The English Proficiency-Social Status Link in China: Insights from the 2021 CGSS Survey

In China, English fluency is more than a linguistic ability; it is increasingly seen as a strategic asset linked to improved educational and career prospects, with implications for social mobility (Wei, 2013). As China's economic and cultural dimensions rapidly align with international norms (Goldstein, 2020; Mingjiang, 2008), this study will investigate the relationship between English language proficiency and social standing. By examining the connection between language skills and social hierarchy, the research's value stems from its potential to enhance our grasp of China's sociolinguistic trends and the wider impact of English learning in a globalized era. Moreover, the findings may inform policymakers and educators by offering insights into the societal effects of English education policies, shedding light on educational approaches that could foster equity and effectiveness in a milieu where English, despite being a non-native language, holds considerable value.

Research Question: "What is the Relationship between Social Status and English Proficiency in China?"

Hypothesis: "In China, individuals with higher English proficiency tend to have a higher social status, compared to those with lower or no proficiency in English."

This hypothesis asserts that English proficiency transcends linguistic ability, encompassing educational and cultural effectiveness in China. It suggests that mastery of English, the lingua franca of global discourse, likely aligns with increased educational prospects, superior employment opportunities, and broader international engagement—elements typically indicative of elevated social standing. Such status can be quantified through income, professional acclaim, academic success, and the breadth of social networks (Barrett, 2003). English competence can be gauged via standardized test performance, scholastic English accomplishments, or practical usage in daily and vocational contexts (American Educational Research Association et al., 1999). This analytical framework scrutinizes the nuanced interplay between English fluency and societal position, providing a platform to probe variations of this dynamic across China's diverse regions, age brackets, and urban-rural divisions.

Data Description

My analysis focused on the comprehensive 2021 CGSS dataset, which encompasses survey responses from more than 8,000 individuals throughout China. This dataset is rich with approximately 700 variables, offering a broad spectrum of insights. Demographically, the sample exhibits a balanced gender representation with 45% male and 55% female respondents. Educational background was assessed on a 13-point scale, with the average participant education level aligning with high school completion (level 5). However, as will be subsequently discussed (Figure 1), the dataset reveals significant variation in educational attainment among participants.

Figure 1: Education Level Distribution

The standard deviation of 3.314 in educational attainment among participants, shown in Figure 1, points to considerable heterogeneity. This dispersion likely mirrors the varied socio-economic contexts within the sample and indicates a broad cross-section of educational experiences in Chinese society. The significant diversity in education levels within the dataset suggests it covers a wide spectrum of the population, which is critical for our investigation into the nexus between English skills and social standing across varied educational echelons. Yet, we must recognize that such diversity might also introduce bias, as access to and interest in participating in surveys can be unevenly distributed across educational tiers.

Respondents assessed their English listening (A51) and speaking (A52) proficiencies using a 1-to-5 scale, excluding responses such as 'I don't know' and 'refuse to answer' (see Figure 2 below). Descriptive analysis revealed an average listening score of 1.65 and a standard deviation of 0.33. The mean speaking score was marginally lower at 1.55, coupled with a smaller standard deviation, indicating less variability in speaking ability reports. The chart demonstrates that at Level 1 Proficiency, a notable number of participants self-report minimal listening and speaking abilities, with listening slightly ahead. At Level 2 Proficiency, a considerable percentage acknowledges moderate comprehension and conversational skills in English, with a leaning towards listening. As proficiency ascends from Levels 3 to 5, there is a discernible decline in self-assessment, revealing a concentration of lower-level proficiencies amongst respondents. Evaluating listening against speaking across the spectrum, it emerges that respondents consistently claim better understanding than expressive language skills. This pattern aligns with typical trends in linguistic development.

The observed trends in English proficiency among the Chinese student population might be influenced by multiple factors. The educational emphasis on English suggests broad exposure; however, this does not necessarily equate to high speaking proficiency (Wei, 2013). The evidence points to a disconnect between the passive understanding and active use of language in educational settings, underscoring a potential need for curricula with a stronger focus on conversational skills. The substantial linguistic divergence between Chinese and English also likely plays a role, as effective speaking requires more intensive practice and immersion than is currently available for many learners. English proficiency levels hold significant implications for social mobility within China, correlating with enhanced educational and vocational prospects. Despite widespread instruction in English, attaining fluency remains a hurdle for many, which may affect social standing and access to opportunities in an increasingly global economy.

Figure 2: English Proficiency

The survey also employs a social ladder metaphor to gauge participants' self-assessed societal status and its evolution. Initially, individuals locate their current position on a 10-rung societal ladder, which depicts the top as high status and the bottom as low status. Subsequently, they evaluate their status a decade prior and project their standing a decade hence. Additionally, participants recall their family's status at age 14.

A43_a: Overall, in today's society, which social stratum do you personally belong to?

A43_b: Which level do you believe you were at 10 years ago?

A43_c: Which level do you believe you will be at in 10 years?

A43_d: What social class do you believe your family belonged to when you were 14 years old?

Figure 3: Perceived Social Stratum

The data reveals a notable clustering of individuals identifying with the central social tiers (levels 4 and 5). A retrospective view uncovers a modest positioning in the past decade, with the majority aligning themselves with the lower-middle strata (levels 3 and 4). Forward-looking expectations indicate an upward aspirational trend, with a peak at level 6. Furthermore, reflections on family status during early adolescence (age 14) indicate a prevalent association with the lower echelons of the social hierarchy, primarily levels 2 and 3.

