[DEBATE] : (Fwd) Ranking SA's universities - just count stuff

[Patrick Bond]

(FinWeek does a commodification exercise on SA universities this week. 
It's a dog's breakfast combining research beancounting and finances and 
student throughput and subsidies and whatever all else can be quanted, 
plus toss in a racialised subtext. I wouldn't take it too seriously. A 
few of the articles are below, ending with a valid punchline diminishing 
the merits of this exercise: "Quantitative indicators, as well as 
qualitative judgments on issues such as leadership, management, 
governance and institutional culture were needed to develop a template 
to appraise institutions and strategies for restructuring them and to 
measure actual outputs against the expected outputs of the reconfigured 
system.")

At a glance
Feb 26 2009 00:00
 
 
Teaching and learning

UNISA (86%), UCT (86%), Rhodes (85%), Limpopo (83%), Stellenbosch (82%), 
University of KwaZulu-Natal (81%), University of Pretoria (81%), Wits 
(79%), North West University (79%) and the University of Zululand (78%).

Research

UCT (221 of a total 300 points), Stellenbosch (232), Pretoria (214), 
Wits (174), KwaZulu-Natal (142), North West (85), Unisa (81), Free State 
(70), UJ (69), and Western Cape (67).

Graduate employability

WHERE DO employees come from? Unisa (33%), UJ (30%), Wits (27%), UP 
(24%), UCT (18%), Stellenbosch (16%), KwaZulu-Natal (15%), Cape 
Peninsula University of Technology (10%), North West (9%), and Tshwane 
University of Technology (9%).

Corporate perceptions: UP (7,79 out of a total 10 points), Wits (7,74), 
Stellenbosch (7,69), Cape Peninsula University of Technology (7,6), 
University of Johannesburg (7,47), North West (7,22), UCT (7,11), Unisa 
(7,06), Tshwane University of Technology (6,44) and KwaZulu-Natal (6,33).

Financial sustainability and productivity

GOVERNMENT FUNDING: VUT (68,53%), Durban University of Technology 
(65,23%), Tshwane University of Technology (61,42%), Walter Sisulu 
University (61,14%), Central University of Technology Free State 
(56,45%), Venda (53,87%), Cape Peninsula University of Technology 
(52,56%), University of Zululand (52,12%), UJ (47,85%) and Free State 
(44,66%).

Expenditure per student: Wits (R97 000,21), Stellenbosch (R74 000,83), 
Rhodes (R74 000,73), UCT (R74 000,24), Limpopo (R55 000,49), 
KwaZulu-Natal (R52 000,66), University of Fort Hare (R50 000,13), UP 
(R47 000,38), Nelson Mandela Metropolitan University (R43 000,38) and 
Western Cape (R43 000,14).

Resources

STUDENT PER academic staff ratio: Rhodes (14,6), UCT (15,61), 
KwaZulu-Natal (18,73), Fort Hare (19,9), Stellenbosch (21,01), Wits 
(21,22), Pretoria (22,92), Limpopo (24,14), Western Cape (24,51) and 
Free State (28,18).

***

Achievements in different categories
Feb 26 2009 00:00
 

Teaching and learning

THE MOST TELLING PARADOX of a restructuring strategy intended to 
increase the teaching capacity of newly merged universities to attract 
black students and increase the quality and quantity of employable 
talent in the labour market is that it appears to be doing exactly the 
reverse. With a growing student population and surging State expenditure 
it would be tempting to think that improvements in teaching and learning 
and undergraduate success rates are most evident in merged and 
incorporated institutions. Not so. On balance, universities left 
virtually untouched by the restructuring strategy produced the best 
results.

With their depth of offerings in teaching and learning, Unisa and the 
University of Cape Town shared the top two spots, scoring 86 each. Says 
Ask Afrika's Richards: "UCT is understandable, because of its tradition 
of quality lecturers and students. In second place, Unisa is a little 
surprising. While its lectures and tutorial material are good it also 
has more responsible, older students that could be the primary reason 
its student pass rate is high.

