Alison Wolf

It appears that many skills associated with high IQ are required for activities such as computer gaming and surfing the net

Have you noticed how the national press has obsessions? At any one time, there will be a few fashionable topics on which any "new" stories are guaranteed blanket coverage, while other whole areas barely register.

For universities, top billing currently goes to fees, student debts and discrimination in admissions. Meanwhile, our steadily rising tide of 2:1s and our plagiarism worries hardly make it beyond the inner pages of The Times Higher . For schools, on the other hand, examination standards and cheating are the main stories. With summer upon us, they will soon reach their climax.

If you have ever worked professionally on public examinations or assessment techniques, summer brings phone calls from journalists desperate for a quote. Does the record number of A grades at A level prove that standards have fallen (again)? The technical answer is that it is pretty well impossible to tell: subject matter and question types change so much over time that you are not comparing like with like. Also, students may work harder; teachers may teach better; and revision guides may have improved.

As an experienced observer, your informal verdict might be a bit different.

Some subjects are almost certainly easier, some have got harder, some seem about the same. But to tell a journalist which ones, on the record? Even academics weren't born yesterday.

Many of these points apply to university standards, too (even though they are not, yet, a story), as does the likely impact of teaching time. In the sixth form, this has fallen with modularisation; in universities, it has fallen still more. On a bad day it is hard to believe that we ask (or teach) as much as we used to, and The Times Higher surveys always elicit plenty of gloom.

Because it is so hard to say anything definite about standards, it is particularly intriguing to be told that, in one area, things are crystal clear. People are getting cleverer. Or, to be more precise, people are getting steadily better at IQ tests.

This finding has been publicised recently in a defence of popular culture by Steven Johnson, who argues in his book Everything Bad Is Good for You that by watching TV or enjoying electronic mayhem we develop these skills.

But the trend was first noticed back in the 1980s, by James Flynn, a professor at New Zealand's Otago University, and is now well documented across the developed world.

Moreover, this is not because the tests have changed. IQ tests do not report raw scores. They measure your performance relative to the rest of the population. They get re-normed regularly, as items are turned over, so that a mid-point score will continue to be reported as 100. However, a large number of items will stay in a given test for years, and new ones are equated carefully with older ones, so researchers can also track absolute performance over time. There seems to be no doubt: people have got better at answering the type of question that IQ tests ask.

No one really knows why this has happened. Computer games have always been my favourite candidate, though trying to follow the dialogue in The West Wing might hone one's IQ faster still. Deciding whether this means we are getting "cleverer" is a bit like measuring standards: a fraught and possibly hopeless task. However, IQ tests certainly measure some real skills - abstract problem-solving or pattern-spotting are common descriptions.

These are just one set of skills among many, but IQ is also a pretty strong predictor of success in many jobs. A whole barrage of psychological studies shows high correlations between people's IQ scores at the time they were hired and their later success. Strictly speaking, such correlations just show that those who are relatively good on one measure (IQ) are also relatively good on another (job performance). It could just be coincidence or that both are related to something else that we have not spotted or cannot measure. But with results as consistent as these, it is hard not to believe that IQ skills also make some substantive contribution to doing your job well. In which case, we are in one real, if limited, sense getting smarter. Companies might want to reconsider what, in their view, is an acceptable IQ score.

For society and for education, there is another wider implication. These results do not just apply to a few favoured people receiving elite education and above-average resources. If it is demonstrably possible for average levels of some quite abstract and demanding skills to rise, then why not others, too? It should be a priority to understand how this happened. And if IQ skills have risen, then why not mathematical skills, too?

Alison Wolf is Sir Roy Griffith professor of public sector management, King's College London.