Page 1 of 1

Finding Statistics

Posted: Wed Nov 29, 2006 7:18 am
by Kelly
Can someone please direct me to some statistics in favour of synthetic phonics. I know of the Clackmannanshire, but what else is there?


Posted: Wed Nov 29, 2006 12:34 pm
by chew8
Kelly -

See Marlynne Grant's article in Newsletter 52 at the RRF website.


Posted: Wed Nov 29, 2006 3:01 pm
by Susan Godsland
A Small Study Comparing National Literacy Strategy Progression In Phonics Tuition (PIPs) with Sounds~Write Tuition (S~W)

Posted: Thu Nov 30, 2006 12:43 am
by Derrie Clark

Posted: Thu Nov 30, 2006 9:36 pm
by john walker
Dear Dick,
Thank you very kindly for your warm acknowledgement of what we are trying to do at Sounds-Write. As we now have data on over 1,700 reception pupils, over 700 pupils in Y1, and over 200 pupils in Y2, all of whom have had S-W since beginning reception, we are doing our best to demonstrate the effectivenesss of our linguistic (and synthetic!) phonics programme.
As for the way in which we do it: teach from sound to print right from the beginning; teach the code from simple to complex (ie one-to-one correspondences; then many spellings to a sound and many sounds to a spelling); teach all THREE skills of blending, segmenting, and phoneme manipulation; provide the instruction within a Vygotskian mediated learning approach; and, teach every day from the moment they arrive in reception.
We find the Diane McGuinness/Vygotskian combination is a winner!
If you would like to know more, do get in touch with us through the website.
Best wishes,
John Walker

Posted: Mon Dec 04, 2006 4:20 am
by AngusM
Dear Kelly:

There's a paper by Marva Wrench of Texas Southern University in the ERIC database - - which mentions varous other 'studies' (in addition to the ones mentioned above) which evaluate synthetic phonics.


Posted: Mon Dec 04, 2006 4:24 pm
by Susan Godsland
No statistics given, but a small study: ... sdale.html

Posted: Thu Dec 07, 2006 10:15 pm
by ian
This has been on the DfES website since 2003. Shame they didn't read it: ... solity.pdf

It is Solity writing about his program, ERR (Early Reading Research). He has published loads of articles about it (summarised the this article and referenced). There are more recent statistics collected by Essex LEA, but unfortnately I don't have them in electronic format.

Posted: Thu Dec 07, 2006 10:28 pm
by maizie

I don't think we've had anyone who uses ERR posting on this board before, so can I pick your brains?

About a year ago there was some publicity about Solity's research which concluded that children should be taught 100 HFWs alongside their phonics. This is a bit of an anathema to many SP people :) But, do you do this as part of the programme, if you do, do you teach them as 'whole' sight words, and, what do you think of it?

Posted: Thu Dec 07, 2006 11:00 pm
by ian
I heard about that article (TES, I think) but didn't read it. I imagine it was relating to research evaluating the whole ERR program, which, even with HFWs, is better than mixed method teaching.

I don't think Solity ever claimed ERR to be synthetic phonics but, on the whole, it is. I've never been sure about teaching 'sight words', but so many 'synthetic phonics' schemes do. In the original ERR scheme the idea was that the 100 words should be taught 'at sight' (yes, as wholes) to help children have access to books before they know the phonics behind the words. Also, to make sure they are fluent at reading these words. Personally, I've never been convinced by this!

Since speaking to Ruth Minskin (at the RRF conference), I've made up my mind. I certainly don't teach regular words (which make up about 60 of the 100 high frequency ERR words) 'at sight' and now I highlight the tricky parts of the words in red. I call them 'naughty letters' (I didn't invent that idea and Ruth calls them 'grotty graphemes'). To begin with, I show the individual graphemes in the word using lines and dots and model how the sound them out. I have the words displayed like this for spelling reference.
So, I still teach the 100 words (well, I only ever through the first 40) but now I tackle them in a phonic way.

ERR has many shortcomings, so I've changed it to include, for example, application of spelling skills. However, is based on some good principles, some principles that other programs could learn from.

Posted: Fri Dec 08, 2006 12:25 am
by Debbie Hepplewhite
I find it most strange that Jonathan Solity appears to be so set on the formula of the Early Reading Research such that he possibly fails to try and change parts of it to see if getting closer to synthetic phonics improves his results further.

What I am suggesting is that this does not seem to represent the characteristics of a researcher.

I am surprised, for example, that he does not compare the ERR formula with NOT learning an initial sight vocabulary.

I am not in touch with Solity, however, so maybe he does try variations on a theme as part of his research activities.

Or maybe he is determined that he is totally right and others are wrong.

This would be a shame because I know of several leading synthetic phonics proponents who have been most interested in the ERR shorter and more often blending and segmenting word level activities and the ERR speed reading activities.

There is such a thing as identifying good 'aspects' of a programme and adopting them to see if our teaching can get better and better!

Posted: Sat Dec 09, 2006 4:17 pm
by ian
I agree with you Debbie. As I said on a TES forum recently, there have been many studies comparing a specific SP program to other approaches and now I think there is a need for more research (not just from Solity) into specific aspects of programmes. Even comparing say, JP to RML would not yield particularly useful results because there are far too many variables. Controlled trials need to be constructed using identical programs but changing one variable; for example:
- teaching an initial at sight vocabulary or not;
- teaching letter names or not;
- using real books or decodables (or both);
- teaching confusable letters (bdpq) together or apart.
One difference probably won’t be massive but combined may help those children who find reading the most difficult. (Such research may already exist, correct me if I’m wrong.)

Solity recognises this need himself. I’ve just re-read an old paper of his and it says:
“Troia (1999) discusses the issues in applying psychological research in reading to classroom practice. Experimental methodology requires that participants are randomly assigned to conditions, that control groups receive a comparable level of contact to experimental groups and that the research investigates the impact of a single variable. This has not been possible within the ERR for a variety of reasons.”
“Further research is being undertaken to establish the impact of the individual components within the framework (Deavers & Solity, 1998; Deavers et al., in press; Solity et al., 1999).”
(Solity, 2000)
I had a series of lectures by him in about 2002 when I was at Warwick University. He said that he couldn’t test individual aspects of the program because that wasn’t very easy with classroom-based research but that each aspect of the program was based on imperial research. At the time the concept of small-unit phonics teaching was far from accepted, so this was probably fair enough.
Although some of the research seems fairly solid, some of it is a bit shaky. For example, his views about use of real books are justified by this: Distinctiveness of Spoken Word Context Predicts Visual Lexical Decision Time (1997). Doesn’t sound much like a classroom comparison between using real books or not. His arguments about distributed rather than massed practice don’t reference research about reading or even children; instead he cites a psychology textbook describing a study into teaching post office workers to touch type postcodes! To be fair, he has gone and done specific research into this aspect (Distributed and Massed Practice: From Laboratory to Classroom.) However, that wouldn’t be the first part of the program I’d have spent time researching.