Understanding demands and pressures of PhD production

There is a real danger that by focusing on national and institutional policies and strategies to expand PhD production, a truism will be lost: “It is that supervisors graduate PhD students,” says Professor Johann Mouton of the University of Stellenbosch in South Africa.

In South Africa’s efforts to triple the number of doctoral graduates from around 1,500 a year, not enough attention is being given to the role of supervisors.

“Of course, they don’t do it in a vacuum. They do it within institutions and departments and scientific fields that give them support and create the necessary conditions. But the bottom line is that irrespective of how many resources you put in institutions or what conditions you have in place, it comes down to the individual who has to supervise and graduate a student,” Mouton said.

“We are starting to understand why some supervisors are more successful in producing PhD graduates of quality, and more efficiently.”

Mouton, director of the Centre for Research on Evaluation, Science and Technology, or CREST, was speaking at a recent workshop, convened by South Africa’s National Research Foundation and the Carnegie Corporation of New York, on “Expanding and Sustaining Excellence in Doctoral Programmes in Sub-Saharan Africa: What needs to be done?”

In his presentation on understanding the demands and pressures of PhD production in South Africa, Mouton said that for five years the team at CREST and others had been trying to understand dynamics around doctoral education.

“One is the relationship between national and institutional policies and what happens within PhD production on the ground. The research has started to give us an understanding of the issues that have to be addressed when Africa attempts to expand PhD production,” he said.

“Our recent work has tried to understand the interplay between policies at different levels and the actual behaviour of supervisors and departments at the micro level. If we understand these dynamics, it might be useful for other institutions, for future actions and initiatives.”

The four imperatives

There have been four main imperatives in policies and strategies on PhD training in South Africa over the past 15 years: quantity, quality, efficiency, and transformation and equity.

Policy and strategy documents by various bodies had articulated the demand for an increase in PhD production in South Africa, said Mouton. While this began to be “symbolically” expressed in the 1997 education white paper, it became explicit in the 2001 National Plan for Higher Education. “But at that point no targets were set.”

In 2003, South Africa’s education department revised the national funding framework for universities, and research masters and PhDs as knowledge outputs were added to subsidies for research outputs. “Suddenly money came into the picture,” said Mouton.

“From 2005, universities in South Africa were rewarded quite significantly for producing more PhD graduates in all fields of science.” Producing a PhD was valued as equivalent to three article units, with the rewards ranging from US$40,000 to US$60,000 per graduate.

It was therefore not surprising, Mouton said, that in 2007 both the Department of Science and Technology and the National Research Foundation set PhD targets for the country, though these targets were “slightly unrealistic and ambitious.”

First, the target was set at 6,000 PhDs per year by 2025 – at that point, South Africa was graduating about 1,200. The most recent target, in the National Development Plan, is 5,000 PhDs a year by 2030. “The latest figures put current PhD graduations at about 1,570 a year.

“It’s very nice to set these targets from the top. My own view is that it’s going to be nearly impossible to more than triple our output in the next 15 years. We haven’t done so in the past 15 years, though we have pretty much doubled output.”

There are only about 5,500 academics at South African universities with PhDs and who can supervise, Mouton told the workshop. The assumption is that they will produce 5,000 PhDs a year. “The current average is one every four years. Only 10% of South African supervisors, the most productive, produce one PhD a year.

Policy-makers, Mouton concluded, needed to check targets against trends and statistics.

While there is no policy imperative that speaks explicitly to quality, Mouton said, it is assumed in all of the documents. “This tacit approach is perhaps because the notion of ‘quality’ is an elusive concept that cannot easily be measured through standard indicators.”

Furthermore, quality in higher education was assumed to be a function of the quality assurance of institutions. The most explicit national intervention to ensure quality in PhD production was through Higher Education Quality Committee audits undertaken between 2003 and 2008.

“These audit reports did focus on ensuring that institutional conditions and practices in support of PhD production were in place and reasonably standardised.”

Policy documents also do not speak explicitly about efficiency, which is seen as improving throughput rates and getting students to complete faster.

“There seems to be an assumption that our system is inefficient. We contest that. But it is clear why there is a demand for efficiency – if you link output (quantity) with monetary reward (subsidy), then you’re in the efficiency game,” said Mouton.

The most explicit statement of demand for greater efficiency was in a 2012 green paper and the National Development Plan, which give a target of a 75% throughput rate for higher education.

“This national demand for efficiency trickles down to the individual supervisor. Academic staff are told by their deans to get students through in the minimum time possible. Concerns about quality are not upfront.”

Transformation and equity
Naturally, the demand for transformation and equity in higher education – including in PhD education – was a major focus in the post-apartheid years, “and translated into having more black students and more female postgraduates”, Mouton said.

“We make a small distinction between the demands for transformation in terms of demographics – race and gender – and also disciplinary transformation, the shift towards science, engineering and technology fields.”

Supervision and students

Qualitative case studies, a web-based survey and numerous supervisor workshops undertaken by CREST identified a number of crucial supervisor-related determinants of PhD production: competence and experience, style, burden, models and support.

There were, Mouton found, “huge differences” in supervisor knowledge, competence and style, a growing supervision burden and differing levels of institutional support in terms of scholarships and bursaries, research facilities and equipment, and institutional policies.

There had been an increase in the numbers of students who were not well prepared for doctoral studies. “As a consequence, we witness a growing demand for ‘remedial’ training of doctoral students in research methods and theory, and even more basic scholarship skills.”

The average doctoral graduate in South Africa is 40 years or older, and 70% of PhD students are part-time, with major implications for student preparedness and commitment to studies – which affects drop-out rates. Most PhD students are female, with most working full-time and having families.

“The bulk of doctoral supervision currently happens at a distance, and the average student gets only two hours of supervision a month.”

These supervision and PhD student dynamics, said Mouton, kicked up their own challenges in efforts to expand doctoral graduation in South Africa.

The macro-micro tension

Mouton’s research has found that supervisors are “first and foremost concerned about the quality of the students they produce” – while national interests prioritise quantity and efficiency and to a lesser degree transformation. Therein lies much tension.

“There is a growing perception among many supervisors that the primary imperative is to get as many doctoral students to complete within the shortest possible time period.”

These perceptions were fuelled by new institutional policies and practices that rewarded supervisors who accepted as many PhD students as possible and graduated them within three years – even though the average doctoral student takes 4.8 years to complete.

“It is essential that universities achieve the right balance between national demands and good practice in supervision.

“Ultimately universities cannot simply slavishly and uncritically follow and implement such demands. They need to protect the academic project (and their supervisors), which is the pursuit of excellence in everything that we do.”