The question of surveys, and employee engagement has been buzzing around the ActKM listserv and more recently the use of Wisdom of Crowds came up. Both positive and negative comments ensued. This morning the Surowiecki’s three conditions were posted. They are
These conditions show why the title “wisdom of crowds” is a bad one though, crowds are herds; we see that with the blogosphere, response to opinion polls and many other examples. Understanding the theory is important here and its a good illustration of the difference between complex and chaotic systems.
The types of examples that Surowiecki talks about (in the main, he gets some wrong) are examples of chaotic systems – all the agents are independent of each other, so if there are enough, and there is diversity then you can use large number mathematics. in Cognitive Edge we now do that with mass consultation on issues of strategy and I am about to start a project in Washington which will use mass consultation of INDEPENDENT agents to look at aspects of the current financial crisis as well as to complete a lessons learn/learning capture. It will also test employee engagement by creating the type of statistics you get from employee engagement surveys, but linked to the narrative of day to day experience to provide an explanation for what would otherwise be numbers without meaning.
On the other hand a crowd, or some of Surowiecki’s examples such as prediction markets are complex systems, the agents, are aware of each others actions so you don’t get the “Jelly Bean phenomena”. This illustrates the point well: if everyone at the county fair guesses how many Jelly beans there are in the jar, writes their answer down on a piece of paper and places it in a ballot box then the average of the group will be the correct answer. On the other hand if the first person writes their guess on the wipe board by the side of the jar then subsequent individuals will start to bracket the first guess, and you will get an emergent dominant pattern and almost certainly the wrong answer.
If nothing else it needs (to return to another theme I have been emphasising of late) an understanding of theory to understand how to scale an observed case. Just repeating what other people report they do is dangerous, simplistic and foolish. Theory should always inform practice.
It’s not the only bad title by the way, the other one which annoys me is “Black Swan”. In practice no one was surprised by the fact that some swans were black, the example was created by a philosopher to illustrate issues of definition. However it sells well, but I preferred Fooled by randomness by the same author, both as a book but also for a title which reflects the main theme of the book.
Cognitive Edge Ltd. & Cognitive Edge Pte. trading as The Cynefin Company and The Cynefin Centre.
© COPYRIGHT 2023
You arrive at a hotel in Washington at around 1930, after getting up at 0330 ...