an advantage in early reading achievement from kindergarten to the end of third grade over US-born second-generation students born in the U.S. to foreign-born parents and third-generation students born in the US to US-born parents. However, this immigrant advantage/paradox may not apply to other demographic groups (i.e., different cohort groups in the same country such as the US) [72] or other language groups in other countries such as the Danish context as documented in Højen et al. [63].

In sum, substantial research evidence has documented that while HLE is significant for bilingual children as it is for monolingual children, the nature of HLE and the relationship between HLE and biliteracy development may play out differently. The relationship may also be affected by different socio-demographic factors such as gender, SES, and immigration status. How HLE affects bilinguals’ bi-literacy development needs to be further explored; and much less is known about the relationships between bilingual early literacy skills and the different components of HLE among Chinese–English bilinguals in North America. This study was designed to investigate not only the patterns of Chinese–Canadian bilinguals’ HLE and early biliteracy development but also how their HLE affects their bilingual literacy skills.

3. Method

3.1. Participants

This study was conducted in one of the most culturally and linguistically diverse areas in Vancouver in the province of British Columbia, Canada where more than one million people speak a language other than English and French at home. Two of the most widely spoken languages are Mandarin and Cantonese [2].

Derived from a large longitudinal research project involving over 200 families, this study was designed to observe the effects of HLE on early literacy development as study participants transitioned from home to formal schooling from December 2020 to May 2022. This study was limited to families with children who were first graders at the time of data collection. Sixty-six first graders who predominantly spoke either Mandarin or Cantonese at home and their parents were included in the final analyses. Their average age was 78.24 months. There was a total of 28 boys and 38 girls; 38 (58%) spoke Cantonese as their home language, and 28 spoke Mandarin. Family SES was categorized by the family’s annual household income and the low-income cut-offs (LICOs) value provided by Statistics Canada [73]. The majority (83%) were non-low-SES families. A total of 26 children (40%) were enrolled in English as a second language (ESL) programs in school, and 31 (47%) attended heritage language learning programs (e.g., weekend Chinese schools). Detailed information is shown in

Table 1. Children’s demographic information.

	N	Percent	Mean	SD
Total	66	100
Average age (in months)			78.24	3.65
Gender
Male	28	42.42
Female	38	57.58
Home language
Mandarin	28	42.42
Cantonese	38	57.58
SES
Low-SES	14	21.21
Non-low-SES	52	78.79
Immigration status
Immigrant	11	16.67
Born in Canada	55	83.33
Language program
English as a second language	26	39.39
Heritage language	31	46.97

.

3.2. Measures and Materials

3.2.1. Children’s Sociodemographic Background and HLE

A parent questionnaire, the Alberta Language Development Questionnaire (ALDQ, [54]), consists of 60 items clustered into five main sections: family background information, home literacy environment (HLE), communication with schools, parental ideology and experience about literacy, and family educational ideology. This questionnaire was used to collect socio-demographic information including the focal child’s home language, age, gender, and immigration status, as well as the family’s annual household income. HLE was measured in four dimensions (i.e., number of books, parent–child shared reading, length of digital device usage, and parents’ direct literacy teaching) in both English and Chinese. Eight items using a five-point Likert scale in the HLE section covered the four different HLE dimensions in each language (four for each language).

3.2.2. Early Literacy Skills

Consistent with the “simple view of reading” [30], children’s early literacy skills in English including reading comprehension, decoding (i.e., letter and word recognition), and oral receptive vocabulary were measured. Due to the lack of available reliable instruments for reading and word recognition in Chinese, only Chinese receptive vocabulary skill was assessed in this study.

3.2.3. English Reading Comprehension and Decoding Skills

The reading comprehension and letter and word recognition sub-tests of the Kaufman Test of Educational Achievement third edition (KTEA-3; [74]) were employed to measure English reading abilities. The KTEA-3 is an individually administered battery of tests assessing core academic skills ranging from pre-Kindergarten to grade 12. The test battery, normed in North America, was reported to have appropriate reliability and sufficient construct and concurrent validity in each of the sub-tests.

3.2.4. English Receptive Vocabulary

The Peabody Picture Vocabulary Test fifth edition (PPVT-5; [75]) was employed to measure English receptive vocabulary. PPVT-5 is an “individually administered, norm-referenced instrument that assesses receptive vocabulary” ([75], p. 1). The test was designed for individuals from 2 years 6 months to 90+ years and is a reliable instrument, having an average internal consistency of 0.97, an average alternative form reliability of 0.86, and an average test–retest stability of 0.84 across age groups.

