Twitter is an unreliable representation of real life emotions
Using Twitter for scientific research might seem like a golden opportunity. In fact, Dr. Eric Allen Jensen from the University of Warwick’s Department of Sociology has called Twitter a “big data gold rush,” given the over 300 million monthly active users on the website.
Jensen’s paper, published in the journal PlOS One, however, warns that any data taken from Twitter and used for scientific research should be done so with a grain of salt.
Jensen cautions against using Tweets for scientific data because there is no guarantee that content on social media is an accurate representation of the people posting it.
People on social media have ample opportunity for living “double lives,” often showing the world a picture perfect version of life that is miles away from reality, or hiding emotions and how they really feel about something.
Twitter, with its 140 character limit, has created a language all its own. Twitter’s users lay claim to a unique use of grammar, conversation, and have distinct identities that shape how content is shared and posted.
“Twitter users present only one side of themselves on social media, shielding their true feelings for good reasons, such as professional reputation. There is clearly a large gap between what people post on social media and how they really feel,” said Jensen.
Using Twitter for scientific studies could also damage the credibility of a scientific study, as drawing conclusions from a group of Twitter users doesn’t mean that group represents the population or their demographic in general.
Celebrities, politicians, big companies, and people who post multiple times a day could also skew data sets collected from Twitter and be over-represented Jensen noted.
“Enthusiasm for accessing digital data should not outpace sound research methodology,” concluded Jensen.