How participation inequality blinds the wisdom of crowds and creates the opinions of the few
(Editor's note: This post is being re-published and highlighted due to its popularity.)
The wisdom of crowds theory as advanced by the New Yorker columnist James Surowiecki contends that the collective opinions of a group of people is often smarter than the opinion of a single member of the group, or for that matter, the opinion of an expert. The central thesis is based on the fact that the diversity, independence, decentralization, disorganization, and aggregation of opinions make the entire group reach a smarter decision. Corollary, when opinions are too homogeneous, too centralized, too divided, too imitative, and too emotional then the crowd usually fails to achieve its collective intelligence. For instance, our judicial jury-based system is based on this seemingly counterintuitive notion.
The Participation Inequality theory as advanced by Will Hill of AT&T Laboratories and later cited by Jakob Nielsen in his Alertbox blog posting entitled "Participation Inequality: Encouraging More Users to Contribute", states that only a tiny minority of users in an online community contributes for a disproportionately large amount of content creation. Specifically, user participation in an online community follows the "1% rule", sometimes also referred to as the "90-9-1 rule":
90% of users are lurkers who tend to read content but rarely, if ever, contribute anything whether creating original content or commenting on any content.
9% of users contribute occasionally by either writing original content or commenting on existing content, even though we now know that most tend to do the latter.
1% of users contribute the most by regularly creating new original content or commenting on existing content.
Since those two theories came about, and rolling forward to 2005 when a wave of entrepreneurs, including myself admittedly, attempted to exploit the potential of the wisdom of crowds, I now candidly recognize how we were all blinded by the light of Participation Inequality. To put it even more bluntly, those of us who knew about the Participation Inequality chose to ignore it and preferred to be carried by the excitement of the wisdom of crowds.
Current rating/review/vote/reputation/meritocracy systems collect the opinions of the few - that 1%. Some may argue that there are ways to reduce such participation inequality as suggested by Nielsen himself, nonetheless, inequality remains even if reduced. Thus, on close analysis, the minority represented by that tiny 1%, enhanced or otherwise, consists of the extreme. For example, in a political blog, the minority consists of diehards who fall on the left or on the right. The great majority of people who fall in the center belong to the lurkers group - they never bother to express their opinions. In an example of a product commentary, consumers who are likely to comment are the ones who either love the product or hate it. Thus, in order to be enticed to comment, passion (for or against) is the motivating factor to engage by expressing one's opinion.
It is indeed this passion which is the main cause of breach of the wisdom of crowds assertion. Specifically, since the great majority of reviews come from the tiny minority of diehards who fall on either side, then the collective opinion is too homogeneous, too centralized, too divided, too imitative, and too emotional to be a wisdom of any kind.
In retrospect, and very candidly, those of us who claim to be building systems based on the wisdom of crowds aren't doing so. That doesn't necessarily mean that what we're building is useless, but let's admit that it isn't the wisdom of crowds but the opinions of the few.
If my analysis is correct, then the next big question, begged to be asked is obviously the following: "is the opinion of the few valuable?"...