3.2.5. Chinese Receptive Vocabulary

The Peabody Picture Vocabulary Test-Revised (PPVT-R, [76]) was used to measure Chinese receptive vocabulary. Derived from the English PPVT, the PPVT-R was revised by Lu and Liu [76] to assess the vocabulary skills of individuals aged from three to twelve years. It was standardized in 1988 on a group of 886 children who spoke Chinese as the dominant language in Taiwan. It was reported to have over 0.90 split-half and over 0.84 on test–retest reliability, and is widely used in assessments among children of Chinese heritage in North America.

3.3. Procedures

Parents were asked to complete the ALDQ questionnaire at home once they agreed to take part in the study and returned their consent and assent forms. All information was de-identified with an assigned family and child code immediately after data collection.

Early literacy skills in English and Chinese were assessed individually via Zoom by trained bilingual researchers during the COVID-19 pandemic. The KTEA-3 reading comprehension test consists of 30 items. Children were first asked to read simple instructions and to react appropriately (e.g., pointing to the hat or looking at the door). After reading the instructions, children read short passages of increasing difficulty and were asked to answer questions. The test was un-timed and stopped when the examinee answered five consecutive questions incorrectly. In the KTEA-3 letter and word recognition test, which includes 100 items, children started by pointing to the letters they heard, and then they were asked to pronounce the letters shown. Finally, they pronounced words with increasing difficulty, from single-syllable words (e.g., in; he; sure) to multi-syllable words (e.g., front; fifteen; leftovers). No time limit was set; however, the test was discontinued after four consecutive incorrect answers were produced.

The PPVT administration guide was utilized to choose the beginning item based on the examinee’s age. The PPVT-5 consists of 240 English word items and the PPVT-R included 175 Chinese word items. Each item included a word and a board showing four different pictures. The administrator pronounced the word clearly and asked the examinee to indicate the picture describing the word. There was no time limit, but testing was stopped after six consecutive incorrect answers. The same procedures were followed for the tests in English and Chinese.

3.4. Data Analysis

Descriptive analyses based on standard scores of the five early literacy assessment results and raw scores from parents’ HLE questionnaire were conducted using SPSS (Version 28). All raw scores of the early literacy assessments were recorded. An overall reading score was calculated by summarizing raw scores of reading comprehension and letter and word recognition. All reading, decoding, and receptive vocabulary raw scores were converted to standard scores (M = 100, SD = 15) based on the first-grade levels indicated in the test manuals.

The average scores of the four HLE sub-scales (frequency of parent–child reading, daily length of parents’ direct teaching, frequency of digital device use, and the number of books available), along with overall scores, were included. We calculated an overall HLE score in English and Chinese by summarizing the corresponding four sub-scales. In total, there were five HLE scores collected for each language. We calculated Cronbach’s alpha to measure internal consistency. Among the 55 parents who responded to all four English HLE items and 51 parents who responded to all Chinese HLE items, Cronbach’s alpha for English and Chinese HLE was 0.66 and 0.75, respectively, which indicates a moderate internal consistency [77] of the two scales.

Pearson correlation coefficients were computed to measure the linear relationships amongst socio-demographic data, HLE in Chinese and English, and early literacy assessment results. All socio-demographic information (L1, gender, SES, and immigration status) and early literacy assessment results remained in the two analyses.

Finally, we conducted five independent hierarchical regression analyses, in which the five early literacy assessment results were added separately as dependent variables. Children’s socio-demographic characteristics were added as control variables in the first step of each analysis. Then in the second step, we added overall HLE scores, number of books, frequency of parent–child shared reading, frequency of digital device usage, and daily length of direct language teaching as key variables one at a time, to explore how the variable might potentially affect early literacy skills, after controlling for demographic factors.

3.5. Missing Data

Some data attrition occurred, as not all parents and children completed all questionnaire items or assessments. All 66 children completed three English literacy tests, but 4 did not complete the Chinese receptive vocabulary test (

Table 2. Descriptive statistics of early literacy skills.

