The teaching-only stream

In the end, it all comes down to what students are actually learning.

A debate is under way in Ontario and across Canada, on the pages of this magazine and elsewhere, over the future of postsecondary education. More specifically, how can governments best meet the need to produce well-educated, global citizens in an age of significant budgetary constraint? Or, better, what methods, means and strategies will enable students to learn more yet, at the same time, cost the system less in the years to come?

Three highly regarded figures in the Canadian higher education community, Ian Clark, David Trick and Richard Van Loon, have answered this question by calling for a dramatic increase in the number of tenured university professors whose primary, if not exclusive, duty is to teach undergraduates. More emphasis on teaching, it follows, will result in greater student learning.

Writing about the Ontario postsecondary system in particular, they claim that the “research university model” is an unsustainably high-cost construct and that it is poorly designed to meet the needs of the next generation of students. Their book, Academic Reform (McGill-Queen’s University Press, 2011), maintains that by releasing a significant number of professors from their research obligations, governments could not only refocus on students but also make the postsecondary system cost-effective.

Clearly, not everyone agrees. To date, most prominent among the critics’ concerns has been the willingness of the authors to separate a university’s research mission from its teaching function.

“Universities teach in a scholarly environment that is informed, stimulated, and enriched by research,” wrote Carleton University’s provost and vice-president academic, Peter Ricketts, in an opinion piece in this magazine. In the Canadian Historical Association’s newsletter, The Bulletin, Professor Craig Heron said: “The symbiosis between teaching and research is essential and would be seriously threatened by a teaching stream.”

Such assertions are so common – one 1992 study suggested that 90 percent of professors believed them – that they have largely evaded the critical scrutiny that scholars insist on in their own disciplines. In other words, a community that typically rejects unsubstantiated assertions has, in this case, seemingly grown comfortable making two of its own: first, that teaching and research cannot be disengaged without compromising the integrity of the university system; and second, by implication, that students will learn more from faculty who are active researchers.

The data available should make most analysts reconsider assertion number one, but it is the lack of evidence about assertion two that in the end is most troubling.

Herbert Marsh, a professor of educational psychology now at the University of Oxford, and John Hattie, a professor of education now at the University of Melbourne, analyzed close to 60 studies on the relationship between teaching and research at the postsecondary level in a 1996 article for the Review of Educational Research. They write: “Good researchers are neither more nor less likely to be effective teachers than are poor researchers.” Or, as they put it even more bluntly in the Journal of Higher Education in 2002, “it is just not defensible to claim that only good researchers are the most effective teachers, or that good teaching follows from more research.”

Their work is fascinating. But since it is focused on teaching, rather than student learning, it might not tell the whole story, and that’s why proponents of teaching-centred universities should wait before calling for firmer divisions between the research and teaching mandates of Canadian postsecondary institutions.

Professors Marsh and Hattie’s meta-analysis identifies a series of hypotheses that have sought to test the relationship between research and teaching effectiveness at the university level. The authors conclude that while each hypothesis has its proponents in the literature, only one can be sustained after being subjected to scholarly rigour.

Investigators have not been able to prove that the more time professors dedicate to research, the less effectively they teach, or vice versa. Nor have they been able to demonstrate that the qualities that make a good researcher mitigate good teaching, or vice versa. Nor does the evidence suggest that being a successful researcher or teacher requires trade-offs that inevitably compromise effectiveness in the other area.

So, good research doesn’t seem to have to come at the price of good teaching, nor does effective teaching necessarily detract from the production of quality publications.

What about the opposite? Claims that effective teachers must have active research agendas have also been rejected in scientific studies. Nor is there any convincing evidence to substantiate the claim that the personal and professional attributes that contribute to research effectiveness and teaching effectiveness are shared. (There is a similar lack of evidence to support the theory that researchers and teachers are necessarily different types of people.)

Today, then, there aren’t any convincing studies to substantiate the long-held professional belief that research and teaching at the postsecondary level cannot, or should not, be separated.

Indeed, to date, only one hypothesis, what Professors Marsh and Hattie call “the different enterprises model,” seems to have been validated scientifically. It contends that academic research and university-level teaching are so distinct that one cannot expect to find correlations in a professor’s performance in the two domains.

Taken at face value, then, one of the fundamental premises underlying the call for teaching-centric universities – that it is possible for professors to teach effectively without engaging in original research in their areas of expertise – might indeed be credible. But there are methodological and strategic flaws with such a conclusion.

Historically, examinations of the relationship between good research and good teaching have involved subjects who were expertly trained in one domain – research – and often not at all trained in the other. Moreover, even if studies had tried to focus on professors who had equivalent expertise in both domains, they would have been hard-pressed to find enough of them to sample given the paucity of pedagogical training universities have typically required of their faculty.

The closest body of evidence available that considers more equitably the links between research and teaching can be found in studies of the importance of subject matter expertise at the primary and secondary school levels, where virtually all teachers have undergone formal pedagogical training.

Deborah Ball’s work in mathematics, for example, demonstrates convincingly that teachers who have greater mathematical knowledge (a specific, deep understanding of math that Dr. Ball, a professor of education and chair of the Michigan Council for Educator Effectiveness, describes in detail) create more productive learning environments for their students.

Because an increasing number of postgraduate programs have begun to provide students with pedagogical training and teaching experience, there is also a slowly emerging literature exploring the inverse relationship. A recent article in Science magazine noted that graduate students in science, technology, engineering and mathematics who taught in lab settings during their coursework emerged at the end of the year with significantly stronger essential research skills than those who did not teach.

So, there is reason to speculate that the links between research and teaching at the university level might be closer than Hattie and Marsh’s work has suggested. Nonetheless, even if these studies are replicated with faculty as subjects and result in similar findings, by focusing so intently on teaching, an input, and not student learning, the desired strategic outcome, policymakers would be missing the point.

Marsh and Hattie’s primary tool to evaluate effective teaching is course evaluations. And while such surveys are helpful, they are insufficient if the goal is to determine what students have actually learned.

As Susan Ambrose and her colleagues have explained in the excellent and accessible How Learning Works, learning – or what some now call deep learning – is a process, not a product. It involves a significant change in the learner’s knowledge, beliefs, behaviour or attitude. Some of these changes are difficult to appreciate immediately, and likely they are impossible to observe in students in the weeks preceding the end of a course (when course evaluations typically take place).

To identify the relationship between faculty members with active research agendas and student learning in their courses, researchers might want to consider students’ academic performance well after the fact: it might be worth asking whether the ideas and concepts they were supposed to have learned affect how students solved problems in subsequent courses in subsequent years. Researchers might want to conduct effective pre- and post-testing on fundamental concepts and ideas. They might want to use other measures that have yet to be fully developed.

Right now, policymakers know so little about how much students are really learning in university classrooms that it seems too soon to render a verdict on the ultimate merits of teaching-stream positions or institutions.

So, let’s still have that debate about how to support and educate society’s future leaders, but let’s first gather the data to make it worthwhile. Discovering more about what and how students are actually learning is in everyone’s best interest.

Adam Chapnick is deputy director of education at the Canadian Forces College and associate professor of defence studies at the Royal Military College of Canada.