Will you e-mail this column?

Sociologists have developed elaborate theories of who spreads gossip and news who tells whom,who matters most in social networks...

Sociologists have developed elaborate theories of who spreads gossip and news who tells whom,who matters most in social networks but theyve had less success measuring what kind of information travels fastest. Do people prefer to spread good news or bad news? Would we rather scandalise or enlighten? Which stories do social creatures want to share,and why?

Now some answers are emerging thanks to a rich new source of data: you,Dear Readers.

Researchers at the University of Pennsylvania have intensively studied the New York Times list of most-e-mailed articles,checking it every 15 minutes for more than six months,analysing the content of thousands of articles and controlling for factors like the placement in the paper or on the home page.

The results are surprising. I would have hypothesised that there are two basic strategies for making the most-e-mailed list. One,which Ive happily employed,is to write anything about sex. The other,which Im still working on,is to write an article headlined: How your pets diet threatens your marriage,and Why its Bushs fault.

But it turns out that readers have more exalted tastes,according to the Penn researchers,Jonah Berger and Katherine A Milkman. People preferred e-mailing articles with positive rather than negative themes,and they liked to send long articles on intellectually challenging topics.

Perhaps most of all,readers wanted to share articles that inspired awe,an emotion that researchers investigated after noticing how many science articles made the list. They found,20 per cent of articles that appeared on the Times home page made the list,but the rate rose to 30 per cent for science articles,including ones with headlines like The Promise and Power of RNA.

To make sense of these trends in virality,the Penn researchers tracked more than 7,500 articles published from August 2008 to February 2009. A random sample of 3,000 of these articles was rated by independent readers. The computer textual analysis could identify affect-laden articles like Redefining Depression as Mere Sadness or When All Else Fails,Blaming the Patient Often Comes Next. It distinguished positive articles like Wide-Eyed New Arrivals Falling in Love With the City from downers like Germany: Baby Polar Bears Feeder Dies.

Story continues below this ad

More emotional stories were more likely to be e-mailed,the researchers found,and positive articles were shared more than negative ones. Longer articles did better than shorter articles. Surprising articles,like one about free-range chicken on the streets of New York,were also more likely to be e-mailed.