Proxy variables in measuring L2 proficiency: The price of convenience

Language proficiency is a central construct in second language learning research, yet it is frequently operationalised through indirect or proxy measures such as self-ratings, academic achievement, or self-perceived competence. While these indicators offer practical advantages, they risk misrepresenting learners’ actual ability and distorting theoretical inferences. The present study investigated the extent to which learners accurately estimate their own English proficiency and examined the empirical consequences of substituting proxy variables for an objective proficiency measure. A total of 228 English language learners completed the Oxford Placement Test alongside measures of self-rated CEFR proficiency, self-perceived communication competence, and self-reported academic achievement. Only 37% of participants accurately estimated their proficiency, with lower-level learners tending to overestimate and advanced learners to underestimate their abilities. Correlations between the Oxford Placement Test and proxy variables were modest (r = .20–.49), and structural equation models using proxies in place of objective proficiency produced results differing in strength and directionality in explanatory models. These findings illustrate the price of convenience: while proxy variables are practical, their use impacts the findings of studies and may compromise validity. The study calls for renewed methodological caution and greater investment in direct, standardised proficiency assessments within applied linguistics research.