(Mar. 30, 2018) – According to a new study by researchers at the University of Kentucky and University of Colorado Boulder, a majority of Twitter users are unaware that their public tweets could be collected and analyzed by researchers.

Nicholas Proferes, assistant professor of information communication technology in the School of Information Science, a part of the College of Communication and Information at UK, along with lead author, Casey Fiesler, assistant professor in the Department of Information Science at University of Colorado Boulder conducted an exploratory study that surveys Twitter users’ perceptions of the use of tweets in research.

“We found that most Twitter users don’t know that this is happening, and many (incorrectly) think that researchers aren’t permitted to use tweets without user permission. However, that’s not the case,” Proferes said. “At the same time, we also found that most Twitter users would be willing to let their content be used for scientific research if they were asked. This raises a number of questions about how we, as researchers, should handle user content, and how we might go about informing users about research that uses their publicly available content.”

Proferes and Fiesler seek to provide ethical guidance and standards for online data researchers, policymakers, and consumers. Researchers agree that the availability of public social media data from social computing systems like Twitter present ethical challenges and that users can help inform ethical research practices.

The research team surveyed 268 Twitter users and found that 62 percent of users didn’t know that researchers sometimes used their tweets. When asked whether they thought researchers were permitted to freely collect and analyze their tweets, 43 percent of Twitter users said ‘no.’

Researchers also found that users’ attitudes towards the practice depended on factors, such as how the research is being conducted, who is conducting it, and what the study is about. For example, 64.9 percent of respondents said that researchers should not use tweets without consent, which suggests a strong desire to have researchers seek permission.

As a result, researchers identify potential best practices and considerations for those conducting observation and analysis of public social media data, such as: 1) asking for permission if there is a reasonable way to do so, 2) anonymizing identifying information when quoting tweets, 3) request permission to publish the user’s identity, and 4) avoid using deleted content.

Proferes and Fiesler also suggest that researchers should consider informing users about the research and providing them with opt-out options. Future directions for this research include looking beyond Twitter and at other social computing systems, such as Reddit, Tumblr, Instagram, and Facebook.

To view the full article, visit Social Media + Society online here. 

Click here for the UKNow publication.