Statistical Analysis

Before choosing the proper test, normality was tested. Deviations from the expected diagonal in all four Q-Q plots, especially in the tails, suggest distributions with heavier tails than normal. Shapiro-Wilk test results, with p-values near zero, reject normality well below the 0.05 significance threshold. These findings justify using non-parametric methods like Spearman's correlation for further analysis.

Figure 4: Q-Q Plot for Perceived Social Status

Figure 5's Spearman correlation matrix analysis reveals several notable relationships between English proficiency and perceived social status variables. Correlation coefficients range from -1, denoting a strong negative correlation, to +1, indicating a strong positive correlation. English proficiency components (A51 and A52) are strongly positively correlated (r = 0.8864), reflecting their commonality in measuring language proficiency. Current self-assessed social status (A43_a) demonstrates a moderate positive correlation with both listening (r = 0.1593) and speaking abilities (r = 0.1590) in English, suggesting that higher language skills are related to elevated present social status perceptions. Historical (A43_b) and future predicted social status (A43_c) show a strong inter-correlation (r = 0.6924), hinting that past social self-perception is aligned with future expectations. Family status during adolescence (A43_d) exhibits moderate positive correlations with all other social status measures, most pronouncedly with current status (A43_a, r = 0.5289), pointing to family background's significant role in shaping current self-perceived social standing. The p-values corroborate the statistical significance of these correlations, particularly between English proficiency and perceived social status, emphasizing their non-random association and the variation in their linkage strength. Despite not being visualized here, the near-zero p-values allow for a high confidence in the correlations' significance. It is important to note that observing significant correlations between English proficiency and perceived social status does not establish causality. To determine whether a causal link exists, additional investigation is required.

Figure 5: Spearman’s Correlation

Adopting a multivariate regression approach, I probed the interplay among the variables under scrutiny. This method forecasts the dependent variable, herein English proficiency, using several predictors, namely social status indicators (see Figure 6).

Figure 6: Regression

The model's residual standard error is 0.8165, suggesting a moderate fit given that A51 ranges from 1 to 5. With an R-squared value of 0.1191, the model explains nearly 12% of the variance in listening proficiency, indicating significant unaccounted factors. The adjusted R-squared, at 0.1186, implies an efficient number of predictors relative to the sample size. The F-statistic of 241.4 with a negligible p-value strongly supports the model's significance. Contrasting coefficient signs reveal nuanced relationships: negative for A43_a and A43_b, questioning whether higher current or past social status aligns with better listening skills. Conversely, positive values for A43_c and A43_d correspond with anticipated societal standing and familial background, aligning with expectations that these factors contribute positively to listening proficiency. These dynamics underscore the complex interplay between social status and language skills and may reflect underlying educational or experiential differences or potential measurement biases in social status assessment.

Critical Evaluation

The study sheds light on the link between English proficiency and perceived social standing in China, yet caution must be exercised regarding the research's scope and transferability. The CGSS dataset, albeit extensive, may not capture the full spectrum of Chinese demographics, particularly underrepresented rural locales with lower survey engagement — a factor that could skew wider applicability. Subsequent inquiries might benefit from employing a stratified sampling approach to better encompass the variegated strata of the population.

In addition, the current study utilizes correlation-based evidence, precluding definitive causal inferences. The relationship between English proficiency and social status is ambiguous, as it remains to be established whether language skill precipitates enhanced social standing or vice versa. Prospective inquiries employing longitudinal methodologies or controlled experiments could elucidate this causality. Furthermore, reliance on self-reported social status and linguistic competence introduces the possibility of subjective bias, where participant perceptions may not align with their concrete societal positioning or linguistic proficiency. Future research integrating objective criteria – for example, quantifiable income data, professional positions, or certified language assessment scores – would substantiate the findings.

Also, the model leaves a considerable 88.09% of English proficiency variance unaccounted for, indicating the influence of additional substantial factors. Incorporating variables such as educational quality, language resource access, personal motivation, and learners' age might enhance explanatory power. Given China's vast cultural and regional diversity, the emphasis on English skills likely fluctuates across different locales and population segments. Implementing region-specific analyses or adjusting for regional disparities could yield more refined insights. Moreover, the dynamic nature of China’s social and economic landscape suggests that the relationship between English proficiency and social standing is fluid. Periodic updates with fresh data are essential to track evolving trends and reconcile the impact of societal shifts.

By addressing these limitations, future research could provide a more comprehensive and accurate picture of the relationship between English proficiency and social status in China, contributing to more informed policy-making and educational strategies aimed at leveraging language proficiency as a tool for social mobility.

References

American Educational Research Association, American Psychological Association, National Council on Measurement in Education, Joint Committee on Standards for Educational, & Psychological Testing (US). (1999). Standards for educational and psychological testing. Amer Educational Research Assn.

Barrett, A. E. (2003). Socioeconomic Status and Age Identity: The Role of Dimensions of Health in the Subjective Construction of Age. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58(2), S101–S109. https://doi.org/10.1093/geronb/58.2.S101

Goldstein, A. (2020). China’s Grand Strategy under Xi Jinping: Reassurance, Reform, and Resistance. International Security, 45(1), 164–201. https://doi.org/10.1162/isec_a_00383

Mingjiang, L. (2008). China Debates Soft Power. The Chinese Journal of International Politics, 2(2), 287–308. https://doi.org/10.1093/cjip/pon011

Wei, R. (2013). Chinese-English bilingual education in China: Model, momentum, and driving forces. Asian EFL Journal, 15(4), 184–200.