For its annual question for 2014, Edge has asked what scientific idea should be retired. While the responses run the gamut — contributors say that concepts from 'universe' to 'inclusive fitness' to 'IQ' and 'statistical significance' should be phased out — a handful of responses focus on big data or big science.
Gary Marcus writes that while many fields would benefit from the collection of large datasets, "we should stop pretending that Big Data is magic." Researchers, he argues, place to much stock in what big data can deliver. Big data is good at identifying correlations, he notes, but not at uncovering laws. "All the big data in the world by itself won't tell you whether smoking causes lung cancer," Marcus says. "To really understand the relation between smoking and cancer, you need to run experiments, and develop mechanistic understandings."
Similarly, Samuel Arbeson notes that science is getting bigger in another way as more and more researchers work together. Still, he argues, small-scale science of a researcher or two can flourish. The idea that science is now big science is what he says should be retired. "Creative experiments and the right questions are just as important as ample funding and infrastructure, and technology is making this work easier than ever," Arbeson says. "Little science can still prosper."
Melanie Swan, by contrast, argues that big data is changing how science is conducted and that it is forcing the retirement of the idea that there is only one way — through the scientific method — to obtain scientific results. Instead, she says a number of models will supplement the traditional approach. "For the first time longitudinal baseline norms, variance, patterns, and cyclical behavior can be obtained," Swan writes. "This requires thinking beyond the simple causality of the traditional scientific method into extended systemic models of correlation, association, and episode triggering."