"However, an area in which elite institutions such as UCT and 
Stellenbosch can improve is academic/non-academic ratios," she says.

Rhodes came in at a close third (85%) followed by Limpopo (83%) and in 
fifth place Stellenbosch (82%).

Research

THE QUANTUM OF RESEARCH output in many of the larger, newly integrated 
universities has grown robustly in recent years, according to National 
Research Foundation (NRF) data. But in terms of the quality of research 
as a percentage of permanent academic staff, medium-sized research 
institutions - UCT and Stellenbosch - excelled, clinching the highest 
weightings. First in the field of research is UCT, scoring 271 points 
out of 300. According to the Ask Afrika ranking: "UCT especially is 
committed to building on its international reputation by offering 
exciting opportunities for postgraduate research and learning and a wide 
range of postgraduate degrees."

Stellenbosch is second (232), followed by UP (214). The NRF measured the 
research capabilities according to research staff they regarded as 
leaders in their field: young, promising researchers identified by their 
peers as future leaders in the field, and researchers who entered the 
field at a later stage but are currently establishing themselves as 
prominent researchers. Third was UP (214), followed by Wits (174) and 
the University of KwaZulu-Natal (142).

Overall research outputs have risen significantly over the past decade, 
the NRF notes. UP has produced the largest number of accredited journal 
publications since 1997. In 2006 the university had 183 NRF rated 
scientists, four being A rated and 53 B rated. UP also registered 34 new 
patents in 2005 and 11 technologies to industry were licensed. In 2005, 
192 students received doctoral degrees, 60 being black. However, its 
problem is very high student supervisor ratios that undermine the 
quality of the postgraduate experience and overall quality of the theses 
and dissertations, the NRF says.

Graduate employability

PREDICTABLY, larger universities scored favourably among 100 human 
resources managers and corporate executives interviewed in the Ask 
Afrika survey to decide which institutions are optimal suppliers of 
employable skills in the labour market.

Unisa had the largest headcount of 239 581 students and so, naturally, 
is the source of the majority of students entering the work environment 
- other than, the University of Johannesburg, which has a headcount of 
41 740 students. Unisa produced most graduate employees last year, 
scoring 33 of 100 points. UJ scored 30 points, followed by Wits (27).

Where quality is concerned, the results differ markedly. HR managers say 
UP students were invariably equipped with qualitatively superior and 
relevant skills. Notwithstanding the volume of output "UJ is a 
university in decline in fifth place in the top10 HR managers' 
category," says Richards.

Financial sustainability and productivity

THERE'S SOMETHING decidedly contradictory about Government research 
subsidies as a percentage of total university income, which is the 
disproportionately high allocation of subsidies to universities with the 
lowest academic output. The top 10 results show smaller universities - 
principally merged and historically black institutions - received more 
grants from Government than their more productive peers.

VUT scored 68,53%; Durban Institute of Technology 65,23; and Tshwane 
University of Technology 61,42. "In this category, smaller merged and 
historically disadvantaged universities got the most funds from 
Government," says Richards. Fortunately, smaller universities were able 
to overcome the shortfall in expenditure per student. Based on 
Department of Education statistics, Wits had the highest spend per 
student (R97 000,21), followed by Stellenbosch (R74 000,83) and Rhodes 
(R74 000,73). "In some instances, bigger universities such as Wits had a 
head start and also had efficient third income streams in place. They 
set up companies to generate revenue and pump money back into the 
universities," says Richards.

Interestingly, UP - usually in the top 10 in all categories - ranked 
poorly, showing third from last.

Resources

SMALL UNIVERSITIES naturally ranked highly in the staff to student 
ratio. Rhodes has SA's most favourable staff to student ratio, with just 
more than 6 000 students and 300 academic staff, enabling it to provide 
a quality contact-learning experience to students. "Once you interpret 
the results Rhodes's size enables quality contact. It's a smaller 
university," says Richards. Its average number of students per permanent 
staff is 14,5, followed by UCT (15,61) and KwaZulu-Natal (18,73). Fort 
Hare (19,9) took fourth spot, followed by Stellenbosch (21,01).The 
University of Pretoria is the winner in terms of having more (61%) 
academic staff than non-academic staff, although it's not performing 
that well in terms of student to staff ratios. Fort Hare (57,87%), 
another small university, is a close second and Cape Peninsula 
University of Technology (54,18%) third.

