1.3 What Exactly is Open Science?

In the previous sections we discussed the burden of proof and the erosion of social and stakeholder trust in the sciences.

This next section will explore a related issue: the ability to build up evidence for a scientific claim by repeating the original study to make sure that the finding, be it correlative or causal, is a true finding and not just a fluke.

A Replication Crisis

Open Science proposes many solutions to problems with how research is done.

Open Science aims to ensure that research can be confirmed through replication, and that both new discoveries and replications of previous research are respected and rewarded.

Although scientific progress rests on the repetition and reproduction of results, efforts at this are hindered by two main factors.

First, there is a desire for the new and shiny: novel studies are more interesting. Second, many studies cannot be competently analyzed or replicated. This is because critical information about them — design, data, methods, lab notes, analyses and code — may not be made available, or may be poorly communicated.

Flashy Science or Good Science

New, positive, and original findings are exciting. Independently repeating a study and getting the same results, not so much. The consequence for science? New, positive, and flashy findings get published and a lot of attention. Re-testing, or replicating these studies, and non-positive findings often attract much less attention.

The result? Novel research gets funded and published. Sometimes, it appears, the excitement factor overrides the quality (or lack of quality) of the science.

Incentives

Replication studies have traditionally been hard to fund, and many journals are not interested in publishing them. And universities and scientific institutes offer career rewards to those who publish novel findings in prestigious journals.

Academia and industry alike reward the publication of positive and so-called "significant" results. Scientists may unwittingly "bend the rules" of science trying to get results worthy of publication. "Bending the rules" in this context is also known as engaging in Questionable Research Practices (QRPs).

It's important to note that QRPs don’t normally result from deliberate efforts on the part of researchers to do bad science. QRPs arise from a complex system of interactions within and around the research process and from the need for researchers to get published. After all, publications are the bread and butter of a researcher. And unfortunately, this can mean that sometimes style trumps substance, as Daniel Engber notes in his Slate article Cancer Research Is Broken.

Enter Open Science

Open Science is a movement that tries to combat the replication crisis, QRPs, and style trumping substance in two ways. First, by providing different incentives and rewards for research. That is, changing what we measure as a success in research, shifting from a culture that emphasizes novel findings to one that also rewards the many other aspects of practicing good science. Second, by making all parts of the scientific research process transparent and accessible, allowing for a critical review of how a study was conducted, and ultimately enabling that study to be independently replicated.

What sort of changes, then, would we expect Open Science to bring? Imagine a world where

  • publications and data are freely available 1;
  • conclusions and public policy are based on solid information and analysis that are clearly evident to all;
  • studies can be tested and repeated;
  • researchers without extensive funding or resources can still participate and collaborate in science;
  • good science — rather than flashy science — brings rewards to those who practice it; and
  • the public trusts science and has opportunities to participate in research.

  1. There are some necessary and reasonable restrictions on data availability, such as privacy concerns, cultural and intellectual property rights, or even protection of the location of endangered species, to name a few.↩︎