Monday, 24 July 2017

Reflections on the UK 'Carnegie Classification'

I posted yesterday on how UK universities and higher education institutions would map onto the US Carnegie classifications. That post simply presented the data; its worth a little reflection on what the data show.

Unsurprisingly, perhaps, the Doctoral classification is the biggest, but there is more differentiation than I expected. When I did a similar exercise 15-20 years ago there weren’t any UK HEI’s in the Masters or Baccalaureate categories - every institutions was R1 or R2 (in the old Carnegie scheme). This reflects the broadening of the pool in the UK (there are more universities now than there were); but also, perhaps, the more selective approach to funding of research studentships. Some of the institutions which are now Master's universities made the Doctoral cut, if I remember correctly, on the old classification.

Also not present in this list are HE and FE Colleges, which between them would occupy the Baccalaureate category, the Baccalaureate/Associate’s category and the Associate’s category. In the UK the equivalent of the Associate's category is Foundation Degrees and foundation years, but the principle remains the same. Its clear that if you want to understand the breadth of UK HE you need to look at HE delivered in FE Colleges. This is a challenge to most of the usual narratives about UK higher education; perhaps it reflects the university-sector 'ownership' of some sector wide bodies such as UCAS, HESA etc. (I'm using 'ownership' loosely here.)

And of course, as the UK did most of its historical oppressing in countries which are now sovereign, there isn’t an equivalent of the Tribal Colleges category.

Is this a useful analytical framework? At the moment the categories used are often driven by mission groups. Whilst membership of these is in part driven by data, it isn't transparent. And as mission groups are clubs not leagues, there isn't often relegation, although promotion does happen. (Remember the expansion of the Russell Group in 2012). Perhaps we need a UK equivalent, to allow for more transparent analysis of the section and how it is developing?

Sunday, 23 July 2017

Mapping UK universities against the US Carnegie Classifications

In 1970 the Carnegie Commission on Higher Education developed a framework for analysing university and colleges, to facilitate research and policy development on higher education. The framework has continued to be developed by higher education researchers in the US. As England’s HE policy framework begins to approximate the US – with an increasing emphasis on diversity of institution, market entry and exit, and the student as consumer – in this blog I’m looking at the US classifications, what they tell us, and how the UK’s HEI’s would map onto the Carnegie classification.

The current Carnegie Classifications – using a methodology last significantly updated in 2005 – divide universities and colleges into seven broad classes. In some classes there is further differentiation by scale of activity. These are:

Doctoral University (with subclasses R1, R2, R3 defined by scale)
Institutions that award at least 20 PhD/DPhil degrees per year
Master’s University (with subclasses M1, M2, M3 defined by scale)
Institutions that award at last 50 Master’s degrees per year
Baccalaureate Colleges
Institutions where bachelor’s degrees made up more than 50% of degrees awarded
Baccalaureate/Associates Colleges
Institutions with at last one Bachelor’s programme and with more than 50% of awards at the Associate degree level
Associate's Colleges
Institutions whose highest qualification awarded was an Associate degree
Special Focus Institutions
Institutions where more than 75% of degree awards relates to a single field or set of related fields
Tribal Colleges
Institutions which are members of the American Indian Higher Education Consortium

The first four categories are applied hierarchically: if you’re doctoral, you’re not counted in Masters; if you’re Master’s, you’re not counted in Baccalaureate, even though numerically you’d meet both sets of criteria.

An associate degree is a two-year undergraduate qualification. Typically it would equate to the first two years of the four-year baccalaureate degree. 

Using HESA data from 2015-16, it’s possible to match UK institutions to these categories. I haven’t done the detailed analysis required to categorise Doctoral universities as RE1, R2 or R3; or Master’s universities similarly. 

All UK HEI’s – or at least those which reported to HESA in 2015-16 – fall within one of the Doctoral, Master’s, Baccalaureate or Specialist classes. Here’s the classification. 

