Monday, 30th October 2017

E- paper

Bad taste? Don't blame FB - Study negates contagion effect, barring music influence

Read more below

  • Published 20.12.11

New Delhi, Dec. 19: Students on Facebook appear to share interests because they have similar tastes to begin with and not through contagion-like, person-to-person spread of interests, according to a study of friendships on the online social network.

While the findings may come as a relief to sections of parents worried that their children may acquire undesirable interests or tastes through friends on Facebook, the researchers caution that their results may generate other concerns.

“If only people with similar interests and similar tastes tend to become friends, their world views may become narrower and narrower — this may also be a concern,” Jason Kaufman, the principal investigator of the study, told The Telegraph over the phone.

The study by researchers at Harvard University shows that students who share tastes in music and movies are significantly likely to befriend each other and that the spread of interests among friends across social networks occurs only rarely.

The findings, to appear tomorrow in the journal Proceedings of the National Academy of Sciences, challenge the idea indicated by earlier studies that interests — from moods to dietary preferences — can spread from person to person across social networks.

“Scientists may need to rethink this contagion effect,” said Kaufman, a fellow at the Berkman Centre for Internet and Society at Harvard University. Kaufman and his colleagues have analysed the evolution of friendships and tastes in music, movies and books as declared by 1,604 US college students through their Facebook activity over a period of three years.

Their analysis suggests that friends tend to share some tastes not because they pass on information about tastes and influence one another but because these shared similarities prompted them to become and remain friends.

But the researchers found that among the 15 sets of tastes they examined, only one — classical or jazz music — appeared to spread across Facebook friends. Students whose friends had expressed tastes for classical or jazz music themselves adopted this cluster as their own interest. The scientists speculate that this specific interest may be “contagious” because of “its unique value as a high status cultural signal”.

The study is among the first to disentangle the importance of underlying tastes and peer influence in shaping interests — but its findings on why people become and remain friends are similar to the results of earlier studies of other traditional social networks.

“The evolution of friendship is governed by homophily — birds of a feather flock together — with respect to gender, racial background, and socio-economic status,” said Kevin Lewis, a social sciences graduate researcher at Harvard University and first author of the study.

In an analysis of a subset of 1,001 students, the study found that the log-odds (a measure of the probability) of two students becoming and remaining Facebook friends increases by 0.56 if they share the same academic field. It found that a single friend in common increases the log-odds of two students becoming and remaining friends by 0.10 and this effect multiplies with each additional shared friend.

The researchers say their study shows how Facebook allows social scientists to analyse human interaction in a natural setting on an unprecedented scale without people knowing that they are being tracked.

In most traditional studies of preferences and tastes, the participants knew they were being observed. However, Lewis said, the analysis of online social networks raises questions about how the data and the findings should be interpreted.

The analysis of such electronic data from online social networks “eliminate problems relating to data accuracy and scale, while replacing them with problems of analysis and interpretation”, Lewis told The Telegraph.

“We’ve shown that, among this group of college students, taste-based social selection is more important than taste-based peer influence, and that a number of forces that have nothing to do with tastes shape the evolution of friendships on Facebook,” Lewis said.

But it is still unclear what exactly a Facebook friendship corresponds to in “real life” and, Lewis said, to what extent tastes reported on Facebook are the authentic preferences of those making those claims.