	Reading Overall			Reading Comprehension			Letter and Word Recognition			English Vocabulary			Chinese Vocabulary
	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD
Total	66	106.42	19.43	66	104.85	18.55	66	107.09	19.05	66	90.88	11.42	62	88.63	18.33
Male	28	104.57	20.62	28	101.86	20.28	28	106.25	20.21	28	87.61	13.24	24	85.00	18.34
Female	38	107.79	18.67	38	107.05	17.11	38	107.71	18.39	38	93.29	9.33	38	90.92	18.18
Low-SES	14	101.71	18.62	14	99.64	17.94	14	103.50	17.33	14	84.64	9.34	14	93.79	23.70
Non-low-SES	52	107.69	19.62	52	106.25	18.63	52	108.06	19.53	52	92.56	11.42	48	87.13	16.44
Immigrants	11	108.91	20.87	11	103.54	22.43	11	112.91	17.07	11	88.91	10.90	11	97.00	19.49
Born in Canada	55	105.93	19.29	55	105.11	17.90	55	105.93	19.35	55	91.27	11.58	51	86.82	17.75

). We calculated final overall scores in English and Chinese HLE by summarizing results only from parents who responded to all four questions, resulting in 55 English HLE scores and 51 Chinese HLE scores (

Table 3. Descriptive statistics of home literacy experiences in English and Chinese.

	Number of Books			Shared Reading			Digital Device Usage			Direct Teaching			Overall			Cronbach’s α
	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD
HLE in English
Total	63	4.24	1.20	65	4.54	1.23	61	2.26	1.03	60	1.50	0.95	55	13.04	2.10	0.66
Male	26	3.92	1.41	28	4.25	1.51	27	2.48	1.19	24	1.92	1.32	22	13.00	3.07
Female	37	4.46	0.99	37	4.76	0.93	34	2.09	0.87	36	1.22	0.42	33	13.06	1.12
Low-SES	13	4.00	1.22	14	4.64	0.93	14	2.57	0.85	12	1.58	0.90	11	13.18	1.47
Non-low-SES	50	4.30	1.20	51	4.51	1.30	47	2.17	1.07	48	1.48	0.97	44	13.00	2.24
Immigrant	10	4.40	1.26	11	4.64	1.21	10	2.40	0.84	9	1.33	0.50	8	13.75	1.28
Born in Canada	53	4.21	1.20	54	4.52	1.24	51	2.24	1.07	51	1.53	1.01	47	12.92	2.20
HLE in Chinese
Total	64	3.25	1.65	60	3.98	1.57	61	1.36	0.55	62	1.34	0.65	51	10.26	2.98	0.75
Male	26	2.81	1.77	25	3.72	1.74	27	1.41	0.64	28	1.54	0.79	22	9.77	3.45
Female	38	3.55	1.52	35	4.17	1.42	34	1.32	0.47	34	1.18	0.46	29	10.62	2.57
Low-SES	13	3.08	1.80	12	3.58	1.83	14	1.43	0.65	11	1.57	0.85	11	9.82	3.63
Non-low-SES	51	3.29	1.63	48	4.08	1.50	47	1.34	0.52	48	1.27	0.57	40	10.38	2.82
Immigrant	10	3.70	1.57	9	3.89	1.69	10	1.60	0.84	10	1.60	0.84	6	12.17	3.31
Born in Canada	54	3.17	1.67	51	4.00	1.56	51	1.31	0.47	52	1.29	0.61	45	10.00	2.88

). We retained all parental responses if at least one question was answered. In hierarchical regression analyses, list-wise cases were excluded so that only those who had a complete set of data were analyzed in each regression model.

4. Results

4.1. Descriptive Statistics of Early Literacy Achievement Patterns

First, we computed descriptive statistics of early literacy achievement to address the first research question. On average, first graders score 106.42 (SD = 19.43) in reading overall, 104.85 (SD = 18.55) in reading comprehension, 107.09 (SD = 19.05) in letter and word recognition, 90.88 (SD = 11.42) in English receptive vocabulary, and 88.63 (SD = 18.33) in Chinese receptive vocabulary (Table 2). The first-grade children achieve age-group norms in reading overall, reading comprehension, and letter and word recognition tests, but slightly lower than the norms in English and Chinese receptive vocabulary. Girls consistently score higher than boys in all five tests. Children from low-SES families have lower average scores on all four English literacy tests, but they (M = 93.79, SD = 23.70) outperform their non-low-SES peers (M = 87.13, SD = 16.44) in Chinese receptive vocabulary. The difference between immigrant children and children born in Canada is more complex than other sub-categories: immigrants have higher average scores in letter and word recognition and Chinese receptive vocabulary, but lower average scores in reading comprehension and English receptive vocabulary.

