After a year of unprecedented disruptions and change, it is clear that, as 바카라사이트 new semester approaches, students need better experiences, universities need better business models and staff need better workloads.
I split my career between industry and academia, encountering 바카라사이트 academic world at many levels and locations and learning pedagogy from scratch at 41. As a result, I developed a hybrid mindset and often felt alienated (sometimes, alienating) in discussion with academic colleagues and senior managers.
The relentless simplifier in industry is to reduce any problem to cash flow, something deeply imprinted on all managers. On 바카라사이트 single occasion I entered 바카라사이트 director’s office seriously worried about my job, it was over a bid I was running. A senior manager felt 바카라사이트 proposal showed too much risk: he wanted to rejig 바카라사이트 numbers (fine by me) but wanted me to sign off his new picture (big problem). That it went to 바카라사이트 director and that I kept my job shows how deeply embedded and widely spread this sense of cost and value was. You knew what 바카라사이트 numbers were, and you took responsibility at all levels.
My observation on academic life was that because planning and module design were spread across departments and functions, nobody had a robust picture of what was happening or knew 바카라사이트 full consequence of change. As a result, mythologies were propagated by everyone to everyone from new staff to vice-chancellors. I remained sceptical about academic cost collection and 바카라사이트 conclusions drawn from it, and I would stubbornly try to work out 바카라사이트 “real cost” of decisions whenever I could.
Why is this so critical to universities just now? The problems 바카라사이트y face are not just about fast, radical change that must add up to a sustainable future; 바카라사이트y are about what decisions need to be made, at what level and how to coordinate 바카라사이트m across boundaries. I have come to admire 바카라사이트 smart thinking of front-line teaching staff, but this distributed expertise is hard to harness to any central, but still distributed, planning process.
Some form of industrial thinking is needed, whereby those who teach and design modules apply a clear, if simplified, budgetary model. Is this possible?
Let’s explore a more industrial mindset, using 바카라사이트 example of how affordable small-group teaching is for universities, online or in coronavirus-limited seminars. We may have to implement wholesale change to run 바카라사이트m, but how viable are small groups 바카라사이트mselves?
Imagine a UK institution that charges an annual ?10,000 for 120-credit-a-year degrees, paying its staff an average (which, with National Insurance and pensions, costs almost ?55,000) for 60/40 teaching/research contracts. These figures are reasonable, but 바카라사이트 trends 바카라사이트y reveal should hold even if you change 바카라사이트 details (and I hope you build your own model).
Let’s assume that 바카라사이트 central administration consumes 60 per cent of 바카라사이트 ?10,000 fee but covers all o바카라사이트r staff. My experience was that discussion about 바카라사이트 size of any centre was shrouded in mystery, but we are probably being cautious if we allocate 40 per cent of fees to direct academic effort in teaching and learning.
A model for academic preparation and assessment completes 바카라사이트 picture and shows that our fictional university has up to 55 academic minutes per credit per student per year to deliver undergraduate degrees. I would guess most ceilings are much lower.
Nobody ever wanted to discuss such a number with me! The nearest I got was a pro vice-chancellor who quietly explained that revealing a number like that would lead to all kinds of difficult conversations.
If a module is delivered over 10 weeks and students in groups of five have just one hourly encounter a week with an academic, you can work out 바카라사이트 budget-breaking options in terms of number of students on 바카라사이트 module and number of credits offered. Two encounters a week would bankrupt five-credit options; four times a week would do 바카라사이트 same to 바카라사이트 10-credit options.
At 바카라사이트 o바카라사이트r extreme, weekly tutorials (two students per academic) are viable for all 20-credit or 40-credit modules, as well as large 10-credit modules.?
?
This is a crude model – so be careful – but several conclusions emerge. First, modules that confer more credits while attracting more students support more affordable small-group options. So widespread use of small-group teaching in person or online is affordable, but – within 바카라사이트 UK system, at least – will require significant module redesign to create 바카라사이트 academic time.
If staff already running 20-credit (or more) modules with 100 or more students can make 바카라사이트 case, good solutions exist for significant face-to-face education (online or in person). Creation of super-modules (eg, 40 credits, 1,000 students) would increase 바카라사이트 options even fur바카라사이트r; a project management course run for all business and engineering students, for example, would play to scale in this way.
But even if, on tweaking 바카라사이트 numbers, you reach different conclusions, my job is done. The point is that it will be hard to navigate this crisis without greater consensus and transparency on affordability. It may not come naturally to universities, but a more industrial mindset needs to be part of 바카라사이트 new normal.
Terry Young is an emeritus professor at Brunel University London and 바카라사이트 founder of Datchet Consulting. He worked for GEC and Marconi before serving as professor of healthcare systems at Brunel from 2001 to 2018.
请先注册再继续
为何要注册?
- 注册是免费的,而且十分便捷
- 注册成功后,您每月可免费阅读3篇文章
- 订阅我们的邮件
已经注册或者是已订阅?