eBangor

Inferencing unknown words in reading.

Davies, David Rees. (1991) Inferencing unknown words in reading. PhD thesis, Prifysgol Bangor University.

[img] Text
Davies signed declaration.pdf
Restricted to Repository staff only

Download (18kB)
[img]
Preview
Text
DX095181_1_0001.pdf

Download (16MB) | Preview

Abstract

We've probably all had the experience, when reading, of coming across an unfamiliar word and trying to guess some of its meaning. This thesis is a study of the guessing strategy used. Independent variables are: word meaning (known/unknown), form of presentation (cloze/pseudoword), word class (noun/verb), amount of information (3 amounts), orders of types of information (6). Dependent variables are: accuracy, confidence (belief in one's accuracy) and uncertainty (the number of alternative hypotheses held). Subjects are native speaker university students. The main result is that subjects tend not to guess unknown meanings. They treat them as known meanings (I e. they guess a familiar single word rather than a new meaning) by regarding the meaning cues, as they appear across varying amounts of information, as inconsistent items of information. Whilst there are interesting differences for form, the presence or absence of an unfamiliar form does not materially affect this process. There are also interesting differences for order. However, an interpretation of this finding in terms of a principle of costs and benefits suggests subjects would not employ an order based strategy in real life. The effectiveness of guessing as a communication and as a learning strategy is evaluated in the light of these findings.

Item Type: Thesis (PhD)
Uncontrolled Keywords: Linguistics Education Psychology
Subjects: Degree Thesis
Departments: College of Arts and Humanities > School of Linguistics and English Language
Degree Thesis
Date Deposited: 14 May 2015 04:10
Last Modified: 01 Sep 2016 08:39
URI: http://e.bangor.ac.uk/id/eprint/4078
Administer Item Administer Item

eBangor is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.