The “Wisdom of Crowds” and Social Media Research Strategy

Architecting Better Data-driven Digital Experiences

The “Wisdom of Crowds” and Social Media Research Strategy

A few posts back, I wrote about the interdependence of communication and feedback in a social media listening strategy. That’s why I couldn’t pass up some commentary on this recent Wired article on a study that finds that sharing information actually reduces the “wisdom of crowds”.

This finding provides an intentional instance of the “observer effect” bias addressed in the earlier post, in which interaction with a population shapes the results coming from that population. In social media engagement, interaction with the population we want to learn from (through our listening program and other social media research) is unavoidable, thus, we must design our interactions to elicit the best feedback possible by accounting for the observer effect. This recent insight into the accuracy of data culled from crowds must clearly be taken into account by anyone conducting opinion or preference research from social media conversations.

The article succinctly defines crowd wisdom as “the statistical phenomenon by which individual biases cancel each other out, distilling hundreds or thousands of individual guesses into uncannily accurate average answers.” However, this only works when the answers provided by each individual have been independently established. Once respondents become aware of the answers provided by others, the average result becomes less accurate.

From the article:

The researchers attributed this to three effects. The first they called “social influence”: Opinions became less diverse. The second effect was “range reduction”: In mathematical terms, correct answers became clustered at the group’s edges. Exacerbating it all was the “confidence effect,” in which students became more certain about their guesses.

“The truth becomes less central if social influence is allowed,” wrote Lorenz and Rahut, who think this problem could be intensified in markets and politics — systems that rely on collective assessment.

I suggest that this problem can also be intensified in social media “communities” – where the collective assessment of a positive or negative attribute of a product/brand will present an inaccurate evaluation of the true average of individual views since opinions at the extreme will be observed, shared and amplified.