The paper investigates factors underlying identification of synonymic adjectives big-large-great by students of English as a second language whose native tongues are Russian, Serb and Chinese. The results of the experiment have revealed a considerable difference in the identification of the adjectives: the best percentage of correct recognitions was obtained for "great", the worst for "large" as was predicted on the basis of semantic analysis. The native tongues are seen as another source of variability. The data of the research can be used in developing courses on the synonyms in teaching English as a foreign language.