Data storage policy can't be enforced
Survey shows 70 percent of researchers are unaware they're required to archive data collected from SSHRC-funded research
by Moira MacDonald
|
| Carol Perry at U of Guelph says Canada doesn’t have a well-defined mechanism to archive research data. Photo: Krys Mooney |
Social Sciences and Humanities Research Council grant recipients have been required for the last 17 years to permanently archive their research data, but a recent study found most recipients it surveyed were unaware of the requirement.
The study, by University of Western Ontario graduate student Carol Perry, found that more than 70 percent of respondents were unaware of SSHRC's "Research Data Archiving Policy." Some of those who did know about the policy were concerned that it contradicted their own university's requirement to destroy data after a fixed period of time.
SSHRC's policy requires that "all research data collected with the use of SSHRC funds must be preserved and made available for use by others within a reasonable period of time." The policy, in existence since 1990, lists 11 sites across the country where researchers can archive their data if their own university cannot.
Chad Gaffield, SSHRC's president, told University Affairs that while all recipients should be aware of SSHRC's archiving policy, it was not yet enforceable. "It is impossible to try to police this in an aggressive way," Dr. Gaffield said. "We know the structures are not in place."
Through her studies and work in information services at the University of Guelph, "I pretty much knew that people weren't archiving their data," said Ms. Perry, a candidate for a master's degree in information and library sciences. "In Canada there [isn't] a well-defined mechanism in place [to archive research data], and there's no Canada-wide policy, although most institutions are discussing it and recognizing the need to go toward setting up a data repository."
Recently, the main recommendation of a report into a case of academic fraud at Memorial University called for universities to set up formal repositories for raw data that all researchers would be obliged to use, as a way to prevent scientific fraud. In May, the Fonds de la recherche en santé du Québec, a health research granting agency, recommended changes to Quebec laws governing data banks and biological data banks, to take into account modern-day realities of how information is shared while still protecting the rights of research subjects.
Ms. Perry's study was small, with 75 researchers from all parts of the country completing a mail-in survey. Her study is planned for publication in a forthcoming issue of Government Information Quarterly.
Ms. Perry found that only 41 percent of respondents had plans to permanently archive their material, using methods that ranged from storing their research in a data repository to using boxes in their offices. Some 54 percent said they had no way to permanently archive their data.
Concerns about ethics and confidentiality were the biggest obstacles against respondents permanently archiving their data. Some respondents said that SSHRC's policy was in direct conflict with their own university's ethics policy requiring destruction of data after a certain period of time, again to protect personal information.
Nonetheless, Ms. Perry's results appear to be a slight improvement on those in a 2001 study done by SSHRC. It found that 80 percent of SSHRC-funded researchers who had created data sets during their projects had not archived them, and just 10 data sets had been deposited at any of the SSHRC-listed sites in 11 years.
Most researchers agree that there is value in making their research data - and not just the end result - available to others. SSHRC's policy came out of the recognition that data collected through public funds should be publicly accessible. But the stumbling block has been a lack of common standards about how the information should be compiled and archived and a shortage of facilities capable of permanently archiving digitized material.
The issue of archiving standards is an international debate, said Dr. Gaffield, and one that will need to continue to be worked out. "There's a lot of work that's being done on this score," said Dr. Gaffield. "We're making progress but we're not there yet."
He said he expected it would take 10 years to reach full compliance, adding that SSHRC is "50 to 60 percent of the way there." Discussions are also underway around the creation of a national data archive, to allow a centralized access point for research information, with national standards.
Chuck Humphrey, head of the data library at the University of Alberta, one of the 11 listed with the SSHRC policy, said such a national archive is needed to preserve data that could help answer research questions of the future. When people discard research data, they're treating it as worthless "research refuse," said Mr. Humphrey, instead of as a potentially rich source of answers to future inquiries.