Surprisingly, UCT's main weakness is its academic to non-academic ratio. 
Stellenbosch is perhaps the biggest disappointment in that category. 
Nevertheless, more alternative funding innovations continued to impress. 
While small classes and student numbers helped Rhodes secure the highest 
spot in the category larger institutions managed to make the top 10 "due 
to innovative measures of revenue generation," Richards says. 
Universities of technology - lacklustre entrants whose resource bases 
are low - were in the bottom 10.

***

A dangerous divergence
Feb 26 2009 00:00

TAKE TWO TERTIARY INSTITUTIONS, both heavily dependent on Government 
subsidies to make their performance look good. Give the smaller, 
non-performing one a disproportionately favourable share of the State 
subsidy allocated annually to institutions of higher learning. Now watch 
it embrace the funds and consume it in inordinately high salaries and 
routine administrative bungles. Hand the other over to highly successful 
but smaller academic programmes in the larger university.

Result? The larger institution - University of Cape Town, in this 
instance - continues to grow its graduate output and quality of research 
at a blistering clip while (by contrast) Mangosuthu University of 
Technology (MUT) plods along, lurching from one emergency to another.

The most telling example of that wild discrepancy in the allocation of 
funds is the divergent salaries of vice-chancellors. In November last 
year a salary review by the Mail & Guardian found that suspended 
vice-chancellor Aaron Ndlovu, of MUT (incidentally, one of the worst 
research producers in the Ask Afrika survey), topped the pile, receiving 
R3,68m in 2007 for running a campus with just 9 828 students . The 
university had expenditure of R257m - R115m of which came from the State.

The packages of heads and State allocations to revenue of four of the 
biggest researching-producing universities for 2007 - University of Cape 
Town, Stellenbosch University, Pretoria University and Wits University - 
were, by contrast, relatively low. For example, the vice-chancellor of 
UCT - the top overall performer in our survey - earned R1,55m for 
heading a financially stable campus with 21 188 students (see salary 
graphic).

Seem paradoxical? Not if the measure of financial reward is racial 
transformation. In a bizarre twist, such comparisons are usually invoked 
by Government to defend the current subsidy formula and validate its 
restructuring and transformation agenda by grousing at a legacy of 
inequality between historically disadvantaged institutions and their 
elite counterparts.

In response to the exposure of the salary discrepancies, Education 
Minister Naledi Pandor has said the funding guidelines have been 
operating in a vacuum. "In the absence of self-regulation I have no 
option but to produce a policy framework," Pandor recently stated.

But the current policy framework of staccato allocations to poor 
performers has long been a reality. In fact, since 2000 - when the 
formula was officially revised by the Education Department (see separate 
report). To be sure, the formula provided for an incentive system based 
on the quantity of research output.

As things stand, all universities receive an annual subsidy based on 
their research outputs: the number of journal publications and the 
number of masters and PhD graduates. However, the sting in the tail is 
that there's no "qualitative measure" to determine whether subsidy 
allocations to universities are commensurate with output, says 
Christopher Vaughan, deputy dean (research and post graduate affairs) in 
UCT's faculty of health sciences.

Meanwhile, as the revenues of comprehensive universities have soared, 
stellar performers such as UCT, Stellenbosch and UP - all old school 
elites left untouched by the mergers - have seen their budgets dwindle 
due to Government's attempt to finance its restructuring and 
transformation efforts. UCT experienced a dip in research outputs from 2 
848 in 2001 to 2 496 in 2003 throughout faculties. There was also 
concern about workloads and lack of incentives experienced by 
supervisors and students, according to a recent audit of UCT's performance.

Top research universities are obviously displeased about being 
"penalised for over-performing" by a Government subsidy system designed 
to help former polytechnics and historically black institutions to catch 
up on research capacity, the Ask Afrika survey found.

