Přejít k hlavnímu obsahu

Přihlášení pro studenty

Přihlášení pro zaměstnance

Publikace detail

Designing, Compiling and Analyzing Learner Corpus
Rok: 2014
Druh publikace: ostatní - přednáška nebo poster
Strana od-do: nestránkováno
Tituly:
Jazyk Název Abstrakt Klíčová slova
eng Designing, Compiling and Analyzing Learner Corpus The presentation describes a part of a bigger multidimensional project called “Aspects of English Language Acquisition of Czech Students on the Onset of Teacher Education” which is focused on the processes of acquiring communicative competence in speaking English as a foreign language by Czech university learners and is financially supported by the Czech Science Foundation. As argued by many linguists, writing and speech are two different systems (e.g. Carter and McCarthy 1995, Biber et al. 1999) and the specific features of dynamic on-line production in conversation lead to a certain degree of dysfluency which brings structures typical of grammar of speaking. Thus the process of designing the data collecting tools reflected (a) the declared differences between speech and writing and (b) the preliminary smaller scale studies carried out by the team. The form of students’ oral performance was brought closer to the “natural” conversation so that the analysis could include the following areas: (a) spoken language features; (b) discourse management structures; (c) grammatical accuracy of the performance; (d) potential Czech native language negative transfer. One of the areas of analysis is focused on the occurrence and distribution of non-clausal units in spoken discourse (i.e. elliptic replies, condensed questions, echo questions, etc.) with the aim to find out how far the learners’ spoken performance differs from the native speakers’ one. Finally the conclusions made on the basis of the analyses could be used for modifications of course syllabi and curriculum of the TEFL study programme. learner corpus; English acquisition; spoken language; annotation