The influence of teaching on research in economics.

• Author: Becker, William E.

1. Introduction

   Can you think of an instance in which your teaching enhanced your research? We have had several such experiences and wondered if it was a common phenomenon among scholars in economics. To find out, we conducted a survey, asking several prominent (and several active, but less prominent) researchers in economics if their teaching had ever enhanced their research. We asked, in particular, if they could cite a specific study for which this was the case. The original goal of this study was to see if examples illustrating how teaching could play a positive role in research were common. We expanded our goal, however, when we were swamped with examples. The surprising range of unsolicited comments provoked by the survey convinced us that reporting the substance of these comments would make a much more interesting paper than simply reporting examples.

   The main message is that teaching plays a far more important role in enhancing research than the existing literature suggests, that this influence is recognized by a very large fraction of active researchers, and that this positive effect of teaching on research occurs through a wide variety of channels.

2. Literature Review

   There is much controversy and confusion in the general higher education literature on the issue of teaching versus research. Feldman (1987, pp. 274-5) summarizes an extensive survey nicely with the following:

    "Strong opinions about the connection between research

   (or scholarship) and teaching, based on informal observation alone, are not hard to find. Writing in School and Society in 1958, George B. Cutter was quite certain from his experiences in academe that 'the more research a professor has done, the more books and articles

   he has written, the better teacher he is supposed to be. But the

   opposite is more likely to be the case.' (p. 372). In direct response to this contention, Lewis Leary (1959), with equal certainty, proclaimed: 'The popular image of scholar as pedant immersed in library or laboratory has about the same validity as the popular image of Mr. Chips, Miss Dove, or Mark Hopkins and his log ... The fact is that our best teachers are almost without exception our best scholars.... Scholarship is not at a different pole from teaching' (p. 362). Extant research supports neither position."

   This controversy has continued into more recent higher education literature. Barnett (1992), for example, has the provocative title "Teaching and Research Are Inescapably Incompatible"; McCaughey (1992) rebuts, defending the view that teaching and research can coexist. McCaughey (1993) provides a literature survey and presents empirical evidence indicating that teaching and research are complements--relatively productive researchers tend to be relatively good teachers. This result notwithstanding, the consensus in the literature is that there is little or no connection between teaching and research, as stated in the last sentence of the Feldman quote above, and as articulated by Centra (1983, p. 387): "The belief that teaching and research performance are related is undoubtedly stronger than this or any other study has shown." Hattie and Marsh (1996, p. 529) provide an excellent survey of the theoretical and empirical work on this issue, concluding that "the common belief that research and teaching are inextricably entwined is an enduring myth." They also suggest reasons for why this myth endures.

   One peculiar feature of this literature, from our point of view, is that it is couched entirely in terms of research enhancing teaching, ignoring any possible causality in the other direction. In only a few studies is the possibility of teaching enhancing research even mentioned, and then it is restricted to considering a possible correlation between excellence in teaching and excellence in research. Our study differs markedly from this literature in two substantive ways. First, we are not interested in the question of whether excellence in teaching is "correlated" with excellence in research--we have no way of knowing about the quality of our survey respondents' teaching. Our interest is focused on the role of teaching in enhancing research, regardless of whether that teaching is "excellent." And second, as our title suggests, we look explicitly for causation in the direction typically ignored in the general higher education literature.

   Discussion of whether research and teaching are substitutes or complements is harder to find in the economics literature. Thorstein Veblen (1918/1957, p. 80) provides an interesting beginning by opining that teaching has importance "only in so far as it is incidental to an aggressive campaign of inquiry." By doing any undergraduate training at all, Veblen asserted that the university was in danger of becoming a body of secondary-school teachers masquerading as something much grander. Frank Knight (1947/1982, p. 386) added: "I fear that the urge to teach is largely a conceit and is a species of the lust for power and prominence, hence one of the immoral ingredients of human nature, a form of original sin."

   Becker (1975, 1979) pioneered economists' theoretical investigations of the issue of whether research and teaching are complements or substitutes. Wood and DeLorme (1976) appear to be the first authors in economics to investigate the correlation between teaching ratings and research output. Their finding of a positive correlation was not consistent with the "no apparent relationship" conclusion of the bulk of the comparable research reported in the education literature. Allen (1980) and Needham (1982) provide reviews of this early evidence.

   Although there are exceptions, as with the education literature, throughout this economics literature the emphasis is on the influence of research on teaching, as opposed to teaching on research. Paul and Rubin (1984), for example, discuss how research can improve teaching, but they are silent on the issue of whether teaching can improve research. A dramatic illustration of this is the paper by Noser, Manakyan, and Tanner (1996). They report the results of a national survey of a thousand economics professors who were asked (among other things) if their research enhanced their teaching, but they were not asked if their teaching enhanced their research.

   Although there is little analysis of teaching enhancing research in the traditional literature, economists' views on how teaching affects research do occasionally appear in biographical essays and like literature. For example, Martin Bronfenbrenner (1971), in the preface of his authoritative book on income distribution, acknowledges the importance of teaching as an input to research. More recently, Roger Backhouse (1996, p. 8) wrote that "teaching and the writing of textbooks are an integral part of the process whereby economic knowledge is created." A more explicit view is expressed by Richard Schmalensee (1998, p. 246): "I have also gotten good topics from students' questions to which 'I don't know' is a correct but unsatisfying answer. Mostly, though, I have found teaching valuable because it regularly requires me to organize, distill, and evaluate a body of literature. A number of my papers address problems uncovered while preparing lectures." But perhaps the most cogent statement in the economics literature comes from Gregory Mankiw (1998, pp. 182-3): "A less obvious benefit of classroom teaching and textbook writing is that they stimulate ideas for research. Whenever you have to explain something to someone, either in person or on a printed page, you have to think it through more thoroughly than you otherwise would. Preparing a lecture or drafting a textbook chapter reveals holes in your understanding. And, sometimes, as you try to fill these holes, you get ideas for research. Put simply, imparting knowledge and creating knowledge are complementary activities. That is why these two forms of production take place in the same firms, called universities."

   These statements from respected economists suggest that at least some think teaching enhances research; yet, we were not able to find many such written statements. The book from which the Schmalensee and Mankiw quotes were taken, for example (Szenberg 1998), contains 20 autobiographies of prominent economists whom Szenberg had asked to produce an essay on the theme "How I work." Remarkably, only four contributors mentioned how their teaching responsibilities affected their working life or their research. Comparable books yielded similar results. In Szenberg (1992) only 2 of 22 economists referred to teaching. In Heertje (1993) only one of six referred to teaching. In Breit and Spencer (1995) only 2 of 13 referred to teaching. Perhaps the mandates given these economists by these editors steered them away from teaching. (Our evidence suggests this may have been the case; there is some overlap between our respondents and these authors.) Or perhaps the Schmalensee and Mankiw statements are anomalies; the purpose of our survey was to see how widespread their experience is.

3. The Survey

   Our survey took the form of a personal letter (1) sent by one or the other of us to a wide range of economists (approximately 150) whom we knew to be productive researchers. In some cases one or both of us knew the individual personally, and in other cases they were selected because they held some specific status such as Nobel laureate, Clark Medal recipient, or past president of the American Economic Association. In this respect it was deliberately neither a random nor a representative sample of all academic economists. Similar to many such surveys, the response rate (65 of 150) was less than desired. Under this circumstance a natural worry is that those responding were those who had something positive to report, causing our results to be unrepresentative. We took steps to protect ourselves from this sample selection problem by forcing a 100% response rate from a specific subset...