Leading research institutions such as Stellenbosch, UCT, Rhodes, UP and 
Free State, which exceeded their research output targets in 2005 and 
2006, have received the basic subsidy only. Unfortunately, the same 
can't be said for larger, merged institutions. There was an unspoken 
assumption among those interviewed by Finweek in the Ask Afrika survey 
that larger universities were more concerned with getting the numbers 
out of the system than quality and were willing to earn a lower-calorie 
throughput. With regard to research output, Vaughan says six 
universities (Walter Sisulu, Vaal, Venda, Limpopo, Durban and 
Mangosuthu) received more in developmental grants than in actual 
subsidies for their research output in 2005. The same applied to their 
2006 output.

"Counter-intuitively, if their performance was to improve their total 
research grant would decrease," he says. In other words, if the current 
development grant were removed all institutions with a delivery score of 
more than 100% would be major beneficiaries (UCT, Stellenbosch and 
Wits), whereas those with qualitatively inferior output would see a 
substantial reduction in their total grant allocation.

"The development grant needs to be set up to assist those that are 
performing. The weaker universities need to be assisted in other ways," 
Vaughan suggests.

Trouble is, other priorities have complicated and scuppered efforts to 
change the higher education landscape for the better. Last year a 
strategic plan published by the National Research Foundation (NRF) 
reported that by 2015 it would like to double the number of PhD 
graduates. A 10-year innovation plan has set even more ambitious 
targets, suggesting that by 2018 SA should increase the annual output of 
PhDs by a factor of five. The plan identified a burgeoning staff crisis, 
as almost half of SA's most senior academics retire in the coming 
decade, combined with growing student numbers as higher education 
expands to achieve a 20% participation rate.

That's not overstating the fact that of the 15 319 permanent academics 
at universities nearly 27% are over the age of 55 and will probably need 
to be replaced over the coming decade. Among professors and associate 
professors - the most highly qualified and productive researchers - 
almost 50% are due to retire within 10 years. Meanwhile, even more are 
required in an expanding system as participation rates increase from the 
current 16% to 20% by 2016 at least.

Judging from enrolment figures, it already has. On the eve of democracy, 
total enrolment in higher learning institutions was around 17%. 
Participation rates were highly skewed by race: approximately 9% for 
Africans, 13% for coloured, 40% for Indians and 70% for whites. 
According to the National Student Financial Aid Scheme, the statutory 
body responsible for allocating student grants, the ratio of black and 
white enrolment has since been virtually inverted. The scheme noted in a 
recent submission to Parliament that 91% of the students that benefit 
from the grants were black.

With increases in enrolment figures already soaring to the 20% mark this 
year, the pressure on universities to fast-track more unprepared black 
academics through the system is on, with all the destructive effects of 
disjointed financial allocations on the quality of output. The shock is 
just starting to hit home as university bureaucrats such as Rhodes 
Vice-Chancellor Saleem Badat call for "urgent interventions".

So far, those interventions are what education academic Jonathan Jansen 
calls a progressive "dumbing down" of the quality of output by "awarding 
professorships to graduates without any record of scholarship, without a 
track record in research and without any credibility in the competitive 
world of research journals, research conferences and research programmes".

In an earlier 2002 critique of Government's plans ('On the state of 
South African universities'), Jansen spared no punches when he wrote 
(quite stingingly) that differences between sub-par institutions and the 
hard intellectual architecture of top performers aren't primarily 
related to differences in resources and quantity of output. Rather, 
they're attributed to a "growing malpractice" in (especially technikons) 
and some universities to create a new class of (mainly) black professors 
in response to "an unseemly haste to propel these institutions, 
overnight, into university status, such as the so-called universities of 
technology - on the back of professorial appointments made, in the main, 
to young black academics".

The proof is in the numbers: when democracy was achieved in 1994 SA's 
academic cohort was overwhelmingly white (83%) and male (68%). While 
black Africans constituted 89% of the population, they made up only 17% 
of academic staff. By 2006 the number of black academics had soared: 62% 
of academic staff were white and 58% male, while 38% were black. The 
percentage of black academics has since been rising exponentially.

