The influence of instructional approach on the reading strategies of beginning readers
MetadataShow full item record
There are a number of models of reading development which propose that reading develops in a set sequence of stages (e.g. Frith 1985, Marsh et al 1981), and that each child must pass through one stage before it can move onto the next. It is been pointed out that these models very rarely take into account external factors such as the method of instruction that the children receive (Stuart and Coltheart 1988, Goswami and Bryant 1990) and what effect such factors would have on progression through the stages. This study investigated how the factor of instruction influenced how children read. Young children taught by two different methods were studied. Scottish five and six year olds taught by a phonics method, where they were shown the correspondences between letter segments and their sounds, were compared with New Zealand children of the same age taught by a language experience approach. Samples were matched for reading age, chronological age, time at school, vocabulary knowledge and digit span. Error analyses of responses to single words showed a marked divergence in reading strategies. The Scottish children were much more likely to attempt to read unfamiliar words, whereas the New Zealand children often failed to attempt to read items they did not know. The errors the Scottish children made were also qualitatively different to those of the New Zealand children. The Scottish children were better at pronouncing nonwords and were more advanced in spelling performance. The Scottish children were also superior at a test of simple phonological segmentation. They also produced a word length effect when reading words. The New Zealanders, however, were better at pronouncing irregular words and were faster readers, especially with familiar classroom words. They did not produce a word length effect even when words were distorted. Overall the Scottish children showed more evidence of a grapheme to phoneme conversion strategy, which in turn was correlated with good reading performance. The New Zealanders displayed signs of a more visual approach to reading. There was some overlap between the national groups particularly regarding the prevalence of errors incorporating beginning and end letters. The older children in each national group also showed a greater convergence of strategy use than the younger readers. This work therefore has implications for the efficacy models of reading, such as Frith's (1985). Matched groups of children should display the same reading strategies if reading skill is accomplished in universal stages, in this study they do not. Future models of reading development will need to take into consideration how the child is taught to read.
Thesis, PhD Doctor of Philosophy
Items in the St Andrews Research Repository are protected by copyright, with all rights reserved, unless otherwise indicated.