Researchers still need to adhere to the scientific method, regardless of how large the datasets are or how complicated the experiments become.
Publishing better science through better data journalism competition winner Erica Brockmeier
The life of today’s scientific researcher doesn’t look like it did in the 1940s. One of the papers I cited in my dissertation, published in 1941 by Dr. C.L. Turner, describes the efforts of a solo scientist manually counting bone segments in female fish fins after treatment with anabolic steroids. Turner was one of the first scientists to show that female mosquitofish exposed to androgens exhibited the type of fin growth which was normally only found in male mosquitofish.
At the time it was a novel discovery for animal physiology, but modern scientific research has become much more complex since. Scientists now rarely work on these types of solo endeavors, and the stereotype of the hermit scientist has been replaced by the reality of large international projects, multi-institute grants, and numerous co-authors which accompany each manuscript. Modern research efforts now focus on capturing large, comprehensive datasets which can more accurately mirror the complexity of the world we live in.
Molecular biology is a good example. The number of gene sequences on the National Center for Biotechnology Information (NCBI) website has been exponentially increasing since the 1980’s, from just 4,954 sequences in 1985 to nearly 190 million at the end of 2015. High-throughput gene expression analyses such as RNAseq and microarrays have also dominated this field, and many of these large datasets are now available on the NCBI Gene Expression Omnibus website at nearly two million samples and counting. Experiments conducted with computer algorithms to analyze existing large datasets are becoming fruitful for scientists, and we seemingly have all the data we need to make new discoveries sitting on a computer, with no need to go back to the lab. Science has changed in the past 70 years, but has an increased amount of data changed what it means to be a scientist?
Science by definition requires us to do both “intellectual and practical activities” which we use to test hypotheses based on existing knowledge. PhD students and early career researchers are usually heavily involved with the practical activities of science, whether it’s collecting 12 hour time points in the middle of the night or writing hundreds of lines of code. But the intellectual activity involves us knowing exactly why we do these activities in the first place.
While research looks different than it did 70 years ago, the scientific method itself hasn’t changed. Regardless of how large the datasets are or how many co-authors end up on a publication, science is still done by asking good questions and knowing how to find the answers. The increased ease of doing technically-challenging experiments, coupled with easier access to large datasets, has changed the practical activities of scientists — but scientists still need to remember to focus on the intellectual.
This means adhering to the scientific method and having a plan for answering specific hypothesis-driven questions instead of jumping down the rabbit hole of new experiments and unexplored big datasets. Successful scientists will be the ones who continue to recognize the value of good scientific questions, and who embrace the definition of science as both an intellectual and practical activity. Whether our work is done on large high-throughput datasets or manually counting bones in fish fins, the core of scientific research, and what it means to be a scientist, still remains the same.
Erica Brockmeier is a post-doc in computational toxicology at the University of Liverpool. She is also the lead writer of Science with Style, a weekly professional development blog for PhD students and early career researchers. You can follow her story of transitioning towards a career in science writing at @EKBrockmeier.
This piece was selected as one of the winning entries for the Publishing Better Science through Better Data writing competition. Publishing Better Science through Better Data is a free, full day conference focussing on how early career esearchers can best utilise and manage research data. The conference ran on October 26th at Wellcome Collection Building, London.
Suggested posts
How is the rise of data-intensive research changing what it means to be a scientist?
So you want to be a data scientist?
So you want to be a data scientist (again)?
Recent comments on this blog
African astronomy and how one student broke into the field
From Doctorate to Data Science: A very short guide
Work/life balance: New definitions