The ubiquity of social media is such that, when Facebook does a study that angers its users, many begin voicing their outrage… on Facebook.
The recent study, which manipulated users’ timelines to gauge whether it would affect their own posts, showed both the remarkable research potential inherent in social media (their study involved a stunning 689,000 users), as well as the tremendous backlash that can result when people believe they’ve been involved in a study that they didn’t explicitly sign up for.
While much of the ire has been directed at the social media giant, the scientific world has expressed some curiosity regarding the Institutional Review Board at Cornell University, where researchers performed some data analysis for Facebook’s study.
In late June, Cornell issued a statement, which said that the university’s IRB did not review the study because its researchers were not the ones collecting the information, nor did they have access to any individual, identifiable data.
But regardless of whether Cornell should share any blame — and some believe they should not — the publicity is likely to change how IRBs operate.
“Even though what was done at Cornell may not have been technically incorrect, IRBs, including ourselves, tend to be conservative,” said Kelly O’Donoghue, director of UR’s Office of Human Subject Protection. “So it’s going to change practice. Nobody wants to be on the front page of the newspaper.”
“Social media is evolving so quickly,” said Tiffany Gommel, director of UR’s Research Subject Review Board. “Each institution has to take it as it comes and figure out how they want to handle it, because there’s no guidance from the government right now. So it’s definitely been on a case by case basis.”
When it comes to IRBs, social media policies aren’t the only things that researchers can struggle with. In a recent chat with the CTSI Stories Blog, O’Donoghue and Gommel shared the three most common problems they encounter when reviewing research study submissions.
“Be consistent across all documents,” said Gommel. “Protocol, consent forms, recruitment materials, study measures, etc. That’s definitely one of the biggest issues we see is that people don’t pay close enough attention. Maybe they’ll have a protocol from earlier that was similar, and they’ll cut and paste it, but they won’t edit it, or update things on every page. This leads to errors and we have to send it back to the study team requesting changes.”
“One problem we run into is that researchers have been working in their field for a long time, and they know what they’re going to do, and what order they’re going to do it in,” said O’Donoghue. “But when they translate it to paper, sometimes they leave big holes in the study. They’re inside it and they know it so well that they assume other people will, too. But they are the experts, and we’re not. So using our protocol templates can really help with making sure your submission is comprehensive.”
- Inadequate Amendments
“Researchers are required by federal regulations to submit all changes to their study to the RSRB before implementing them,” said O’Donoghue. “But sometimes researchers don’t give enough information within their amendment form, or enough rationale for the change, or they won’t make the requested changes in their application. This will slow down the review process for amendments and require the RSRB to send it back to the study team.”