CORPATT: A SCALE FOR MEASURING ATTITUDES TOWARDS CORPUS USE

Sa？etak This study reports on the development of an instrument for measuring learners’ attitude towards the use of a language corpus in language teaching. The use of corpora in teaching is known as data- driven learning (DDL) and research has shown that students can benefit from corpus consultation to develop their language learning competences because it presents an important source of real target language discourse (Luo, 2016; Mizumoto, Chujo & Yokota, 2016). In addition to studying the effects of DDL, research has also shown interest in learners’ perceptions of such a learning experience, an issue addressed in a number of studies in a range of contexts and with a wide range of methods. This study aims at adding to this research and filling the gap by providing an instrument that measures the actual attitude towards such learning experience rather than asking students about the perceived benefits and problems of using corpora. The participants were students majoring in Tourism at the Faculty of Economics, Business and Tourism, University of Split whose attitude is measured using the newly developed instrument. We also compare the results of the two sub-samples: students of the undergraduate study programme of tourism and students of the professional study programme of tourism. Students’ attitude towards the experience was assessed using a 41-item questionnaire that was submitted to item analysis which resulted in the reduction of the instrument to the current 24 items. The principal components analysis indicated two components as sub constructs on this instrument: (i) the cognitive-behavioural component, and (ii) the affective component of the measure of attitude. Internal consistency of the two multi-item scales proved to be high indicating that a psychometrically valid scale for measuring attitude towards corpus use was constructed. Finally, we found that the students showed an average attitude towards the experience of corpus use and no significant difference was found between the two sub-samples compared using the independent samples t-test. Limitations of the study and suggestions for further study are emphasised at the end of the paper