And, as an aside in passing, there is a workable alternative to using the discrepancy criterion. In response to the understanding that an IQ/achievement discrepancy indicates nothing particularly useful, and certainly nothing ‘diagnostic’, a different approach has been developed in the USA. It is known as RTI (Response To Instruction) (See Fletcher 2005, Fletcher, Francis, Morris & Lyon 2005, Fletcher, Denton & Francis 2005 & Kovaleski 2004). RTI avoids psychometric testing (in particular IQ) and focuses instead on a student’s response, in terms of learning, to instruction received. This has the advantage of relating to an individual’s progress or otherwise as it happens (rather than after failure has already occurred), thus being child-centred, immediate and relatively easily operationalised in the real world. It is based directly on practice, so leads directly to appropriate teaching responses. Indeed, as its title indicates, it focuses, quite deliberately, on instruction as well as response. Where there appears to be a problem, educational consideration is immediately drawn to the teaching offered to the student as an individual as well as to his or her learning.

In the words of Fletcher, Denton & Francis (2005 p. 545) ‘Hybrid models combining low achievement and response to instruction most clearly capture the LD [learning disabilities] construct and have the most direct relation to instruction’. (In my opinion it is unfortunate that the RTI tool should be used to continue the search for disabilities, so perpetuating the deficit approach and its attendant maladaptive attributions, but it is also my opinion that RTI does not actually have to be thus. The tool could simply be deployed to find, and address, any learning difficulty in any subject. I imagine that many of the better teachers subvert it exactly thus, in real classrooms.)

Some writers use a different discrepancy to diagnose dyslexia - that between chronological (actual) age and ‘reading age’ - usually demanding a discrepancy of two or more years (e.g. Williams & O’Donovan 2006). However, such a discrepancy is merely a sign. It is just a symptom. It says nothing, in itself, about the underlying cause or nature of those influences which might be causing it. It is like observing that at any one time a certain percentage, say 10%, of people are likely to be lame and then going on to claim that this indicates that 10% of us has a sprained ankle. The one fact does not at all reliably indicate the other. Thus it is with ‘dyslexia’. If someone has a chronological age / reading age discrepancy it tells you only that they read poorly for their age. It tells you nothing of why this is so. We may not properly deduce any aetiology from the discrepancy itself. It is ‘diagnostic’ of nothing whatsoever other than an unfortunate and as yet unexplained discrepancy.

Those few researchers who attempt to diagnose dyslexia without reference to a discrepancy criterion are beginning to use performance related signs which they claim are pathognomic. There are many tests of literacy abilities to choose between, including single word reading, spelling, pseudoword reading, irregular word reading, homophone/real word choice, homonym choice, phoneme manipulation, rapid automatic naming and reading speed - all mentioned, for a single example, in Paracchini et al (2007). They go on to write that ‘Unfortunately, there is not universal agreement on which precise ascertainment criteria and psychometric tests should be applied, and different research groups typically use a specific selection of them, often dependent upon the language of the population under study’ (ibid. p. 59). There is no consensus, as you can see, as to which test, or which battery of tests, enables a ‘diagnosis’ reliably to be made. Researchers literally pick and mix. ‘Performance deficits’ selected by the researcher are said, by the researcher but without evidence (as there is none we can rely on), to be typical of and pathognomic for dyslexia. Researchers differ widely as to which tests are appropriate and what their results may indicate. Such highly subjective criteria for sample selection and such individual interpretation of findings will, however, not stop considerable reliability of diagnosis, validity of result and comparability of conclusion being claimed.

A further means of ‘diagnosis’ of dyslexia and selection of sample ‘dyslexics’ is simply to throw in the sponge, deploy the ‘bell curve’ of reading ability and define those in, say, the lowest 10% as ‘dyslexic’ (e.g. Olson 2006, Paracchini et al 2007). As you will know, the normal distribution curve, or bell curve, is the curve which can be plotted for any attribute which is normally distributed across a population (height is the usual example). The curve looks like a bell, hence the name. Reading ability is normally distributed. If you plot reading ability across the population you get the familiar bell shaped curve. It is easy to select, say, the bottom 10% of such a population from their results on reading tests and consequent place on the curve. As Paracchini et al write ‘RD [reading disability] represents the lower tail of a normal distribution of reading ability found in the general population’ (ibid. 2007 p. 59). Kate Nation (2006 p. 2) reaches the same over-extended classification when she writes about ‘… individuals who are at the low end of distribution – individuals who are reading disabled’. Olson further claims that ‘the positive consequence of the bell curve in reading research is that it allows us to apply powerful statistical methods in our genetic analysis of dyslexia and individual differences that depend on normal distributions …’ (Olson 2006 p. 3).