4.2. Descriptive Statistics of Home Literacy Experiences

To answer the second research question about the patterns of early bilingual home literacy experiences, we calculated HLE scales. On average, parents rate higher in their English HLE than in their Chinese HLE across all four sub-scales (Table 3), resulting in a higher average score in overall HLE in English (M = 12.86, SD = 2.26) than in Chinese (M = 10.26, SD = 2.98). In the sub-categories, parents provide more books and more frequently engage in shared reading with girls in both languages but provide more direct language teaching to boys and report more daily digital device usage for boys in both languages. Additionally, we find that parents from low-SES families have a higher average score in overall English HLE (M = 13.18, SD = 1.47) than parents from non-low-SES families (M = 13.00, SD = 2.24). They have higher average scores in all the English HLE sub-scales except the number of books at home. Finally, parents with immigrant children have higher overall scores in HLE in both languages (English: M = 13.75, SD = 1.28; Chinese: M = 12.17, SD = 3.31) than parents with children born in Canada (English: M = 12.92, SD = 2.20; Chinese: M = 10.00, SD = 2.88). Both groups have higher average HLE overall scores in English than in Chinese.

4.3. Correlation Analyses

To address the third research question on the role HLE in early biliteracy development, we computed correlations between HLE and early biliteracy achievement. The analyses results (

Table 4. Correlations of socio-demographic factors, home literacy environment, and early literacy skills.

			Socio-Demographic Factors				HLE					Literacy Skills
			1	2	3	4	5	6	7	8	9	10	11	12	13	14
Socio-demographic factors	Age	1	-	−0.18	−0.08	0.02	0.02	0.00	−0.21	−0.04	−0.12	0.06	0.10	0.01	−0.03	−0.22
	Gender	2		-	0.06	0.17 *	0.16	0.10	0.14	−0.05	−0.36 **	0.08	0.14	0.04	0.25 *	0.16
	SES	3			-	0.16 *	−0.18	0.14	−0.06	−0.14	−0.05	0.31 *	0.34 *	0.27 *	0.40 **	−0.07
	IS	4				-	0.08	0.09	0.01	−0.03	−0.02	−0.06	0.03	−0.14	0.08	−0.21
HLE	Overall	5	0.06	0.03	−0.02	−0.07	-	0.45 **	0.63 **	0.20 *	0.24 **	−0.07	−0.04	−0.08	0.03	0.09
	NB	6	0.07	0.09	0.15	−0.04	0.80 **	-	0.20 *	0.01	0.01	0.02	0.04	0.02	0.17	0.37 **
	SR	7	−0.21	0.04	−0.03	−0.05	0.72 **	0.42 **	-	−0.21 *	0.06	−0.05	−0.04	−0.04	0.08	0.09
	DD	8	0.02	0.02	−0.05	−0.01	0.57 **	0.24 **	0.21 *	-	0.31 **	0.16	0.19	0.09	0.12	−0.01
	DT	9	−0.01	−0.28 *	−0.33 **	−0.06	0.51 **	0.23 **	0.10	0.16	-	0.06	0.07	0.05	−0.05	−0.20
Literacy skills	Reading	10					−0.22	−0.02	−0.05	−0.13	−0.35 **	-	0.94 **	0.94 **	0.57 **	0.08
	RC	11					−0.27	−0.02	−0.09	−0.15	−0.37 **		-	0.78 **	0.58 **	0.04
	LW	12					−0.14	−0.02	0.01	−0.07	−0.29 *			-	0.51 **	0.12
	V-E	13					−0.16	0.10	−0.09	−0.05	−0.27 *				-	0.09
	V-C	14					0.35 **	0.34 **	0.29 *	0.12	−0.03					-

Note. Coefficients on the up-right were calculated by using English HLE, while coefficients on the down-left were calculated by using Chinese HLE. * p < 0.05; ** p < 0.01. IS = immigration status; NB = number of books; SR = parent–child shared reading; DD = digital device usage or TV program watching; DT = parent-directed literacy teaching; reading = reading overall; RC = reading comprehension; LW = letter and word recognition; V-E = receptive vocabulary in English; V-C = receptive vocabulary in Chinese.

