Literate, or just more computer literate?

Our spelling is getting better. But why, wonders William Hartson

The A-level results published last week showed improved pass rates on last year. Indeed, we are told they have been improving for the past 15 years, but the debate continues about whether this indicates a general improvement in the country's education or is merely a sign that the exams are getting easier.

Well, there's one simple way to tell: spelling. If education is getting better, then our spelling should surely be improving too. We have accordingly monitored the frequencies of certain common misspellings in the national press over the past three years. The table below records the misspelling rate per typical 1,000 usages in each year:

Word 1994 1995 1996

accommodate 19.9 17.5 14.6

embarrass 7.0 9.7 7.3

consensus 19.2 11.4 16.1

inoculate 203.9 153.3 132.7

minuscule 224.6 156.7 89.3

pavilion 14.5 14.9 15.7

separate 5.8 5.1 4.8

supersede 101.4 98.6 79.4

The results confirm a steady improvement in our spelling, with five of the eight words registering significantly fewer misspellings so far this year than in 1994, and the other three only marginally worse. The curious blip suffered by "consensus", which had such a bad year in 1995, is difficult to explain.

Despite the steady improvement in "inoculate", this word (and its derivatives) has now displaced "minuscule" from the top of the most frequently misspelt list. For minuscule to have dropped from a rate of greater than one in five to less than one in 10 misspellings is something in which the nation can take pride.

Or can it? Might not the above simply be a reflection not of our educational standards but of our increasing use of spell-checkers? To test that hypothesis, we have monitored the frequency of the word "spell-check" and its slightly less common variant "spellcheck" (without a hyphen) in the same newspaper database.

Word 1994 1995 1996

spell-check 12 17 12

spellcheck 20 17 18

This gives a total of 32 incidences of spell-checking in 1994, which grew slightly to 34 last year and, at 30 so far this year, may be expected to reach a figure close to 50 by the end of December. If this 50 per cent increase in the use of the term correlates to a similar increase in the use of spell-checking, this may account for all the improvement in the first table. So even if our general education is not improving our ability to use computers is. There would be, at least, some comfort in that.