The perils of league tables

Cambridge summer rituals don’t end with the exams, or even with the fancy dress of degree day. The final pantomime is the college league tables. The version that gets most publicity is the “Tompkins Table”, which is basically a first-past-the-post style of ranking. It gives 5 points for a first, 3 points for a 2.1 and so on. Then it produces a score for each college by reckoning their actual points total against what they would have got if every student had been awarded a first. A bit of “weighting” goes into the process, to cancel out the undue influence of subjects which score a lot of firsts. Otherwise any college could “win” by packing the place with (male) mathematicians.

This isn’t an official university ranking (we do have our own internal version, the “Baxter Tables” driven by every imaginable statistical correctness – but they are not made public till September). Peter Tompkins is a Cambridge Maths graduate (what else?) who compiles this table for fun.

Our public line is that is not something that much bothers us. Winning colleges may blazon their success on their website (but even some of those affect a more lofty disdain, and boast only in private). Winners and losers alike join together in patiently explaining that this is all very misleading, that it gives an inappropriate weighting to firsts (when for most of the students it’s a 2.1 that counts), and that anyway very little separates the top from the middle (if not the bottom).

But underneath we’re all a little bit more anxious.

The truth is that we all want our colleges to win. Many of us have grown up in this competitive system – and show us a race and we’re programmed to want to come in first. Eager Senior Tutors across town are rumoured to make efforts to “weed out” (whatever that means) the thirds. Some of them try to enhance the performance of their flock by closing the College bar early at exam time (as if the kids didn’t have the wit to go to the local pub). The idea of giving the best rooms in college to those who get firsts (rather that the captain of the boat) is coming back into fashion.

None of this matters very much. And, in fact, this sense of college loyalty and inter-college rivalry is one of the things that helps break down a very big University such as Cambridge into manageable and productive smaller units. But, all the same, I have some nagging unease.

First, these tables rank our students on the sole criterion of exam grade just as they leave the university. That’s one point to test them. But actually I’m more interested in what it all looks like 10 or 20 years on. Will all these clever firsts seem like the college’s greatest successes after a decade or so. Some yes, some no. But it’s harder to measure achievement once exams are over, and harder to know what contributed to that achievement. And you can be certain that a few of those blots on the Tompkins Tables today will be those we will be boasting about tomorrow. Take a look at our websites and ask yourself how many of the “famous alums” we are so proud of got firsts. Not all that many.

But the other worry is what it would do to our admissions practice if we did take it all too seriously. Obsession with League tables inevitably leads to a low-risk strategy in who you choose to study here in the first place. It means not daring to take a gamble on some sixth-former you are more or less convinced is brilliant but hasn’t quite got the exam results to prove it. The best universities are those that have the confidence to make mistakes every now and then (just as the best hospitals happily tolerate a few octogenarian bed-blockers). If the League table culture stopped us taking risks, Cambridge would be a worse university.

My college, in case your wondering, normally appears at the “lower end”.