This particular experiment was referring to the influence Twitter has on Google rankings. Interesting because I just conducted an SEO class a couple weeks back and social influence on rankings was one of the questions brought up. My take was (and still is) that social services like Twitter do not have a direct effect on Google rank. Twitter can certainly help your site be discovered more quickly, but with that discovery comes links and no one will dispute that links improve rank.
So the SEO junk science comes in when the results of the experiment are reported that Twitter helped increase the ranking for the page in question from 7-8 to 1. The problem is there was no control in the experiment. Even the Mythbusters know that to have an experiment you need to have a control to be able to isolate what is actually making the differences observed.
A page, absent the influence of Twitter, on a well-ranked site will also rank well on Google for obscure terms within a day or so assuming all other on-page and on-site factors are optimized.
So this page is set up to rank for "SEO junk science." Believe it or not, there are no other pages optimized for this term directly, so my prediction is that this page, on the little ol' IPG site, will rank #1 for a search for "SEO junk science" within a few days. Do not tweet or share this article or you will negate the control aspect. Since the experiment is now complete, please feel free to share this post.
Check back here and I'll update with the results...
This page ranked #6 on Google for the term SEO junk science with just a weekend and no sharing anywhere. Put SEO junk science in quotes and it ranked #1. The post was originally published July 8, 2011, and the results were viewed on July 11, 2011.