Doctoral Universities

  • Aberystwyth University
  • Anglia Ruskin University
  • Aston University
  • Bangor University
  • Birkbeck College
  • Birmingham City University
  • Bournemouth University
  • Brunel University London
  • Canterbury Christ Church University
  • Cardiff Metropolitan University
  • Cardiff University
  • City, University of London
  • Coventry University
  • Cranfield University
  • De Montfort University
  • Edinburgh Napier University
  • Glasgow Caledonian University
  • Goldsmiths College
  • Heriot-Watt University
  • Imperial College of Science, Technology and Medicine
  • Keele University
  • King's College London
  • Kingston University
  • Leeds Beckett University
  • Liverpool John Moores University
  • London Metropolitan University
  • London School of Economics and Political Science
  • London South Bank University
  • Loughborough University
  • Middlesex University
  • Newcastle University
  • Oxford Brookes University
  • Queen Margaret University, Edinburgh
  • Queen Mary University of London
  • Roehampton University
  • Royal Holloway and Bedford New College
  • Sheffield Hallam University
  • St George's, University of London
  • Staffordshire University
  • Swansea University
  • Teesside University
  • The Institute of Cancer Research
  • The Manchester Metropolitan University
  • The Nottingham Trent University
  • The Open University
  • The Queen's University of Belfast
  • The Robert Gordon University
  • The School of Oriental and African Studies
  • The University of Aberdeen
  • The University of Bath
  • The University of Birmingham
  • The University of Bradford
  • The University of Brighton
  • The University of Bristol
  • The University of Buckingham
  • The University of Cambridge
  • The University of Central Lancashire
  • The University of Dundee
  • The University of East Anglia
  • The University of East London
  • The University of Edinburgh
  • The University of Essex
  • The University of Exeter
  • The University of Glasgow
  • The University of Greenwich
  • The University of Huddersfield
  • The University of Hull
  • The University of Kent
  • The University of Lancaster
  • The University of Leeds
  • The University of Leicester
  • The University of Lincoln
  • The University of Liverpool†
  • The University of Manchester
  • The University of Northampton
  • The University of Oxford
  • The University of Portsmouth
  • The University of Reading
  • The University of Salford
  • The University of Sheffield
  • The University of Southampton
  • The University of St Andrews
  • The University of Stirling
  • The University of Strathclyde
  • The University of Sunderland
  • The University of Surrey
  • The University of Sussex
  • The University of Warwick
  • The University of Westminster
  • The University of Wolverhampton
  • The University of York
  • University College London
  • University of Abertay Dundee
  • University of Bedfordshire
  • University of Chester
  • University of Durham
  • University of Gloucestershire
  • University of Hertfordshire
  • University of Northumbria at Newcastle
  • University of Nottingham
  • University of Plymouth
  • University of South Wales
  • University of the Arts, London
  • University of the Highlands and Islands
  • University of the West of England, Bristol
  • University of Ulster
  • University of Wales Trinity Saint David
Master's Universities
  • Bath Spa University
  • Buckinghamshire New University
  • Edge Hill University
  • Falmouth University
  • Glynd┼Ár University
  • Liverpool Hope University
  • Newman University
  • Royal Agricultural University
  • Southampton Solent University
  • St Mary's University, Twickenham
  • The University of Bolton
  • The University of Chichester
  • The University of the West of Scotland
  • The University of West London
  • The University of Winchester
  • University College Birmingham
  • University of Cumbria
  • University of Derby
  • University of Suffolk
  • University of Worcester
  • York St John University
Baccalaureate Universities
  • Bishop Grosseteste University
  • Leeds Trinity University
  • SRUC
  • St Mary's University College
  • University of St Mark and St John
Specialist Focus Institutions
  • Conservatoire for Dance and Drama
  • Courtauld Institute of Art
  • Glasgow School of Art
  • Guildhall School of Music and Drama
  • Harper Adams University
  • Heythrop College
  • Leeds College of Art
  • Liverpool School of Tropical Medicine
  • London Business School
  • London School of Hygiene and Tropical Medicine
  • Norwich University of the Arts
  • Plymouth College of Art
  • Ravensbourne
  • Rose Bruford College
  • Royal Academy of Music
  • Royal College of Art
  • Royal College of Music
  • Royal Conservatoire of Scotland
  • Royal Northern College of Music
  • Stranmillis University College
  • The Arts University Bournemouth
  • The Liverpool Institute for Performing Arts
  • The National Film and Television School
  • The Royal Central School of Speech and Drama
  • The Royal Veterinary College
  • The University of Wales (central functions)
  • Trinity Laban Conservatoire of Music and Dance
  • University for the Creative Arts
  • University of London (Institutes and activities)
  • Writtle University College


Monday, 17 July 2017

How many overseas students are there?

With more-than-usual amounts of chatter around UK HE policy at the moment, its useful to remind ourselves about evidence and actuality. With this in mind, I’m presenting two views of UK HE and its reliance upon students from other countries.

The chart below shows this – it’s drawn from HESA data, and shows the headcount of students – all levels, all modes – from outside the UK. This includes other EU and EEA countries, and students from all other countries.  It’s about domicile recorded for fees purposes, which practitioners will know is not a simple correlation with passport.


The data show that there’s a wild and fluctuating market, with sharp declines and even sharper rises. It must be a nightmare for universities in such a market – no wonder the emphasis given to recruitment activities.

Let’s look at a different chart. Again, students from outside the UK. The data here tell a different story – a stable market (you might even say strong and stable). Universities can concentrate on what that do best – teaching and supporting students learning; and conducting research. Our global partnerships look safe and secure.