That, it seems, is the key issue. In the Education Department's haste to 
pummel more black students through the system as quickly as possible 
there was growing disquiet, raised in the early 2000s in public 
statements such as the National Plan for Higher Education of the 
Education Department and the National Research & Development Strategy of 
Science and Technology, about the dominance of an ageing white 
population of academic output.

Says Jansen: "For both technikons and universities the common additional 
pressure is employment equity - and the need to advance black scholars 
within those institutions... The instrument of choice has been to give 
away professorial status and salary levels to those young and ambitious 
black academics."

Hence the State's skewed allocation of funds to the worst performers 
must be viewed in the context of the system-wide effects of that 
strategy to alleviate the deleterious effects of decades of poor 
schooling. And since in the SA context universities are funded on the 
basis of student enrolment rates and research volumes it's small wonder 
elite institutions are penalised for over-performing.

Though that may be good for transformation the consequent downward 
aggregation in standards and quality is apparent from graduate 
unemployment rates in SA, which doubled in the first decade of democracy 
despite a worsening skills shortage, according to economists at UCT.

Certainly, while SA's Budget allocation to newly merged and historically 
disadvantaged universities has increased considerably on the back of a 
rising tide of black student enrolment, the academic progression rate is 
faltering. Clearly, that's more than a simple black:white numbers game: 
the incentive system, driven by Government's push for more black 
academics and students over the past decade, is one example of 
perilously skewed priorities that now threaten the productivity of 
quality-driven universities. "The effects of such practices on the 
higher education system are incalculable," says Jansen.

All such changes have made more acute a longstanding asymmetry: 
Government's bias towards development in its subsidy formula is a boon 
for mediocrity and a nightmare for the elites.

What that actually proves is the transformation of higher education is 
complex, with historical redress just one contributory chapter in a much 
bigger story. At issue is a transformation agenda that doesn't see black 
institutional and individual academic advancement as a national asset in 
which Government makes investments based on the quality of output - for 
example, by incentivising universities to produce top-flight black PhDs.

Jansen says: "The most important intervention would be to make massive 
investments in the development of a new generation of young scholars." 
Pursuing, as Vaughan has recommended, a different "demand-driven model" 
where incentives for funding are determined by the quality of output.

That seems to be sinking in as the Council on Higher Education slowly 
warms to the idea of a new framework for rewarding quality. Currently, 
most education bureaucrats in Government are asking whether it's time to 
rewrite the funding textbook.

That's good news for universities credited with quality output. Vaughan 
estimates a shift in the funding formula towards quality output would 
mean his own institution (UCT) would be one of the main beneficiaries, 
with its research subsidy increasing by at least R40m - an increase of 
almost 25% - and even greater increases with the removal of the current 
development-based grant.

But not everyone (especially the elite campuses) is applauding yet. 
Despite the council's avowal, there are signs that smaller 
non-performing (comprehensive) universities may yet come out ahead. In a 
recent announcement, Pandor said R3,6bn would be allocated to a number 
of universities that had the potential to increase their enrolments and 
graduates - especially black graduates - during the planning process for 
the three years to 2010.

If that's the case, then merged "comprehensive" universities - with the 
largest black student population - are in luck. The downside, say 
education analysts, is that increasing the allocation to smaller 
comprehensive universities may worsen the crisis if the tendency for the 
administrative affairs and balance sheets of comprehensives and 
universities of technology, such as Mangosuthu, to run right over the 
cliff is accentuated by weak pass rates and poor research output. 
They'll simply enrol more black students that, in turn, would enable 
them to raise more subsidies.

And so the cyclical dependency can potentially feed on itself.

Unlike teaching institutions, research-based universities face a 
resources dilemma: fast-track more black students through the system - 
with all the attendant risks of declining quality and under-capacity - 
or maintain the status quo and risk losing State funds needed for 
quality research.

