It wasn't so long ago that "data sharing" meant talking over your findings with a trusted colleague. Now it's shorthand for describing the future of scientific research.
Scientists, funders, publishers and institutions are taking steps to make science more transparent. Sharing research data is a big part of that process, along with steps such as reporting study methodology in greater detail and publishing results in publicly available open-access journals. "There is a culture shift happening," says Sangeeta Panicker, PhD, director of research ethics at APA.
In June, the APA's Data Sharing Working Group, convened in 2014 by the Board of Scientific Affairs, released a report that affirmed the benefits of sharing data, while detailing the practical and ethical considerations involved in doing so.
"In general, data sharing is a great idea, but it's not a blanket issue we can just approve," says working group member Michael Ross, MD, PhD, MPH, a professor of sexual health education in the department of family medicine and community health at the University of Minnesota. "For many of us, data include clinical records, interview transcripts or GIS [geographic information systems] data" that can be sensitive or even dangerous if they fall into the wrong hands, he says.
Many reasons to share
The last few years have seen a parade of high-profile reports of scientific fraud, including the 2011 scandal surrounding Diederik Stapel, PhD, a Dutch psychologist who was found to have manipulated data and fabricated whole experiments. This year, another scandal made headlines when the journal Science retracted a widely reported 2014 political science study. That paper, which reported that a conversation with gay canvassers could change voters' minds on same-sex marriage, was retracted following allegations that lead author Michael J. LaCour, a graduate student at the University of California, Los Angeles, faked the data.
Making public raw data could help prevent such scientific misconduct, but that's not the only driving force behind data sharing.
As the working group noted in its report, sharing data encourages accountability. Accessible data make it easier for researchers to replicate experiments, allowing them to better identify robust trends and connections. That, in turn, helps psychologists and other health-care providers make more informed, evidence-based decisions. Data sharing also makes older data available to new and more powerful analyses, and allows researchers to build on past work to maximize the impact of every data point.
"It's more bang for the buck," says working group member Janine Jennings, PhD, a cognitive psychologist at Wake Forest University.
That's likely one reason that major funding agencies are helping to make data sharing the rule rather than the exception. In 2013, the White House Office of Science and Technology Policy (OSTP) announced a policy requiring federal research funding agencies to develop plans to increase public access to research results. Among the directives in that policy was a charge for agencies to "maximize access" to data.
At the time, the National Institutes of Health (NIH) already had a public access policy that met many of the new criteria spelled out in the OSTP public access policy. In 2015, the National Science Foundation, Department of Defense and the Department of Health and Human Services released their own plans for sharing publications and data from studies that received federal funding.
Publishers, too, are marching toward transparency. Scientific journals across fields are increasingly moving toward open-access models, including the newest APA journal, the Archives of Scientific Psychology, which launched in 2013.
Yet transparency isn't a new idea. APA Publisher Gary VandenBos, PhD, says he's been involved in initiatives to expand data sharing since 1990, following a discussion at an NIH conference. At the time, he says, practical and technical challenges made sharing data difficult. "Improvements in computer technology over the last seven or eight years have made it much easier to be able to clean, anonymize and share data," he says.
Recently, VandenBos participated in a collaborative effort among a group of publishers, academic leaders and funding agency representatives, organized by the Center for Open Science in Charlottesville, Virginia, to develop a document known as the Transparency and Openness Promotion (TOP) guidelines. Published in Science in June, the guidelines were designed to help scientific journals across all disciplines promote transparency, openness and reproducibility.
Already, more than 500 journals are signatories of the guidelines, says Brian Nosek, PhD, a psychologist at the University of Virginia and co-founder of the Center for Open Science. While APA has not yet decided whether to officially become a signatory of the guidelines, VandenBos says, the association's Publications and Communications Board is using the TOP guidelines as a tool in ongoing discussions of transparency.
"The transparency movement is moving fast," Nosek says. "This isn't just within a particular discipline, or a particular nation. The momentum is upon us."
