That article failed to consider the echo chamber effect inherent to passive citation acquisition. There's an echo chamber effect to the way people share information on the web that I believe distorts not only these kinds of signals but run of the mill link signals, too.
For example, there is a kind of forum spammer that pretends to be knowledgeable by running the OP's query into Google then cutting and pasting their "discussion response" from one of the top five ranking websites. Non-spammer forum members do a similar thing. I see this every day in what people are linking to. Someone asks a question, another person Googles it. The link to an existing top five ranked site goes up.
Once a site ranks in the top five the site will begin to receive casual links simply from ranking in the top five. While it's important that the ranking page answer the query in order to get the link, once it satisfies this need the ranking becomes self-reinforcing, the link acquisition goes on autopilot.
I suspect Facebook sharing and other methods of spreading awareness of a site follows this trend as well. Google probably doesn't have to filter for this but if someone is going to do a study on Facebook ranking influence, I believe the better way is to actually move the SERPs (or not) rather than attempting to extract meaning from correlating the SERPs with Facebook data.
The above is one scenario Rand did not and could not investigate, as this was a study over the course of weeks. Still, he could have mentioned this as a possibility.
Link data was present for nearly every result we examined (99.9%+), which is to be expected, but social data? Of this magnitude? Even for plenty of weird, uninteresting queries? Shocking. If you had asked me to guess, I would have said we'd find Facebook share data on maybe 5-10% of the results - 61% is mind-boggling.
It's not mind boggling or shocking if you consider that citation patterns can follow a self-reinforcing pattern.
I noticed this phenomenom many years ago when I acquired a link from the ZEAL directory and started receiving traffic from sites that had copied and pasted those links into their website "links" pages. After Google became popular my sites started passively acquiring links from top five rankings in Google. As mentioned at the beginning of the post, I can see this kind of self-reinforcing linking in the links that forum members post. It's a common pattern and not surprising to see correlation of SERPs in FB, where FB data reflects the top ranked websites, not the other way around.
I am not denying that FB/Twitter can move the SERPs.
This post [webmasterworld.com] is a better read about how that can happen. I'm just pointing out a flaw in how the data was interpreted at SEOMoz.