The stakes are high. On average, says Higher Education SA spokesman 
Patrick Fish, universities in SA earn around 43% from direct public 
finding, 29% from tuition fees and 28% from "third stream" private income.

That cold truth prompts Vaughan to pose the question: "Should we be 
rewarding universities whose academics produced the greatest number of 
publications, without regard to quality, or should our emphasis be on a 
system that inspires our academics to aim for a level of scholarship 
that can withstand the scrutiny of an international audience? I believe 
it's the latter."

How the grant system works

IN TERMS OF Government's formula, each institution's success in meeting 
its research output target is known as "delivery" - the ratio of 
weighted output to "normed" output expressed as a percentage. A delivery 
of more than 100% indicates the institution has exceeded its target. The 
"shortfall" is simply the normed output less the weighted output and is 
set equal to zero when delivery exceeds 100%.

According to 2007 statistics in a recent report by Vaughan, the actual 
grant earned by an individual institution in 2007 was equal to the 
institution's weighted research output multiplied by the total research 
subsidy (R1,385bn) and divided by the normed output for all institutions.

"In 2007, one research output credit was worth R85 026. Those 
institutions with a delivery of less than 100% earned a development 
grant - sharing approximately R148m - where the amount awarded was 
linearly related to their shortfall. Each institution's total grant for 
the year was the accrual grant, based on research outputs, plus 
something called a development grant."

In simple terms, what all that means is the development grant has 
obviously been included in the funding formula to encourage institutions 
to develop a stronger research culture, says Vaughan. But here's where 
the problem gets interesting. Vaughan says several universities earn a 
greater development grant than actual grant, thus establishing a 
perverse incentive. "The lower their delivery, the greater their total 
grant. The unintended consequences of that policy certainly need to be 
evaluated," Vaughan says.

***

Research model
Feb 26 2009 00:00

 
A NUMBER OF sequential steps were taken in the development of the 
research model. The first was the construction of a baseline picture in 
which several areas of performance were estimated by Department of 
Education and National Research Foundation (NRF) data. The assessment 
model was broken into five key performance indicators based on 
weightings by university vice-chancellors themselves of typical areas of 
competence and output.

"Five universities responded regarding the weightings. We used NRF and 
Education Department independent rating systems. So universities didn't 
rate themselves. Their involvement was their willingness to participate. 
The weights allocated to each performance indicator were reduced to an 
average. In terms of human resources manager interviews universities 
weren't involved," says Ask Afrika's Amelia Richards.

Teaching and learning, which relates to data regarding undergraduate 
success based on 2007 Education Department statistics, was the first 
indicator allocated 0,33 points by V-Cs. The research category 
contributed 0,28 points of the total weighting. Third was rating 
universities in terms of graduate employability (0,15). Fourth was 
expenditure and funding, which relates to Government funding as a 
percentage of total income spent per student based on 2007 data (0,15); 
and fifth was resources, which focuses on student and staff ratios 
(0,09). A sixth factor - sport and cultural performance - was excluded 
from this survey due to insufficient data.

The next step was to distribute updated figures for the Education 
Department and NRF throughout the five indicators. "The overall winner 
was an average calculation for all five factors. Universities that won 
didn't necessarily take first place every time," says Richards.

"If the university was ranked first for a certain category (say, 
research) it would get 100 points for that factor. If it came second, 90 
points, and so on."

Considerable time was then spent distilling data from relevant secondary 
research from the universities themselves and interviewing a wide 
variety of sources, including human resources managers in the top 
JSE-listed companies and students to arrive at a more accurate 
assessment of the top and bottom performers.

"We interviewed HR managers of top JSE-listed companies in order to get 
a sense of employability of graduates from different universities and, 
lastly, interviews with students to get a sense of their perceptions of 
institutions they attended," says Richards.

However, Richards concedes the research project is limited. Quantitative 
indicators, as well as qualitative judgments on issues such as 
leadership, management, governance and institutional culture were needed 
to develop a template to appraise institutions and strategies for 
restructuring them and to measure actual outputs against the expected 
outputs of the reconfigured system.