Yacapaca

The Battle of Waterloo, it is said*, was won on the playing fields of Eton. Building character, not imparting knowledge, was traditionally seen as the primary task of the British public school and, subsequently, its state-funded inheritors. The quintessential attitude was sportsmanship. Not winning or losing, but playing the game right.

As time went on knowledge and skills came to be valued equally, as they should be, but somehow that pendulum just kept on swinging and now the development of character gets little more than lip service. Time for the pendulum to swing back.

Enter Dr Nigel Newton, and a super article he wrote recently promoting Character for learning & life in schools. The whole read is well-worth your time, but here’s the core; qualities, values and learning dispositions collated into a table that allows for practical syllabus-building. I am pretty sure you will be nodding along as you imagine how this might be implemented in your school.

But.

But, but, but. The very large bluebottle in the ointment is that competition between schools is measured in results. Exam results, not the hypothetical ability to win 19th Century battles. If Character for Learning and Life is to find space in a crowded curriculum, it has to be examined, and lead to a qualification. Impossible? I don’t think so.

I would like to lay out for you how I think it might be done. I should be clear that I don’t intend to do this; my aim is to demonstrate to you that by using modern tools, we can assess attributes that were previously thought un-assessable.

How to assess character, reliably and fairly

You may be known (I hope you are!) to be ‘a good sport’, or ‘of good character’. That is to say, you have a reputation. Reputation can be somewhat amorphous, but it does not have to be. Philosopher Cory Doctorow proposed Whuffie as an hypothetical digitally-enabled ‘reputation currency’. This has found practical expression in Stack Overflow, the coders’ community. If you are a software engineer, your next job depends far more on your Stack Overflow reputation than it does on any paper qualification. Reputation is already being measured and recorded; we just have to apply this to character rather than skill. Here’s how…

Let’s take the very first value from the table above: conscientiousness. Don’t think about students, but rather about one of your own peers; a fellow teacher perhaps. If you know them well, you already have a well-formed opinion of their conscientiousness. Note, please, that this is a comparator. You see this person as more or less conscientious than average.

Now imagine that I set a task for you and a group of peers. At the end of the task, I ask you to rank the other participants according to how conscientious each has been. I may also ask you to rank them on other criteria relevant to the task – empathy, say. I keep these rankings secret.

Your syllabus consists of multiple tasks or projects, each with a different group of peers. At the end of each, a ranking exercise. At the end of the whole course, we have a huge pile of data that we throw into an algorithm known as Thurstone Case 5, to produce an overall ranking.

There is some additional data which must be added. We want to assess the quality of your ranking judgements. The ability to assess the character of others is in itself a valuable character trait. In addition, we wish to encourage diligence in the ranking process by rewarding it. These data, too, will contribute to the final grade.

You may be wondering how we convert the rank to a grade. We need to include the results from your school into a national or global results set. The problem is that we cannot assume that the average levels of two separate schools are the same. Ideally, we use inter-school competitions (perhaps piggy-backing on sports fixtures) to gather data on this. Using this data for levelling between schools is the sort of task statisticians love. Converting a global ranking to grades is fairly easy; you just have to decide where the grade boundaries lie.

The cynical may object that a peer assessment process is open to cheating; wealthy students could simply buy high ranks from their peers. We can actually make this very difficult. The ranks given are secret and students are given progression data only in aggregate and after a randomly-determined delay. There is no way to know if the ranking you ‘bought’ was actually delivered.

Additionally, statistical tools can be applied, such as correlation to grades in other subjects, and anomalies investigated.

If you are an experienced Yaquapacista, you will already have recognised this as an application of our patented Structured Peer Assessment technology. Creating or implementing a suitable syllabus is well outside the scope of our organisation but I hope I have demonstrated that if it were to be done, then it could be assessed.

A-level Good Character is perfectly possible and would hugely help to re-balance our knowledge-skewed education system. Be a good sport and do your bit to make it happen.

*Apocryphal. Wellington staunchly denied having said it, and hated Eton anyway. But it’s a great quote.