É«ºÐÖ±²¥

There can be no better time than One Nation Tory Week to note the continuing relevance of Disraeli's even better-remembered coinage - the contention that there are three types of lies: "lies, damned lies and statistics". This is not to imply that anyone is lying in the debate over drop-out rates. But there are almost as many statistics as there are anecdotes. What it all means is debatable.

The latest contribution is the Committee of Vice Chancellors and Principals' Survey of Student Financial Support, published to coincide with this week's Commons debate on student loans. This recognises that there is a problem: drop-outs have increased by almost two-thirds over the past four years. Why? One explanation may be that growth in modular courses and transferable credits is allowing students to drop in and out at their convenience.

Another is expansion. One reason traditionally given for Britain's historically low drop-out rates is that barriers were erected pre-entry. Increase the intake, while maintaining the exit standard, and arguably more people should fail. Those who think, like Kingsley Amis, that we have "tapped untalent" will doubtless be confirmed in their view by the fact that numbers of drop-outs are growing twice as rapidly as the intake and academic drop-outs have grown twice as fast as other types over the last year.

And, of course, lack of cash must make a difference.

The odds are that dropping out has many causes. Our front-page graph shows exam failure now overtaking other causes but academic failure itself is rarely monocausal. Anything contributing to pressures, or taking up study time - debt, part-time work, family disruption - is likely to impact on academic performance and ultimately on drop-out rates. High attrition among mature students, generally highly motivated but particularly prone to such pressures, seems to support this.

But we do not know. Dropped-out students are hard to find and apt to tailor answers to the questioners' perceived agenda. Until the effort is made to find them, and to establish why they go, the debate will flounder amid anecdote, rhetoric and special pleading. Calling for more research looks like a kneejerk academic response. But in this case it is the right one as the Higher Education Funding Council for England is aware. Its deadline for bids to study the issue was last week. Subject to maintaining research quality, they should now look for rapid results. With firm data perhaps the Government will listen - though it would be unusual.