Cassar, M., Treiman, R., Moats, L., Cury Pollo, T., Kessler, B., 2005. How do the spellings of children with dyslexia compare with those of non-dyslexic children? Reading and Writing (2005) 18:27-49. The authors were interested in the question of whether children diagnosed as dyslexic made spelling mistakes of a different kind from those made by other children. They conducted a study in which they compared two groups of children who performed at the same level on a standardised spelling test: a group of dyslexics with a mean age of 11 years 7 months and a group of non-dyslexics with a mean age of 6 years 8 months. Their analysis showed no differences in the types of errors made by the two groups. In case there were subtle differences that their measures had failed to pick up, they had the spellings scrutinised by 44 teachers who were experienced at working with children with serious reading and spelling difficulties. The children’s spelling attempts were typed out so that handwriting would not show which were the younger and older children. It was found that ‘even the most experienced teachers could not reliably determine, based on a child’s spellings alone, whether that child was a typical beginner or an older child with dyslexia’. The authors comment that ‘the teachers’ poor performance, rather than reflecting poorly on them, shows that the two groups of children indeed produce very similar spellings’. They conclude that the same kinds of spelling difficulties (e.g. with consonant clusters) are experienced by relative beginners and older dyslexics and that ‘good instruction that focuses on the kinds of difficulties that are experienced by typical children should help all children’.
Editor’s comment: This study is particularly interesting in the light of the Channel 4 Dispatches programmes on dyslexia which was broadcast in September 2005. The focus there was reading rather than spelling, but similar conclusions were reached: that it was difficult to tell dyslexics and other poor readers apart and that teaching methods which were good for one group were good for all.
Hulme, C., Caravolas, M., Málková, G., Brigstocke, S., 2005. Phoneme isolation is not simply a consequence of letter-sound knowledge. Cognition 97, B1- B11. This is a further contribution to the debate about phonemic awareness. Specifically, the authors evaluated a suggestion made by Castles and Coltheart in Cognition 91, 2004 (see Research Digest in RRF Newsletter 52), that ‘children may be able to perform phonemic manipulations on those sounds for which they know the corresponding letter, but not...for those sounds for which they have not yet acquired the graphemic link’. In experiments carried out with groups of English and Czech children, Hulme et al. found that 22 out of 24 Czech children and 13 out of 16 English children were able to identify at least one phoneme (either the initial or the final phoneme in a spoken single-syllable non-word) for which they did not know the corresponding letter. 11 out of 25 Czech children and 5 out of 16 English children managed this for 5 or more phonemes for which they did not know the corresponding letters. In a second study, the researchers found four Czech children who knew none of the 15 letter-sounds used in the experiment – one of these children could not isolate any phonemes but two succeeded in all 15 phoneme-isolation tasks and one succeeded in 10. Unfortunately, however, the researchers could not be sure that these children knew no letter-sounds at all – it is possible that they knew some which were not among the 15 used in the experiment. The researchers conclude, as others have done, that there is a reciprocal relationship between phonemic awareness and learning to read and spell.
Editor’s comment: While there is little doubt about the reciprocal relationship, the nature of this relationship may vary to some extent according to whether children are taught to read words by phonics alone or by using the NLS type of range of strategies. A study which was interesting in this respect appeared in 1987. It was by Perfetti, Beck, Bell and Hughes: Phonemic knowledge and learning to read are reciprocal: A longitudinal study of first-grade children (Merrill-Palmer Quarterly, Vol. 33 No. 3). In talking about ‘children taught by direct code instruction’, these authors suggested that ‘their improvement in decoding may depend less on phonemic analytic abilities than does the improvement of children not taught coding directly’. The implication is that if something is directly taught, the ability of children to deduce it for themselves becomes much less important. If children are taught to link phonemes with graphemes from the start and are taught to read every word by sounding and blending, then they are being taught exactly what they need for word-reading purposes and there is much less margin for error.
Leppänen, U., Aunola, K., Nurmi, J-E., 2005. Beginning readers’ reading performance and reading habits. Journal of Research in Reading, Vol. 28 No. 4, November 2005. This research was carried out with Finnish children. In Finnish schools beginners are taught to read by relying on letter-sound correspondence knowledge. There are 21 correspondences in Finnish and just one is taught per week; it therefore takes about 6 months to teach all 21, but children start reading CV and VC syllables and words after the first few correspondences have been taught, which means that word-reading based on the taught letter-sound correspondences starts after just three or four weeks. The researchers investigated the effects of early reading skills on the extent to which children engaged in out-of-school reading and vice versa. They found that ‘the stage of reading acquisition is of importance in the formation of reading habits: it was particularly after the basics of reading had been learned that out-of-school reading began increasingly to contribute to children’s subsequent reading skills’.
Editor’s comment: This surely suggests that Frank Smith’s idea that children ‘learn to read by reading’ applies only after they have mastered the basics.