SciELO journals
Browse
1/1
14 files

Recurrent error analysis in a learner corpus of Spanish as a foreign language (Corpus CAELE)

dataset
posted on 2021-03-26, 23:41 authored by Anita Ferreira Cabrera, Jéssica Elejalde Gómez

ABSTRACT The CAELE is a collection of 418 written texts produced by 62 learners of Spanish as a foreign language (SFL). These texts were collected between years 2014 and 2015 and are saved and processed in digital format. The main objective of this study is to determine the most frequent and recurrent errors from that corpus with the aim of supporting the selection of adequate errors for the processes of grammatical correction with Written Corrective Feedback (WCF) strategies. Results indicate that the most frequent and recurrent errors in the CAELE are those of false selection of grammatical genre and the omission of orthographic word stress. Regarding proficiency levels A2 and B1, the most recurrent error in both levels corresponds to the omission of orthographic stress in last-syllable-stressed words. Although there are some differences: omission of orthographic stress in last-syllable-stressed words and hiatus in level A2, and false selection of grammatical genre and stress omission in diacritics at level B1. These results suggest an important tendency in the order in which errors should be treated according to proficiency level in the process of written correction.

History

Usage metrics

    Revista Brasileira de Linguística Aplicada

    Categories

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC