MulTeC (Multimodal Teletandem Corpus)
A telecollaborative corpus (ENG-PORT) open to researchers around the world
The MULTEC corpus (ARANHA & LOPES, 2019), built from the Teletandem databank, structures the data of 282 Teletandem participants from 224 instantiations of the integrated Teletandem modality and 58 instantiations of the semi-integrated Teletandem modality. It comprises 581 hours 19 minutes of video data from toss, 666 learning diaries, 351 chat conversations used during TOS, 956 texts (both in Portuguese, written by l2 students and commented upon by l1 native speakers and in English, written by l2 students and commented by a proficient English speaker), 91 initial questionnaires and 41 final ones. The corpus comprises 145 GB of data.
MULTEC organization can contribute significantly with the researchers interested in working with telecollaborative multimodal learning environments and with oral and written aspects of foreign language learning. Instead of previously establishing a specific research question to design the corpus – as proposed by CHANIER and WIGHAM (2016), we defined a research purpose of compiling, anonymizing, and filing data collected by ARANHA, LUVIZARI-MURAD and MORENO (2015) and elaborating documents to specify characteristics of Teletandem context. Nevertheless, this process resulted into a researchable corpus that will allow internationalization and an open data source once we share the corpus with colleagues around the world.
Building MULTEC indicated the necessity to standardize the documents, to broaden theoretical background to include the concepts of pedagogical and learning scenarios (FOUCHER, 2010), to consider all the steps of corpus design for further collection to minimize the problems faced during this process.
The MULTEC corpus (ARANHA & LOPES, 2019), built from the Teletandem databank, structures the data of 282 Teletandem participants from 224 instantiations of the integrated Teletandem modality and 58 instantiations of the semi-integrated Teletandem modality. It comprises 581 hours 19 minutes of video data from toss, 666 learning diaries, 351 chat conversations used during TOS, 956 texts (both in Portuguese, written by l2 students and commented upon by l1 native speakers and in English, written by l2 students and commented by a proficient English speaker), 91 initial questionnaires and 41 final ones. The corpus comprises 145 GB of data.
MULTEC organization can contribute significantly with the researchers interested in working with telecollaborative multimodal learning environments and with oral and written aspects of foreign language learning. Instead of previously establishing a specific research question to design the corpus – as proposed by CHANIER and WIGHAM (2016), we defined a research purpose of compiling, anonymizing, and filing data collected by ARANHA, LUVIZARI-MURAD and MORENO (2015) and elaborating documents to specify characteristics of Teletandem context. Nevertheless, this process resulted into a researchable corpus that will allow internationalization and an open data source once we share the corpus with colleagues around the world.
Building MULTEC indicated the necessity to standardize the documents, to broaden theoretical background to include the concepts of pedagogical and learning scenarios (FOUCHER, 2010), to consider all the steps of corpus design for further collection to minimize the problems faced during this process.
All in all, as it is now, MULTEC is a researchable multimodal corpus composed of telecollaborative interactions among foreign language learners participating in different groups of Teletandem.
For more information, please contact: [email protected]
