The uncertain position, status and impact of negative results in marine ecology: philosophical and practical considerations

MEPS Theme Sections (previously referred to as Comment Sections) represent integrated expert analyses highlighting an important, cutting-edge topic. Theme Sections (TS) are organized by a Contributing Editor (e.g. Browman 1995, 1996). The present TS addresses a topic which has been discussed, condemned or defended by authors and editors for decades: negative results of research efforts. A rejected manuscript was my motivation for organizing this section. There was nothing wrong with the manuscript itself; the hypothesis was clear and concise and the experiment to test it was appropriate. However, both the editor and the reviewers objected to the fact that the result reported was negative. It was the first time in my career that I had prepared a manuscript about a non-result. The potential value of data which do not support a research hypothesis is often not appreciated. Nevertheless, ‘negative’ results may be very important for several reasons: they may provide more balance for a subject area thus far supported only (or primarily) by positive results (e.g. the impacts of solar ultraviolet B radiation). They may indicate that a subject area is not as mature or clearly defined as previously suspected (e.g. the first reports of reverse diel vertical migration). They may show that a particular line of research is not worth further efforts (e.g. trophodynamic modelling), or that our current methodologies are inadequate for producing a definitive result (e.g. predicting recruitment in groundfish populations). The concept of negative results is rather fuzzy. In order to provide broader coverage of its many possible meanings, I sought contributions from long-standing editors of marine science journals, senior scientists/ educators, and historians/philosophers of science. Only the latter category responded with enthusiasm. It proved difficult to recruit marine ecologists (fortunately, A. J. Underwood accepted the challenge). Further, it was impossible to get any editor onto my hook (believe me, I tried). Thus, this TS itself produced a largely negative result. Hence, the coverage is not as comprehensive as I would have liked. Declining my solicitation to contribute, Stephen Jay Gould (Professor of Geology, Museum of Comparative Zoology, The Agassiz Museum, Harvard University) wrote: ‘Your suggestion for a forum on publishing negative results in science represents a most important project. In my opinion, this is perhaps the most important effectively undiscussed subject in the entire methodology of science.’ In fact, Professor Gould had already written an essay on this topic (Gould 1993) — the main reason for declining my request. In this essay he states: ‘The importance of negative results — nature’s apparent silence or nonacquiescence to our expectations — is also a major concern in science. Of course, scientists acknowledge the vitality of a negative outcome and often try to generate such a result actively — as in trying to disprove a colleague’s favored hypothesis. But the prevalence of negative results does pose an enormous, and largely unaddressed, problem in the reporting of scientific information.’ Professor Gould asserts that positive results tell more interesting stories than negative results and are, therefore, easier to write about and more interesting to read (a privileging of the positive). He contends that this may lead to a bias which acts against the propagation of negative results in the scholarly literature. This has been borne out in recent surveys of the medical literature which discuss ‘publication bias’: studies showing positive results from drugs are published faster and more often than studies showing neutral or negative results, producing a bias that shows drugs in a favourable light (e.g. Johansen & Gotzsch 1999, Rennie 1999). These articles, and Gould’s essay, deal with several aspects of the negative results issue not addressed in this TS, and so I encourage readers to look them up. The issue of negative results remains complex. It reflects our training, our thoughtfulness about what we do as scientists (and how we do it), and our humanity, with all its inherent biases. Hopefully, the essays that follow will provide MEPS readers with a more concrete introduction to the issue.