) indicate that female students’ early literacy skills have negative correlations with parental language teaching in English (r = −0.36, p < 0.01) and in Chinese (r = −0.28, p < 0.05), indicating parents are less frequently engaged in language teaching with girls than boys at home. Gender (female) is positively correlated with English receptive vocabulary (r = 0.25, p < 0.05). Additionally, family SES is positively correlated with all four English early literacy tests: reading overall scores (r = 0.31, p < 0.05), reading comprehension (r = 0.34, p < 0.05), letter and word recognition (r = 0.27, p < 0.05), and English receptive vocabulary (r = 0.40, p < 0.01), but not with Chinese receptive vocabulary. Results reveal that the family SES is a significant factor associated with first graders’ English literacy skills, but not with their Chinese receptive vocabulary.

HLE factors are not strongly and positively correlated with any of the English early literacy skills. In contrast, HLE factors are positively correlated with Chinese receptive vocabulary, as three out of five Chinese HLE sub-scales have significant correlations with the test results, including the Chinese HLE overall scores (r = 0.35, p < 0.01), number of Chinese books (r = 0.34, p < 0.01), and shared reading in Chinese (r = 0.29, p < 0.05). These results reflect a strong effect of HLE on children’s Chinese receptive vocabulary.

4.4. Hierarchical Regression Analyses

4.4.1. The Relationship between Overall Home Literacy Environment and Early Literacy Skills

Since the relationship between HLE and early literacy skills was the main focus of this study, we conducted hierarchical regression analyses to further examine the effects of HLE on biliteracy development after controlling for children’s socio-demographic factors. In the first model, HLE overall scores were set as the key variable to examine its effect on early literacy skills after controlling for socio-demographic characteristics (

Table 5. Hierarchical regression analyses of overall home literacy environment and early literacy skills.

Variable	Reading Overall		Reading Comprehension		Letter and Word Recognition		English Vocabulary		Chinese Vocabulary
	B	SD	B	SD	B	SD	B	SD	B	SD
Constant	17.14	64.09	10.60	59.00	64.07	61.96	62.38	36.37	201.03 **	58.35
Age	0.77	0.77	1.02	0.74	0.34	0.75	0.05	0.44	−1.49	0.73
Gender	2.03	5.43	5.18	5.20	−0.69	5.25	3.34	3.08	3.12	5.50
SES	3.87 *	1.75	3.81 *	1.65	2.95	1.69	2.42 *	0.99	−2.34	1.74
Immigration status	−2.54	7.40	−0.16	7.04	−5.75	7.16	3.45	4.20	−13.63	7.89
Model fit
$R^2$	0.17		0.16		0.06		0.14		0.32
F (df1, df2)	1.34 (4, 50)		1.37 (4, 50)		0.84 (4, 50)		1.34 (4, 50)		3.14 (4, 42) *
Key variable (after controlling demographic factors and language program attendance)
HLE	1.42	1.27	1.67	1.20	0.94	1.22	0.98	0.72	1.94 *	0.90
Model fit
∆$R^2$	0.02		0.02		0.02		0.04		0.08
∆F	1.22		1.31		0.88		2.13		3.38 *

Note. * p < 0.05; ** p < 0.01.

). Overall English HLE scores have no significant effect on any of the English early literacy skills. However, overall Chinese HLE scores have a significant effect on Chinese receptive vocabulary: B = 1.94, p < 0.05, after controlling for socio-demographic factors. In addition, adding the overall Chinese HLE score significantly improves the model fit (∆F = 3.38, p < 0.05) and explains eight percent extra variance in Chinese receptive vocabulary.

4.4.2. The Relationships between HLE Sub-Scales and Early Literacy Skills

In the second model, the number of books was added as the key variable. As shown in

Table 6. Hierarchical regression analyses of number of books at home and early literacy skills.

