If you read papers on the philosophy of science you will very quickly come across the concept of replication, the requirement to test the same hypothesis twice or more before you become too attached to your conclusions. As a new student or a research scientist you face this problem when you wish to replicate some previous study. If you do replicate, you risk being classed as an inferior scientist with no ideas of your own. If you refuse to replicate and try something new, you will be criticized as reckless and not building a solid foundation in your science.
There is an excellent literature discussing the problem of replication in ecology in particular and science in general. Nichols et al. (2019) argue persuasively that a single experiment is not enough. Amrheim et al. (2019) approach the problem from a statistical point of view and caution that single statistical tests are a shaky platform for drawing solid conclusions. They point out that statistical tests not only test hypotheses, but also countless assumptions and particularly for ecological studies the exact plant and animal community in which the study takes place. In contrast to ecological science, medicine probably has more replication problems at the other extreme – too many replications – leading to a waste of research money and talent. (Siontis and Ioannidis 2018).
A graduate seminar could profitably focus on a list of the most critical experiments or generalizations of our time in any subdiscipline of ecology. Given such a list we could ask if the conclusions still stand as time has passed, or perhaps if climate change has upset the older predictions, or whether the observations or experiments have been replicated to test the strength of conclusions. We can develop a stronger science of ecology only if we recognize both the strengths and the limitations of our current ideas.
Baker (2016) approached this issue by asking the simple question “Is there a reproducibility crisis?” Her results are well worth visiting. She had to cast a wide net in the sciences so unfortunately there are no details specific to ecological science in this paper. A similar question in ecology would have to distinguish observational studies and experimental manipulations to narrow down a current view of this issue. An interesting example is explored in Parker (2013) who analyzed a particular hypothesis in evolutionary biology about plumage colour in a single bird species, and the array of problems of an extensive literature on sexual selection in this field is astonishing.
A critic might argue that ecology is largely a descriptive science that should not expect to develop observational or experimental conclusions that will extend very much beyond the present. If that is the case, one might argue that replication over time is important for deciding when an established principle is no longer valid. Ecological predictions based on current knowledge may have much less reliability than we would hope, but the only way to find out is to replicate. Scientific progress depends on identifying goals and determining how far we have progressed to achieving these goals (Currie 2019). To advance we need to discuss replication in ecology.
Amrhein, V., Trafinnow, D. & Greenland, S. (2019) Inferential statistics as descriptive statistics: There is no replication crisis if we don’t expect replication. American Statistician, 73, 262-270. doi: 10.1080/00031305.2018.1543137.
Baker, M. (2016) Is there a reproducibility crisis in science? Nature, 533, 452-454.
Currie, D.J. (2019) Where Newton might have taken ecology. Global Ecology and Biogeography, 28, 18-27. doi: 10.1111/geb.12842.
Nichols, J.D., Kendall, W.L. & Boomer, G.S. (2019) Accumulating evidence in ecology: Once is not enough. Ecology and Evolution, 9, 13991-14004. doi: 10.1002/ece3.5836.
Parker, T.H. (2013) What do we really know about the signalling role of plumage colour in blue tits? A case study of impediments to progress in evolutionary biology. Biological Reviews, 88, 511-536. doi: 10.1111/brv.12013.
Siontis, K.C. & Ioannidis, J.P.A. (2018) Replication, duplication, and waste in a quarter million systematic reviews and meta-analyses. Circulation: Cardiovascular quality and outcomes, 11, e005212. doi: 10.1161/CIRCOUTCOMES.118.005212.