Researchers at McGill University in Montreal and Carnegie Mellon University in Pittsburgh have suggested that studies claiming social media data can predict online and offline human behaviour may be flawed.
Over the years, studies conducted have found that social networks can be used to predict everything from the performance of summer blockbusters to fluctuations in the stock market.
But in a new article published in the journal Science, Derek Ruths, an assistant professor of computer science at McGill University, and Jürgen Pfeffer of Carnegie Mellon’s Institute for Software Research, have pointed to a number of flaws in using social media data sets.
For example, researchers say that because different social media platforms attract different demographics, they do not accurately reflect the feelings of an entire population.
In addition, the design of a social network can influence user behaviour. Researchers pointed to the lack of a ‘dislike’ button on Facebook as making it difficult to gauge negative responses. They also highlighted the problem of spammers and bots that get mistakenly incorporated into the data.
“The common thread in all these issues is the need for researchers to be more acutely aware of what they’re actually analysing when working with social media data,” said Ruths in a statement.
Ruths has pointed to the infamous ‘Dewey Defeats Truman’ headline of 1948 – when telephone polls that under-sampled supporters of US presidential candidate Harry Truman led to The Chicago Tribune’s famously incorrect banner headline – as an example of how surveying techniques have been changed to improve accuracy before.
“Rather than permanently discrediting the practice of polling, that glaring error led to today’s more sophisticated techniques, higher standards, and more accurate polls,” said Ruths.
“Now, we’re poised at a similar technological inflection point. By tackling the issues we face, we’ll be able to realise the tremendous potential for good promised by social media-based research.”
Social media image via Shutterstock