Variable	Reading Overall		Reading Comprehension		Letter and Word Recognition		English Vocabulary		Chinese Vocabulary
	B	SD	B	SD	B	SD	B	SD	B	SD
Constant	41.00	56.08	27.36	52.70	67.66	55.60	63.63 *	31.85	170.85 **	47.97
Age	0.59	0.69	0.70	0.65	0.34	0.68	0.14	0.39	−1.10	0.60
Gender	2.69	5.18	4.68	4.87	0.88	5.14	3.94	2.94	5.78	4.94
SES	4.40 **	1.65	4.26 **	1.55	3.90 *	1.64	2.85 **	0.94	−2.36	1.60
Immigration status	−3.76	6.66	0.89	6.26	−7.85	6.61	1.04	3.79	−6.65	5.97
Model fit
$R^2$	0.18		0.18		0.21		0.22		0.18
F (df1, df2)	2.03 (4, 58)		1.99 (4, 58)		2.24 (4, 58)		2.56 (4, 58) *		2.12 (4, 55)
Key variable (after controlling demographic factors and language program attendance)
Number of books	2.70	5.76	2.53	5.41	2.53	5.71	1.91	3.27	3.60 *	1.38
Model fit
∆$R^2$	0.02		0.02		0.00		0.01		0.10
∆F	0.03		0.04		0.01		0.34		6.82 *

Note. * p < 0.05; ** p < 0.01.

, the number of books at home is not significantly associated with any of the English literacy assessment results. However, this key variable is a significant factor associated with Chinese receptive vocabulary results: B = 3.60, p < 0.05. Adding the variable significantly improves the model fit (∆F = 6.82, p < 0.05) by explaining an extra 10 percent of Chinese receptive vocabulary score variance.

The parent–child shared reading variable (

Table 7. Hierarchical regression analyses of parent–child shared reading and early literacy skills.

Variable	Reading Overall		Reading Comprehension		Letter and Word Recognition		English Vocabulary		Chinese Vocabulary
	B	SD	B	SD	B	SD	B	SD	B	SD
Constant	65.48	56.55	48.84	53.08	91.26	55.91	66.78 *	31.69	130.60 *	55.77
Age	0.46	0.69	0.60	0.64	0.21	0.68	0.15	0.38	−0.59	0.68
Gender	2.34	5.06	4.38	4.75	0.61	5.00	3.87	2.83	6.39	5.25
SES	4.09 *	1.61	4.03 **	1.51	3.57 *	1.59	2.80 **	0.90	−2.03	1.68
Immigration status	−4.73	6.40	−0.31	6.01	−8.41	6.33	0.89	3.59	−12.87	6.60
Model fit
$R^2$	0.11		0.14		0.10		0.19		0.13
F (df1, df2)	1.93 (4, 60)		2.02 (4, 60)		2.49 (4, 60) *		3.61 (4, 60) *		1.96 (4, 56)
Key variable (after controlling demographic factors and language program attendance)
Shared reading	0.92	2.02	0.87	1.90	0.85	2.00	0.38	1.13	6.53 *	3.24
Model fit
∆$R^2$	0.00		0.00		0.00		0.01		0.07
∆F	0.21		0.21		0.18		0.11		4.07 *

Note. * p < 0.05; ** p < 0.01.

) is not significantly associated with any of the English early literacy skills. However, it is a significant factor in predicting Chinese receptive vocabulary: B = 6.53, p < 0.05. Adding the parent–child shared reading variable significantly improves the model fit by explaining an extra seven percent of the variance in Chinese receptive vocabulary (∆F = 4.07, p < 0.05).

The length of digital device usage is found to be significantly related to reading comprehension: B = 5.39, p < 0.05. Results indicate that after controlling for socio-demographic factors, children with more usage of digital devices score significantly higher in their reading comprehension test (

Table 8. Hierarchical regression analyses of digital device usage and early literacy skills.

Variable	Reading Overall		Reading Comprehension		Letter and Word Recognition		English Vocabulary		Chinese Vocabulary
	B	SD	B	SD	B	SD	B	SD	B	SD
Constant	50.23	55.53	18.42	52.10	91.91	55.17	69.40 *	32.00	206.63 **	51.94
Age	0.50	0.70	0.78	0.65	0.10	0.69	0.08	0.40	−1.39 *	0.65
Gender	2.59	5.05	5.33	4.74	0.04	5.01	4.88	2.91	4.88	4.99
SES	3.72 *	1.60	3.88 *	1.50	3.01	1.58	2.64 **	0.92	−1.70	1.60
Immigration status	−4.45	6.42	0.37	6.03	−8.58	6.38	1.62	3.70	−13.71 *	6.10
Model fit
$R^2$	0.10		0.12		0.21		0.17		0.18
F (df1, df2)	1.42 (4, 56)		1.91 (4, 56)		1.25 (4, 56)		3.00 (4, 56) *		3.39 (4, 52) *
Key variable (after controlling demographic factors and language program attendance)
Digital device	3.84	2.38	5.39 *	2.34	2.21	2.36	1.73	1.37	3.33	4.28
Model fit
∆$R^2$	0.06		0.07		0.03		0.02		0.01
∆F	2.60		4.47 *		0.88		1.13		2.81

Note. * p < 0.05; ** p < 0.01.

