This thesis examines the impact of Data-driven learning (DDL) or classroom concordancing on a group of adolescent students in Greece seeking to investigate the degree of motivation to learn grammar when involved in DDL a nd the...
moreThis thesis examines the impact of Data-driven learning (DDL) or classroom concordancing on a group of adolescent students in Greece seeking to investigate the degree of motivation to learn grammar when involved in DDL a nd the effectiveness of DDL in the teaching and learning of grammar. The study introduced concordance-based grammar materials to the experimental group, whereas a conventional grammar book was used with the control group, examining common grammar items and patterns. The analysis of classroom data gathered during DDL sessions offered insights into the improved noticing skills of the participants but also into the difficulties when involved in DDL with regard to unknown vocabulary and the Key Word in Context (KWIC) concordance format, which underlined the need for considerable teacher guidance. The qualitative evidence drawn from questionnaires and interviews suggested that the majority of the participants acknowledged the contribution and potential of corpora, but the degree of motivation to study g rammar further varied. M ost learners also expressed their preference for concordance-based learning, rather than their previous mostly passive learning experience, and further access to corpora, but without total abandonment of the conventional grammar book. The qualitative evidence was supplemented with analysis of test performances of the two groups, according to which more participants of the experimental group scored higher than those of the control group in each test. All these findings pointed to important gains and represent a preliminary step in the development of corpusbased grammar teaching to EFL adolescent learners. As the pre-electronic corpora were applied in lexicography, grammatical studies, dialect and language education (see section 2.2) the current expanded forms of second generation of corpora have even greater potential in additional fields, such as language variation and translation. However, an analysis of lexicography and issues regarding language teaching and learning follows as being related to the present study. Corpus-based dictionaries give a lot of very useful information, such as syntactic patterning, collocation, phraseology. A s mentioned earlier, prior to the introduction of computer corpora in lexicography, all of this information had to be collected manually, which resulted in years of dictionary creation. Nowadays, all information is available in computer-readable form and using concordancing software many stages can be automated, such as word frequency, detecting prefixes, suffixes, irregular forms, sorting words by lemmas and in the case of a tagged corpus, the part of speech of each word can be automatically determined. Corpus-based dictionaries include new meanings which were considered previously to be unimportant but which come to light owing to the results of frequency analysis. The dictionary definitions are typically prioritised, using information from tagged corpora, according to the frequency of use of meanings and sub-senses. The Longman Dictionary of Contemporary English based on British National Corpus includes words that are frequent in spoken and written English and compares their frequency in the two modes, while a large corpus of third wife Suzanne Acosta, 34, is expecting their first child. [p] Th in a phenomenally good Christmas and is expecting to have to make a large number of profit Nylex, its Australian subsidiary. The City is expecting further sales, with the most likely