Patterns of errors in texts written by Costa Rican university English learners: A corpus-aided study

Abstract

The present corpus-aided study sought to identify the grammatical and non-grammatical second language (L2) error patterns of Costa Rican university English learners at all academic levels of a public university. Specifically, a total of 360 English as a foreign language learners, who were enrolled in the B.A in English or B.A. in English Teaching during the second semester of 2019, took the Quick Oxford Placement Test to ascertain their English Proficiency level and composed an argumentative text to elicit their written errors. Results from the placement test showed that the participants’ proficiency level ranged between B1 (low intermediate) and C1 (low advanced). In addition, the quantitative nature of the study required not only converting the handwritten compositions with a speech recognition software but also identifying and tagging all L2 errors with a tagging system. Analyses of a statistical software for data management revealed that the learner corpus contained a total of 33 L2 error patterns, which were classified as follows: 17 grammatical, 10 stylistics, and 6 lexical. Main descriptive findings indicated that although some error frequencies lowered to the point of having none as learners advanced in the major (e.g., capitalization and superlatives), other linguistic problem areas persisted all throughout (e.g., word form errors, fragments, and word order). Concluding remarks highlight that because the error frequencies of some L2 error categories still ranked high over time, learners’ L2 knowledge of lexical, syntactic, morphological, and stylistic domains could need more expert input (in the form of explicit instruction and/or feedback) depending on the complexity of the target structure.