). Furthermore, this key variable significantly improves the model fit. Seven percent of the variance in reading comprehension scores (∆F = 5.39, p < 0.05) is uniquely explained by digital device usage.

In the last model (

Table 9. Hierarchical regression analyses of parent-directed literacy teaching and early literacy skills.

Variable	Reading Overall		Reading Comprehension		Letter and Word Recognition		English Vocabulary		Chinese Vocabulary
	B	SD	B	SD	B	SD	B	SD	B	SD
Constant	24.76	63.79	−14.62	59.64	78.85	62.58	75.46 *	35.76	207.07 **	55.36
Age	0.86	0.77	1.27	0.72	0.29	0.76	0.06	0.43	−1.29	0.68
Gender	3.54	5.81	7.08	5.43	0.04	5.70	3.33	3.26	4.34	5.39
SES	4.23 *	1.71	4.48 **	1.59	3.35 *	1.67	2.47 *	0.96	−1.97	1.71
Immigration status	−5.41	7.04	−2.56	6.58	−7.31	6.91	2.21	3.95	−12.72 *	6.33
Model fit
$R^2$	0.18		0.14		0.08		0.15		0.15
F (df1, df2)	1.53 (4, 55)		2.15 (4, 55)		2.42 (6, 53) *		1.81 (6, 53)		2.35 (4, 53)
Key variable (after controlling demographic factors and language program attendance)
Direct teaching	3.24	3.06	4.11	2.86	1.82	3.00	0.17	1.71	−3.63	5.87
Model fit
∆$R^2$	0.02		0.03		0.01		0.00		0.15
∆F	1.13		2.07		0.38		0.01		0.86

Note. * p < 0.05; ** p < 0.01.

), parents’ direct literacy teaching is not significantly correlated with either English early literacy skills or Chinese receptive vocabulary.

5. Discussion

Our study reveals that the Chinese–Canadian first-graders achieve the same level of reading overall, reading comprehension, and letter and word recognition scores compared with the age-group norm. However, their receptive vocabulary in English and Chinese is slightly lower than their monolingual peers of the same age. These results suggest that their vocabulary skills in English and Chinese should be emphasized in their early literacy development both at home and in school. This is especially the case for children from low-SES families, who score lower than their peers from non-low-SES families across all English early literacy sub-skill tests.

Parents score higher in English HLE than Chinese HLE across all four components: there are more books in English at home, more parental engagement in shared reading and direct teaching in English, and more digital device usage in English. Parents from low-SES families score higher in English HLE but not in Chinese HLE. In comparison with parents from non-low-SES families, they are more engaged in parent–child shared reading and direct teaching in English. Parents who immigrated with their children are more engaged in HLE as they score higher in both English and Chinese HLE than those whose children were born in Canada. These results support the findings of Van Steensel [23], who points out that low-SES ethnic minority parents are not necessarily less engaged in HLE of the mainstream language. In contrast, they demonstrate active engagement in HLE to support their children in mainstream language development.

Our correlation analyses suggest that the children’s overall early literacy experiences in Chinese are significantly correlated with Chinese receptive vocabulary, but overall English HLE scores are not significantly correlated with any of the English literacy skills. Our hierarchical regression analyses echo these findings, as they show that overall HLE is a significant key variable in predicting Chinese receptive vocabulary but not in any of the English literacy skills. Similar to previous studies (e.g., [21,44,48,55]), the results indicate that, for bilingual children, HLE is more salient in affecting their heritage language than mainstream language, especially in early elementary years. Therefore, we conclude that by first grade, bilingual children’s English literacy development is less related to their HLE after controlling for socio-demographic factors. In contrast, HLE in heritage language is still an independent and significant factor affecting heritage language skills. To maintain or improve heritage language proficiency and, hence, support children’s bilingual development, parents should actively engage in their children’s heritage language learning, and provide high-quality HLE in that language at home.

