I said I’d try and say a bit more – this is it...
As Toots has focused quite a bit on the 2005 Ehri article for which I gave a link and has pointed out that it doesn’t directly support the idea of a synthetic start for beginners, I’ll try and explain why I cited it.
I cited it (though only after mentioning articles by Johnston et al. which I regard as more directly relevant) because it clearly deals with the issue of automatic word-reading which Toots had raised, and because Ehri recognises the part played by good grapheme-phoneme knowledge in the development of this automatic word-reading, despite the fact that the kind of teaching she uses in her experiments is a bit different from synthetic phonics as we know it:
Ehri wrote:In our studies, learners did not decode the words taught with spellings. Rather they heard the words pronounced by someone else. In fact, spellings appeared but were never even mentioned. (p. 179)
The context of that comment needs to be taken into account, however: the paragraph starts by mentioning Share’s work, and if one knows the work in question, one realises that Ehri’s second sentence is a reference to the fact that in the three studies cited in her first sentence, Share was
talking about a decoding approach - identifying words by saying sounds for graphemes and blending the sounds. In the 2004 study, for example, he mentions ‘the letter-by-letter sounding out and blending’ which his Grade 3 subjects did on their ‘initial encounter with a new target’ (p. 278). But Share, like Ehri, also sets a lot of store by automatic word-reading and the contribution made to this by grapheme-phoneme knowledge, so although he uses a decoding approach and she doesn’t, there is enough common ground for them to quote each other’s findings as fitting in well with their own*.
In the same paragraph as Ehri says that her subjects didn’t decode, she allows for the possibility that ‘Use of a decoding strategy may help students’ in applying the knowledge that helps them to ‘secure the spellings of specific words in memory’ and thus eventually read word automatically. Moreover, at the end of that paragraph, despite having said that her subjects ‘did not decode the words taught with spellings’, she says it’s possible that ’some of the children decoded the spellings subvocally when they saw them’ (p. 179). It’s that kind of thing, together with anecdotal information that I’ve heard about Ehri, that makes me think that although she doesn’t use a synthetic phonics approach in her own research, she accepts that it may, like her own approach, lead on to good ‘sight’-word reading in her sense, which is a sense which I regard as legitimate.
Here’s more from Share, this time from a 1995 article which Ehri also cites, and which I have long admired:
He wrote:The advantage of an early code emphasis is substantial and not merely statistically significant. In their meta-analysis of research on early reading instruction, Pflaum et al. (1980) found that synthetic phonics (letter-sound instruction with explicit blending) produced an average outcome 35 percentile points higher than the mean of control groups compared to an average 15-point advantage for experimental groups generally. This elegantly dovetails with the conclusions reached above regarding the twin causal roles of letter-sound knowledge and phoneme blending. (‘Phonological recoding and self-teaching; sine qua non of reading acquisition’, Cognition 55. 1995)
I also find Share excellent on the subject of phonemic awareness, but that’s another story.
Toots wrote:Any interpretation is different in quality from an evidence-based conclusion. Where there is documented evidence there is no need for trust. It is important to know the difference. Wouldn't you agree?
Yes, I would agree, but I’d also want to say that background knowledge can affect one’s interpretation. In this case, my knowledge of Share’s work makes me think that although the documented evidence provided by Ehri’s own experimental work doesn’t directly support a synthetic phonics approach, she would accept that s.p. may lead to good ‘sight’-word reading in her sense.
*Share's high opinion of Ehri’s work is evident in his description of it as ‘ground-breaking’ in his 1995 article. The relevant section of that article also appears in an article produced jointly by him and Stanovich in the same year, so Stanovich must also think highly of Ehri. (Issues in Education
Vol. 1 No. 1, 1995)