Social media has ways of affecting our perceptions of what is real and fake, and how others perceive us and our beliefs. The implications of those perceptions are examined in separate research projects by Clemson University business professors.
A study conducted by Marten Risius, associate professor of management in the College of Business, illustrated how social media users can discern legitimate news from a fabrication of it on social media channels. A second study, by Phil Roth and Jason Thatcher, professors of management, determined that the likelihood of a person being hired can be affected when their political beliefs are discovered over social media outlets.
Risius was joined in his research by Christian Janze, a Ph.D. student from Goethe University, Frankfurt, Germany.
“Our study investigates how to automatically identify fake news using information immediately apparent on social media platforms,” Risius said.
The study examined cognitive, affective and behavioral cues in more than 2,000 Facebook news article posts from left, right and mainstream media outlets during the 2016 presidential campaign. Researchers used 230 samples of fake news and an equal number of real news. Using their process, they were able to determine fake from real with 80 percent accuracy overall, when all of the articles were examined. They then trained the algorithm so it could correctly detect 90 percent of the 230 fake stories.
Risius said affective, behavioral, visual and cognitive cues go a long way in determining if a news article is real or fake.
“Emotional cues, such as love or anger, or the amount of likes or shares are telling on a story’s legitimacy, as are exclamation and question marks and the use of quotes,” Risius said. “For example, the word count, or using all caps, exclamation and question marks in a post are strong predictors of a story being fake.”
Roth and Thatcher’s three-year research project involved more than 400 participants and found political beliefs can play a significant role in how a hiring manager assesses a job applicant’s qualifications.
The pair created two versions of a student Facebook page infused with the political leanings of either a Democrat or Republican. The pages were sent to two groups — upperclassmen business majors at a Southern university and employed MBAs. A series of hiring-related questions accompanied the Facebook pages. At the end of the survey questions, the participants were asked to identify themselves politically.
“The bottom line is that decision-makers have a tendency to hire like-minded people,” Roth said. “Based on our experiment, the study showed political similarities correlated with ratings on how a respondent liked the candidate and how they thought the candidate would do their job.”
Roth said that for job applicants, sending clear signals of their political affiliations on social media channels carries risks and rewards.
“If you get a decision-maker who has the opposite of your political viewpoint, there may be consequences. On the other hand, should the hiring manager align with your political beliefs, your chances of being hired may be enhanced,” he said.
Hiring managers should also be cognizant of the power of similar or dissimilar political beliefs among them, applicants or subordinates seeking promotion, according to Roth.
“There is no evidence political affiliation relates to job performance, so a decision to hire or not based on a political ideology could be construed as political discrimination,” he added. “Hiring based on one’s political belief can not only hurt a qualified candidate who didn’t get the job, it can impact the organization if the best applicant wasn’t hired.”