At the start of this July (2014), Facebook hit the news worldwide, all because of research methods. If ever proof were needed that research methods are super-exciting, here it was!
A study conducted over one week in 2012 had altered the extent to which 689,000 Facebook users were exposed to emotional content in their News Feeds. The study found evidence of what it described as ’emotional contagion’: when users were more likely to see more positive content in their News Feeds, they used very slightly more positive words in their posts. When the content they saw on their News Feeds was less likely to be positive, the number of positive words was very slightly reduced.
The News Feed feature was the only part of users’ Facebook pages that was modified, and this is a feature of Facebook that is constantly modified anyway – the content from friends that is shared on News Feed is selected by a Facebook algorithm which is designed to show content that users are most likely to find interesting. All shared content was viewable as usual on users’ wall and timeline. Users’ posts (300 million of them) were monitored, but no researcher read any posts – all the processing of emotional word use was done by computers via Facebook’s own News Feed filtering system. The experiment did not alter in any way the messages that users sent to each other. And all those who were involved in the study were selected randomly, by username and pattern of use (the study made sure they were people who used Facebook regularly). So why was there such a huge reaction to news of the study?
The main problem people had with the study, it seemed, was that they hadn’t been asked if they wanted to take part. As all psychology students know, this is an ethical issue known as informed consent. Participants must be given comprehensive information concerning the nature and purpose of a study and their role in it. Participants can then make an informed decision about whether to participate in the study or not.
The difficulty for researchers is that providing comprehensive information of this kind may reduce the meaningfulness of their research because once participants know what the study is doing, it may then affect their behaviour. That would certainly seem to be an obvious problem for this Facebook-based research into emotional contagion. However, concerns over manipulation of personal information by ‘big data’ are deeply felt by large numbers of people worldwide, suggesting that the ethical issues of this study should have been much more carefully considered.