It’s the same data of course. The trickery is in the Y axis (the horizontal one) – in the first chart the scale is truncated to start at 415k students; in the second chart the scale starts at 0.  The first gives the detail, the second the big picture. The chicanery is that the eye is tempted to focus on the line not the numbers. The conclusions drawn from the two charts are quite different.

There’s two lessons in this.

Firstly, don’t be fooled by bad charts. Darrell Huff’s How to lie with statistics is essential reading for everyone, in my view.

Secondly, the details does matter. Although total numbers are stable, the total is made up of the totals at all of the UK’s universities and HEIs – there’s almost 200 components to this, and maintaining, or growing, your numbers make the difference between adding jobs and giving a better student experience, or retrenchment, retraction and job losses.

So, look at the detail; remember the bigger picture might say something different; and try not to mistake wood for tees, and vice versa!


Monday, 10 July 2017

Cartel schmartel

The question of university tuition fees in England is causing brouhaha again. Part of the issue seems to be some Conservative ministers wanting to catch up on an issue where the Labour Party had the edge on them at the recent General Election. Part of the issue appears to be a Damascene conversion by Andrew Adonis, the behind-the-scenes architect of the 2004-06 increases in tuition fees, and in particular his assertions that universities are operating a cartel about fees. A particular trigger appears to be recent revelations about the amount of debt that students will incur and not repayhttp://www.bbc.co.uk/news/uk-politics-40547740.

At heart, a lot of the issues appear to relate to the consistent £9k fee charged by universities in England. This costs more than the government planned (an estimated cost of £7.5k per student per year underpinned the calculations in 2010 and 2011). So why do universities charge £9,000 per year? And why do students pay?

(Cautionary note: although I graduated from LSE, I am not an economist. But I don’t think that arguments I make are bad arguments. Second cautionary note – the Higher Education and Research Act and the Office for Students changes a lot of the nomenclature in this, but the fundamentals remain the same.)

(by the way, on this general topic here is a really good article from the BBC showing some data about this complex topic)

Firstly, let’s look at the question of why students pay.

The student loans scheme in England is not like a normal loan. Repayments are contingent upon income levels (ie you don’t pay until you earn £21,000 per year; and then you pay a flat rate 9% of income over 21,000 per year.). You keep paying until you’ve paid off the debt, or until 30 years post-graduation.

This means that many won’t fully repay their student loans, as their income levels between 21 and 50 aren’t high enough to have paid for enough years. It also means that for any given level of income, the amount you repay would be the same regardless of the amount of loan you took out. Let’s show how that works:

Student A borrowed £9,000 per year - £27,000 in total – to fund tuition fees at Poppleton University. They now earn £25,000 per year, and so every year they repay £360 – that is, 9% of £4,000, which is the difference between their income (£25,000) and the threshold (£21,000).

Student B went to the Poppleton Metropolitan University which charges £6,000 fees per year, so they borrowed in total £18,000. They now earn £25,000 per year, and so every year they repay £360 – that is, 9% of £4,000, which is the difference between their income (£25,000) and the threshold (£21,000).
You can see that the annual amount of repayments is not related to the amount borrowed. Total amount repaid does relate – in principle – to the amount of borrowing, but you’ll only be expected to repay all of the debt if you earn enough.

This leads to my first proposition: the loans system does not encourage students to be price-sensitive.

The second question relates to why universities charge £9,000. It’s important to understand how this £9,000 is made up. The legislation provides for a standard amount (£6,000 per year) and a variable element (initially an additional £3,000 per year, now £3,250.)

Any approved English HE provider can charge £6,000 per year. Institutions can charge an additional fee if they have an access agreement with the Director of the Office for Fair Access (OFFA).  The Access Agreement sets out what additional fee they plan to charge, and what they’ll do to ensure that this does not militate against fair access. The assumption was that only in exceptional cases would £9,000 be payable; but there was no mechanism to enforce this. OFFA had to judge each application on its merits. And so if, say, University A had got an access agreement for £9,000 fees with fair access spend of £500 per student, then University B which proposed £9,000 fee with £500 fair access spend per student would be able to argue that to deny them the right to charge £9k would be perverse. The quality of the university was not a consideration.

In that initial round, nearly every university had its access agreement approved without conditions; a few had to make revisions. But none were turned down. £9,000 proved not to be exceptional.
Why did universities even suggest such a fee level? In part because fees act as a marker for quality. If your rival charges £9,000, why would you charge less? To do so could be to indicate that you weren’t as confident as your rival in the value of your offer.  It isn’t about greed or excess; simply about market position.

So this is my second proposition: there was no incentive for universities not to charge £9,000.

