I've had social news on the mind this week, especially with the announcement that the social news aspects of Netscape.com will move over to a new site.

It's either synchronicity or serendipity that the Project for Excellence in Journalism (part of the Pew Research Center) has a new study out -- The Latest News Headlines -- Your Vote Counts -- that compares the featured stories on mainstream media sites, with those on social news sites Digg, Reddit and Del.icio.us (as well as Yahoo News) [link via BBC News]

[Actually, looking at their list of sites, I would split a hair and say they looked at social linksharing sites (where users submit, vote and comment on links), which is a broader category that includes social news sites, which are explicitly news-focused: Newsvine, NowPublic, etc.]

During a one-week period this June, they analyzed the featured stories, and came up with some general observations:

* Disparity: There wasn't a lot of commonality between mainstream and social news sites; MSM sites tended to be more internationally-focused, to stick with big stories and come back to them again and again. On the other hand, the social news was more "twitchy" -- rapid cycling, with stories passing quickly out of view with little followup.

* Diversity: Social news sites were also a lot more likely to draw from non-news sites like blogs. In fact, "Digg users dug blogs more than any other source," which was slightly surprising to me, since so many Digg users reflexively profess to hate blogs (especially "blog spam," which is what they tend to call any intermediary link, even if it adds value and isn't simply an unneeded middleman).

* Differences: Additionally, they noted differences between the social news sites: Digg was more tech & science focused (shocker -- not uncharacteristic, even if it was iPhone launch week, too), as well as the most US-centric; Reddit was the lowest-traffic and most political; and Del.icio.us was the most eclectic.

On top of all that, they also looked at Yahoo News' user-influenced categories (Most E-Mailed, Most Viewed, Most Recommended), and also saw that the user picks didn't jibe with editors' picks, with some differentiation between categories --most viewed was most sensational (crime and celeb gossip).

(I would just like to add a personal observation -- for the photo sections, I would also note a healthy and consistent selection of cheesecake in the most popular photos for any given time -- swimsuits, celebrity skin and lingerie fashion shows.)

Anyway, their conclusions are a little soft (we need to keep an eye on these differences), but it's an interesting study.

I'm a little tired, so I may have mischaracterized the results. If I missed something, or if you have your own observations, please leave a comment.

Thanks -- Joe