#35303
Anastasios Asimakopoulos
Keymaster
    @anastasios

    Hi @pennar I particularly find error correction rather tricky. As you mentioned there could be several error types in a student’s work, and each is likely to require a different type of search, for each learners have to be gradually trained. Other factors that might complicate the picture is what kind of feedback is provided, if any, by the teacher. Does the teacher highlight the errors, identify the type of error, instruct student on the type of search required? If you would like to experiment, I would definitely recommend starting on a specific type of error and search. For example, you might work on parts of speech and helping students identify if the correct form has been used. You could show them that ‘will analysis’ gives 1 hit and ask them to type ‘will analys*’ (the asterisk stands for any undefined character) – see results : will analysis vs will analys*.

    I would recommend reading some research on error correction via corpus methods to get an idea of how much corpora can help and how effective they can be. The articles below would be a good start:

    Dolgova, N. and Mueller, C. (2019) ‘How useful are corpus tools for error correction? Insights from learner data’, Journal of English for academic purposes, 39, pp. 97–108. doi:10.1016/j.jeap.2019.03.007.

    Satake, Y. (2020) ‘How error types affect the accuracy of L2 error correction with corpus use’, Journal of second language writing, 50, p. 100757. doi:10.1016/j.jslw.2020.100757.