Also relevant was the actual behaviour of students. When setting higher fees for the first time, many university governing bodies recognised that they were entering the unknown. There was a recognition that student numbers might well fall. So a £9,000 fee, together within internal budgeting for fewer students, would make sense. We put on a brave face for the world, but plan for hard times. As it turned out, student numbers did not (after the first year) decline. And this is the third factor which I think is relevant.

Student recruitment has historically been controlled by the government, through a funding cap. Essentially, universities had a recruitment target set by HEFCE; there were penalties for exceeding this. When the cap was relaxed and then removed, as happened in 2013-14 onwards, universities were free to recruit as many students as they liked. This meant that a university which was growing could afford to spend resources on other activities, as in most cases the £9,000 fee was greater than the full cost of teaching. This enabled development of new subjects; investment in new buildings for teaching and research; and investment to improve reputation in, for instance, the REF.

This creates my third proposition: universities had many incentives to grow; few to remain the same size.

Put these three factors together, and a lot of the features of the current system become clear. £9,000 fees were the natural desire for universities. There was no price competition between universities, because enough students (in a growing market) were not price sensitive. So £9,000 becomes the norm.

This isn’t a cartel. Universities are by habit compliant with law and regulation, and the injunction ‘do not discuss fees with your peers’ was, in my experience, very well observed.  But it was bad regulation. The design of the system did not, beyond pious words, prevent £9k becoming the norm.

What would Nick do?
 Equally, universities had for decades been enjoined to behave more entrepreneurially. In relation to PGT fees, universities were encouraged to see these as a price, not a cost. Would students pay the fee? If so, charge it. If the fee wasn’t enough, stop teaching the programme. And this habit infused undergraduate fee decisions. It would be possible to regulate or legislate for a fee regime that reflected cost not market price. But this wasn’t done.

So what is to be done now? I’ve a simple plan. Ask Nick Barr. He’s Professor of Public Economics at LSE, knows more about higher education funding than almost anybody else in the world, and is wise and fair-minded. Ask Nick Barr how to fund students and universities on a fair and sustainable basis. And do what he says.


Monday, 3 July 2017

Increased tuition fees do not cause increased participation

Tuition fees are back on the political agenda, big time. Arguably a significant component in the unexpected relative success of the Labour Party at the 2017 General Election, there are now calls by senior Ministers for a ‘national debate’ on the issue: see, for instance, Damian Green’s speech to the Bright Blue think tank on the weekend.

One aspect of this which is worth examining is the connection between fees and access: on the one hand there us the fear that debt will put people off university (and hinder their subsequent life-chances); on the other there is the evidence that participation by students from less advantaged backgrounds has grown since the introduction of higher tuition fees in 2012. I've seen it argued - by people including a Vice-Chancellor of a UK university - that fees have helped with this process.

But of course, correlation does not imply causality. And because higher education is a devolved matter, there’s a simple experiment which can shed light on the question about whether fees encourage participation.

Scottish domiciled students pay no tuition fees if they attend Scottish universities. If they attend universities in England they pay the normal home rate – that is, £9,000. HESA data lets us see whether more Scottish students attended universities in England after fees were increased in 2012, which they should if tuition fees encouraged greater participation.

Here’s the data. It shows the number of undergraduate students attending universities in England from each of the four UK nations: England, Scotland, Wales and Northern Ireland (NI). The data covers the years 2008-09 to 2015-16: that is, four years under the £3k tuition fee regime and four years under the £9k regime. The number of English students is several orders of magnitude higher than those from Scotland, Wales and Northern Ireland, so I’ve used two vertical axes. The left had axis show student numbers from Wales, Scotland and NI; the right hand axis shows students from England.

Data from HESA Student data, table N
The chart shows that the number of English students at English universities grew over the eight years, from about 780,000 to 925,000. The number of Welsh and Northern Irish students at English Universities also grew, from 15,000 to 22,000 in the case of Wales, and from just shy of 8,000 to just over 9,000 in the case of Northern Ireland.

The number of Scottish students at English universities did not grow. 4,840 Scottish students attended English universities in 2008-09; 4,255 attended in 2015-16.  And just to be clear, there isn’t a peculiar effect of declining number of eligible Scottish students: the number of Scottish students attending Scottish universities grew from about 84,000 to about 94,000 over the same period.

So it seems that tuition fees do not cause increased participation. The growth in English students can be explained by the availability of places: the cap on recruitment was removed in the couple of years following 2012, giving universities an incentive to recruit as many students as they wished. But if students could study in their home country for free, as in the case of Scotland, they were immune to the charms and marketing persuasions of English universities.

So when the debate plays out, be careful to spot when correlation (growth in student numbers in England) elides into claims of causation. The evidence is that tuition fees do not cause a growth in student numbers.