Our hierarchical regression analyses also reveal that digital device usage (including TV program watching) in English has a significant effect on reading comprehension scores, while the number of Chinese books and the parent–child shared reading in Chinese has significant effects on Chinese receptive vocabulary. The important role of digital device use in early literacy development might be related to the outbreak of COVID-19 in 2019, during which we witnessed a surge of digital device and online resources usage during our data collection period. In the metropolitan area where the study was conducted, many local school districts incorporated online resources in support of children’s literacy learning. This was especially important for bilingual families, as many parents with limited English proficiency were not able to provide age-appropriate literacy instruction to their children. Many researchers argue that whether digital device usage supports children’s literacy learning or not is highly related to the way they use the devices (e.g., [16,26,27]). We conclude, based on our study findings, that with proper implementation and recommendation from schools, digital device usage can benefit bilingual children’s mainstream language learning by providing age-appropriate learning content and literacy activities.

Our analyses also confirm previous findings that HLE is crucial in supporting bilingual children’s heritage language development (e.g., [44,64,78,79]). In addition, the number of Chinese books and parental shared reading are also found to be significantly related to Chinese receptive vocabulary. This finding reflects the importance of HLE in maintaining heritage language. To foster heritage language learning, parents should provide more literacy resources and engage in more literacy activities in the heritage language. Furthermore, the insignificant effect of parent-engaged activities (i.e., shared reading and direct teaching) on English literacy skills may be explained by the fact that in this study, most parents were native Chinese speakers whose English proficiency was limited. As a result, they may not have been able to provide high-quality shared reading or direct teaching experience in English. Hence, the impact of these activities on literacy development may not be significant when parents are not native speakers of the language (e.g., [49,54,80]). Therefore, children’s mainstream language is more likely to be developed in school or through children’s independent literacy activities such as using digital devices or online resources at home, which provide more exposure and input in English.

In sum, Chinese–Canadian first graders may have lower oral vocabulary abilities in English and Chinese, but their decoding and reading comprehension skills in English are on par with their monolingual peers. Despite their dominant heritage language use at home, bilingual families demonstrate more parental engagement in English than in Chinese. For low-SES families, although they have a lower number of books in English at home, parents are more engaged in shared reading and direct teaching in English. However, the children rely more on HLE for their Chinese receptive vocabulary development than for any of their English literacy skills. By examining the relationships of different early home literacy experiences and early literacy skills, we find that children’s digital device usage is significantly associated with their English reading comprehension, while the number of books and parent–child shared reading is significantly correlated with their Chinese receptive vocabulary. This finding reflects the benefits of using digital devices and online resources in promoting literacy learning, especially during the COVID-19 pandemic. It also underlines the differences between HLE activities in supporting bilingual development.

6. Implications

This study’s findings have important implications for bilingual families, especially those who have a desire to develop children’s bilingual abilities. First, children’s heritage language learning relies on HLE. Parents can help their children in heritage language learning by providing high-quality literacy activities (such as parent–child shared reading, singing, and game-playing) and literacy learning resources, especially print books. For parents with limited English proficiency, digital device usage can be a useful tool to support mainstream language development, especially with proper support from schools and parents’ or caregivers’ involvement. Digital devices or online resources in English can provide age-appropriate literacy learning content to children when parents lack the language proficiency to do so. Our findings on the insignificance of parental-direct literacy teaching on either English literacy skills or Chinese receptive vocabulary also suggest that parents should not teach these skills at home, but focus more on quality parent–child shared literacy experiences [8]. Teachers should also encourage parents to continue with these activities at home with their children in their first language. For in-school instruction, teachers should strengthen children’s vocabulary instruction in English while attending to other emergent literacy skills.

This study also has implications for future research. Our study found that the availability of books and parent–child shared reading are significant only to Chinese vocabulary but not English literacy skills. Further research should investigate and compare the nature of parents’ engagement in these activities in each language. In addition, although we find that digital device usage is beneficial in English reading comprehension, future studies are needed to explore whether the benefits are a short-term phenomenon during the COVID-19 pandemic, or a long-term effect observed in other contexts. Additionally, as noted in previous studies (e.g., [18,26,27]), there are different conclusions about the effects of digital device usage on early literacy development; therefore, it is necessary to investigate the nature and patterns of engagement with digital devices to gain a better understanding of how these widespread digital technologies can support children’s literacy learning.