Sticking points
Despite that momentum, there are challenges in putting data on public display — or even in making them available to a limited group of scientists.
One of the first problems will be to adequately define "data," says John Capitanio, PhD, a psychologist at the University of California, Davis, and another member of APA's Data Sharing Working Group. If you score behaviors from a videotaped subject, for example, are the numbers the data? Or is it the video itself? "There are logistical issues, stemming from fundamental questions about what, exactly, are the data," he says.
Then there's the issue of data that are sensitive in nature. "When we talk about data sharing, we tend to use models that are fairly unproblematic," Ross says. Genetic sequences or coordinates of distant planets are relatively straightforward. But "social science data, and in particular psychology data, can have repercussions on individuals and on subject groups," he says.
Ross knows about delicate data firsthand. He studies men who have sex with men in Africa. "These data can be used by governments and organizations to argue against human rights," he says. In one African country, Ross says, a government official recently tried to access his data to help make the case for a bill that would legalize the death penalty for gay men in certain circumstances. "We have to realize that access to these data might actually be quite dangerous for the communities who have given us their trust," he says.
And it's not just a problem in foreign countries. In the United States, a number of Native American communities explicitly say they retain ownership of research data because of historic misuses of such data, Ross says. "It's not as if the question of ownership of data is a simple one."
But even studies that seem fairly benign on the surface can pose ethical problems for data sharing, Jennings says. Sometimes just a few data points are enough to unravel anonymity. A study of underage drinking among college athletes, for instance, might note a student's race, age and sport. In a small college, that could be all it takes to pinpoint which student was engaged in the illegal activity. "Demographic information might make it easy to identify some subjects," she says.
There's also the issue of informed consent. Participants in research studies have signed consent forms describing how the data will be used. But if those data are shared, they may be used for any number of secondary analyses in years to come — analyses to which the subjects have not consented. The scientific community, together with the public, will need to reconsider and redefine the concept of "informed consent," says Panicker.
Practical matters
Data sharing also presents practical problems. Will journal article methods sections need to become more detailed and explicit? Will authors need to learn new procedures for annotating data so that anyone can understand them? How will the data be delivered to journals, or entered into repositories?
"The learning curve could be fairly steep," Capitanio says.
Nosek believes that integrating data sharing into a researcher's daily workflow will be one of the biggest challenges to overcome before data sharing is widespread. "When you get down to the brass tacks, a lot of these initiatives require researchers to do something new," he says. Hoping to make that process less painful, he and his colleagues at the Center for Open Science created Open Science Framework, software that helps researchers store and share their data and other study materials in a systematic way.
But to fully face these problems of process, VandenBos says, the logistics of data sharing will have to be taught to researchers in training. "It's during graduate school that many of the values and standard operating procedures are put in place. They are the foundation that guides a researcher for the rest of his or her career," he says.
And then there's the question of access. How long should primary investigators have exclusive access to data before they are made public? If researchers don't have adequate time to make full use of their data before others can share them, it may be a disincentive to doing the research at all.
"There's been a suggestion that the data should become public after one year," Ross says. "For academics who also have to teach and have other responsibilities, it would make life impossible for them to get the data analyzed and published within a year. It may lead to a new type of researcher who simply cherry picks other people's data to publish."
Still, he believes that with care, the benefits of transparency can outweigh the growing pains. "It's good science," says Ross.
Jennings agrees, and also hopes that the research community will come together to wrestle with the ethical concerns thoughtfully and deliberately. "Let's step back and think about how to do this well, rather than cleaning up the mistakes after we've made them," she says.
Finding the sweet spot would make science more transparent without destroying innovation or creating huge bureaucratic hurdles, Nosek adds. "The real responsibility now is to make sure any changes do not destroy all the good things that are happening in science, but complement them."
Data sharing also presents practical problems. Will journal article methods sections need to become more detailed and explicit? Will authors need to learn new procedures for annotating data so that anyone can understand them? How will the data be delivered to journals?