Abstract
Despite the benefits to investigating learners' interlanguage through corpora, many are not compiled according to the Common European Framework of Reference (CEFR) levels, making it difficult to establish what learners can do at each stage. Furthermore, learners classify their English knowledge based on their years of experience, many times specifying an unattained level of proficiency. Much has been done on written corpora calibrated to CEFR, but research on calibrated spoken corpora is in its nascence. Taking into account the aforementioned considerations, the aim of this research is to compile the Brazilian Spoken English Learner (BraSEL) Corpus calibrated to the CEFR to further investigate, through a corpus pragmatics approach, what the development of spoken grammar and pragmatic features reveals across the levels. This poster will detail the planning phase of designing the corpus matrix, metadata information, and the task variables for the recording sessions.