Using BERT to calculate word predictability in language learner texts and its association with language proficiency.
Measures calculated from two corpora are provided in this repository.
Texts from the Learner Proficiency Corpus (LPC) are provided in the data/ directory.
See the following papers for more information:
- Crossley, S. A., Salsbury, T., McNamara, D. S., & Jarvis, S. (2011a). Predicting lexical proficiency in language learner texts using computational indices. Language Testing, 28(4), 561–580. https://doi.org/10.1177/0265532210378031
- Crossley, S. A., Salsbury, T., McNamara, D. S., & Jarvis, S. (2011b). What Is Lexical Proficiency? Some Answers from Computational Models of Speech Data. TESOL Quarterly: A Journal for Teachers of English to Speakers of Other Languages and of Standard English as a Second Dialect, 45(1), 182–193. https://doi.org/10.2307/41307622
The TOEFL 11 texts may be requested from ETS.
Holmes, L., Crossley, S., Morris, W., & Choi, J. S. (n.d.). Word Predictability as a Measure of Second Language Proficiency. Language Learning. https://doi.org/10.1111/lang.70037