Towards an Authentic `Discourse Cloze'  

Oleh:

Deyes, Tony

Jenis: Article from Journal - ilmiah internasional
Dalam koleksi: Applied Linguistics (Full Text) vol. 5 no. 2 (1984), page 128-137.
Fulltext: Vol 5, 2, p 128-137.pdf (597.82KB)

Ketersediaan

Perpustakaan PKBB Nomor Panggil: 405/APL/5 Non-tandon: tidak ada Tandon: 1

Lihat Detail Induk

Isi artikel

This paper is essentially programmatic. It follows suggestions by Alderson (1979), Jongsma (1980), and others that cloze tests used to evaluate reading comprehension should derive their deletion criteria from aspects of the reading process, rather than the traditional pseudo-random (every n-th word) deletion procedure. Some recent attempts have been made at establishing a 'Discourse Cloze' (e.g. Levenston et al 1982), but, the present author argues, the limitation of deletion items to cohesive features makes these 'Text-Cloze tests' rather than 'Discourse-Cloze tests'. If a truly discourse cloze is to reflect the reader's ability to follow information through the text and use contextual clues as well as co-textual ones then, it is argued, theme and rheme, as units of information, provide criteria for item deletion. Relevance of these units to the comprehension of the discourse can be determined by applying the concepts of'frame' (Minsky 1975) and macro-structure criteria (Kintsch and Van Dijk 1977). 'Cloze-tests' on these principles illustrate and substantiate the claims of the writer.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali