Abstract
Learning to read a second language (L2) can pose a significant challenge for children already struggling with their first language (L1). This study investigated Chinese (L1) and English (L2) reading skills in a representative sample of 1,824 bilingual children in Grades 4 and 5 from urban and rural schools in Beijing. We examined the prevalence of reading difficulty in Chinese only (PC), English only (PE), and both languages (PB), calculating a stable 36% co-occurrence rate. Multinomial logistic regression revealed that boys and rural students face a disproportionately higher risk of reading difficulties. These findings suggest that despite structural differences between logographic and alphabetic systems, L1 and L2 reading skills are interdependent, likely sharing universal meta-linguistic foundations.
Introduction
Learning to read a second language is a monumental task, particularly for children already struggling with their native tongue. While research has established links between reading difficulties in two alphabetic languages (e.g., Spanish and English), the relationship is less clear when writing systems are fundamentally different.
Chinese is logographic, using characters to represent meaning and sound through radicals, whereas English is alphabetic, using letters to represent phonemes. This study sought to determine if reading difficulty "transfers" across these disparate systems and how demographic factors—specifically sex, grade level, and the urban-rural divide—influence the prevalence of these struggles.
Methods
The researchers conducted a cross-sectional study in Beijing, assessing 1,824 bilingual children in Grades 4 and 5. To isolate reading difficulty from general cognitive impairment, only children with a non-verbal IQ above the 40th percentile (measured via Raven’s Standard Progressive Matrices) were included.
Participants were categorized into four groups based on a 25th percentile cutoff on Chinese character recognition and English word spelling tests:
- PC (Poor Chinese readers): Struggled only in Chinese.
- PE (Poor English readers): Struggled only in English.
- PB (Poor Bilingual readers): Struggled in both.
- NR (Normal readers): Proficient in both.
Key Results and Demographic Influences
The Male Disadvantage and Grade-Level Trends
Across the sample, 8.20% were PC, 14.22% were PE, and 4.60% were PB. Boys were significantly more likely to fall into all three "poor reader" categories, being roughly 1.9 to 2.2 times more likely than girls to struggle. Additionally, 4th graders showed a higher risk of Chinese and bilingual reading difficulty than 5th graders, suggesting a developmental "catch-up" as children transition from "learning to read" to "reading to learn."
The Rural-Urban L2 Gap
While L1 (Chinese) reading difficulty was relatively similar across locations, a stark divide emerged in English proficiency. Rural students were 2.17 times more likely to be poor English readers (PE) than their urban peers. This suggests a "resource gap," where urban students benefit from greater English exposure and extracurricular support that helps them overcome potential reading hurdles.
The "Shared Core" of Reading
Despite the visual differences between Chinese characters and English words, the study found a stable 36% co-occurrence rate: if a child was a poor reader in Chinese, their chance of also being a poor reader in English was 36%. This rate remained remarkably stable regardless of sex or location, supporting the hypothesis that "meta-linguistic" skills—the ability to reflect on and manipulate language—are universal and transcend specific writing systems.
Discussion and Conclusion
This study demonstrates that L1 reading difficulty significantly increases the risk of L2 reading difficulty, even across logographic and alphabetic scripts. The findings highlight the importance of understanding the interdependent relationship between different writing systems. Furthermore, the identified male disadvantage and the urban-rural gap in L2 prevalence call for targeted educational interventions for disadvantaged populations.
While the study used English spelling as a proxy for reading and relied on a specific 25th percentile cutoff, the large-scale data provides a robust foundation for future research into the specific cognitive skills, such as phonological or orthographic awareness, that drive this 36% co-occurrence of reading failure.
Summary updated