I have frequently written about the need to decouple learning and credentials, so I love this approach to doing so from Brunel University. It fully decouples learning and credentials by offering ungraded study blocks (in North America the equivalent of courses, in the UK the equivalent of modules) with no summative assessments, followed by integrative assessment blocks, that provide opportunities for students to pull together what they have learned across their various courses/modules in a variety of (mostly) useful integrative learning activities for which marks are awarded. It's neat, simple, practical, and effective.

The summative assessment load (for students and their professors) is reduced by more than 60%, the quality of those assessments increases (in every way), students feel better prepared for employment (and employers agree), it improves retention figures, teachers can focus on teaching, assessments are more authentic, more engaging, and it massively reduces cheating. The only significant downside that I can see in this is that it is not quite as flexible as a completely modular program - there are a few dependencies and limits on when and how students learn, albeit that these are no worse than in most in-person universities.

I learned about this from Peter Hartley, who mentioned it in a quite inspiring IFNTF talk on assessment yesterday. Amongst other things, Peter highlighted a wide range of issues with modularization (i.e. the standard approach used in many parts of the world of splitting up a program into a set of self-contained courses) and assessment, including, from his slides:

1. Not assessing programme outcomes.
2. Atomisation of assessment.
3. Students and staff failing to see the links/coherence of the programme.
4. Modules too short for complex learning.
5. Surface learning and 'tick-box' mentality.
6. Inappropriate 'one-size-fits-all'.
7. Over-standardisation in regulations.
8. Too much summative assessment and feedback - not enough formative.

While I couldn't agree more, for the most part, I have mixed feelings about some of Peter's list of issues. I agree that the traditional 3 or 4 year program(me), in which the course of study is designed to work as a whole, not as a collection of self-contained pieces, is far better for integrating knowledge across a discipline, though I don't see why it should always take exactly that amount of time to achieve mastery, and I am not even sure whether we should be thinking in terms of disciplines at all. There's some value in the notion, for sure, and there are some kinds of subject and learning for which it makes sense, but I think a lot of it is down to centuries' old tradition and post hoc justification rather than careful consideration of fitness for purpose. Also, it seems to me that summative assessment should always be formative, too, so the issue could be partly addressed by simply improving summative assessments, not by scrapping them altogether. However, I think Peter is fundamentally right that, due to modularization, most universities over-assess, that credentials become the reason for learning rather than the measurement of it (with all the very many evils that entails), that the big picture tends to be lost, that there is a ridiculously large administrative burden that results from it, and that learning - the point of the thing after all - consequently suffers. As we and much of the rest of the world start to move towards ever smaller chunks, with associated stackable microcredentials, badges, etc, this is going to be a bigger problem. Brunel's solution is not the only way, but it is a radically disruptive intervention that that many universities could implement without breaking everything else in the process.