Why I think we need a national spelling survey among schoolchildren

The first question people ask about spelling is, "Is it getting worse?" The problem is how to give a proper answer. We can look back at fond memories of our own stunning success or failure at school. Or we can try to see whether there is any factual information on which to base an answer.

One analysis of Standard Assessment Test scores indeed found that children in 1996 did worse on spelling than in 1976, in so far as they could be compared. The National Foundation for Educational Research found that children aged 15 still made a high proportion of spelling errors. Only two out of ten had no mistakes in ten lines of an essay. Mostly they were omitting letters as in "recuring", transposing letters as in "freind", or substituting one letter for another as in "definate". On the face of it, an alarming statistic. But can we really say that this is better or worse than 15-year-olds of earlier days if there are no data to compare them?

This level of mistake seems not unusual from people writing by hand without the chance to check. The first ten lines of the manuscript of Keats' Ode to Autumn include "furuits" (fruits), "sweeness" (sweetness), "hazle" (hazel) and "wam" (warm). From Keats we may call them slips of the pen. But why should Keats have an excuse and not the children? Spontaneous use of language allows a number of slips if the person can't or doesn't edit. In everyday speech such mistakes are simply ignored. Somehow in writing they loom far larger.

A major change in progress is the switch from the pen to the keyboard and from email to text. While today's children may perform less well on a particular spelling task than their predecessors, it is not known how skilful they are at these newer forms of spelling, probably far more so than any adult. But the crucial test of children's spelling is whether education is equipping them for the real use of English today.

Another aspect of children's spelling that has been little studied is how it relates to a child's accent. A major part of spelling is connecting letters and sounds - the letters "r", "e" and "d" go with three sounds to make up the word "red". But obviously this depends on a child's accent. If the child's speech differs from the one taught in the school, there is no longer the same connection between the letters and the sounds. American children who speak the Hoosier dialect of Southern Indiana don't distinguish in speech between "i" and "e". So they naturally spell "king" as "keng", "men" as "min" and "big" as "beg". Perfectly logical in their own dialect but something they will have to overcome to write standard spelling.

The same is true in England. Recent research shows that the majority of children under seven in East Tilbury in Essex preferred to spell "thing" as "fing" and "nothing" as "nuffing". In writing, they spelt "wall" as "wow", "pool" as "pow" and "thumb" as "fum". But how extensive is the link between accent and English spelling problems? There must be similar issues with local accents in different parts of the UK, for example places where the "r" in "card" is pronounced, say the West County or Scotland, compared with the standard British absence of spoken "r" before consonants.

The overall issue of whether spelling is in decline is then unanswerable without major research. We need to know how children are using spelling today across the country and how this varies according to the kind of English spoken by the child. In other words, we need a national spelling survey to see the strengths and weaknesses our children have and how they vary from place to place. Only after this will we be able to say whether spelling is getting better or worse.

Vivian Cook
Chair in applied linguistics
Newcastle University

Vivian Cook's book Accomodating Brocolli in the Cemetary is published by Profile Books on September 2, £9.99.