In an earlier blog post, i took this political slogan, and interpreted in terms of league tables, writing:
I’ve given presentations to two of our faculties on understanding league tables, and what we need to do to improve our position. I always ask people where in a table do they think we should come. The answer lies between 50 and 70, always. Bearing in mind that our position in any of the tables, no matter what the methodology, is not this high the maybe we can take away the following two ideas. Firstly, that there is a will there to work together to change things, and a recognition that improving our position will help with our own feelings of self worth.
The key point of my talk is to explain that league tables are not “something that is done to us”, rather they are just a mirror held up to show us who we are. And if we are uglier than we want to be, then we need to start to do something about it.
Essentially there are two sets of data that go to make up the tables – input and output data. Like all universities, we are now making sure that our input data – staff student ratios, spend per student, entry grades are reflecting us in the best possible light.
A few months on and it’s worth revisiting this, in the light of work I’ve been doing on portfolio performance, and in anticipation of the next round of league tables.
We could now move to a position where we start forecasting more accurately where we are likely to be in a league table, how we should target specific indicators and where our improvements will have a positive effect.
Using the data in Heidi (2 weeks in and I’m still discovering new things with it) we can use the most recent student and staff records, which were added in February 2014, to do two things:
- consider how we have changed since the previous year, and show trends
- consider how we have performed against the sector and in particular our comparator group
This isn’t rocket science – I’ve seen plenty of similar outputs from the planning functions of other universities.
As well as developing better business intelligence using this public data (which could easily be linked to UCAS for undergraduate…) we can use the data, and some very simple stats (the same function that league table compilers like to use) to start to look at how well we perform in the market, relative to others. The simple graphs below give an example of this:
From these simple visualisations, we can see how subject 1 outperforms in terms of market against comparator universities, whereas subject 2 is less prominent, This in itself should give rise to questions about what shape an award portfolio should be.
To make this one step more sophisticated, the same datasets could be used to provide data on student attainment, student entry characteristics and staffing, all of which will enable us to build a predictive model of performance benchmarked against all other institutions.
For a university to fully understand how it performs in league tables, then the first step is to understand how the datasets were compiled. What I am suggesting here, is that by understanding the data more fully, and by comparing in detail against competitors, a university should be able to identify where action can be taken to improve position.
It cannot be emphasised enough though, that league table position is simply a reflection of ourselves, and it should not become the be-all and end-all. We must not end up chasing numerical targets, if that means that we forget the broader goals of a university. I come back to my first point – we can be better than this.