The Proposed Teaching Excellence Framework (TEF) – Some Thoughts on Value Added as a Metric

The purpose of teaching is learning. We measure learning mostly through qualifications. It’s easy to compare a student’s qualifications at entry with their qualifications at exit, if we want to. All we need is a common measurement scale that embraces entry and exit qualifications, differentiating between different grades where applicable. 

Maybe something like this:

A-level – grades A* to E, 100 down to 50 points 

(A-level points for a student would of course be summed across the subjects they took.)

Cert HE – 600 points

DipHE / FD  – 800 points

First Degree, 1st class down to pass

These points are not the same as credit points, which describe number of hours of study rather than qualifications attained.

We might also include higher taught degrees – the same logic applies, I think. 

The actual numbers are arbitrary, although they are ordered – higher always means higher. But this arbitrariness doesn’t matter, as long as the entry and exit qualification scales don’t overlap. All we will be doing is subtracting points at entry from points at exit  – that is, identifying the value added for each student. These differences, these measures of value aded per student, will then be averaged by institutions, Schools, programmes, whoever’s value added we are currently interested in. And, once calculated, they can be compared. We are only interested in difference.

There’s a choice to be made about how we deal with completion rates. Is completion rate a measure of teaching quality, or a measure of how life affects students? Or both? I’m inclined to suggest that we only compute value added for those who complete, which may mean achieving an intermediate qualification, typically CertHe or DipHE.  As always with metrics, we need to be explicit about what we are proposing. There is not always a right answer.

This approach to value added only use existing data – no new data need be collected. Which is good.

Of course, what have historically been called ‘recruiting’ universities would score  higher on value added than ‘selecting’ universities. But the resultant debate would flush out views about the value of degrees from different institutions. These views may or may not may not survive sunlight. But these views are probably best expressed and tested. 

Anyway, value added is only one element of the framework. But it’s a tough one to argue against.

The problems with measuring value-added are not technical. They are political. But let’s have the debate!

This post was prompted by:

“No convincing metrics have been offered so far on how to measure added value without involving time consuming pre-and post tests.” in http://sally-brown.net/wp-content/plugins/download-monitor/download.php?id=238  from Sally Brown and Ruth Pickford, from here, and the subsequent LTHE Tweetchat, Storified here.