Saturday, January 16, 2016

What percent of your speech audience really likes you? The 20-60-20 rule

I heard that people can be divided into two categories: those who like to divide things into two categories, and those who do not. The first category would be content to hear that perhaps a famous comedian said:

“Half the audience will like you, but the other half won’t.”

The second category instead would be more intrigued by a LinkedIn Pulse article on January 4, 2016 by Alf Rehn titled The 20/60/20 rule - or why you shouldn’t worry about frowns in the audience. It can briefly be paraphrased as follows: 

“In every audience, there will be 20 percent of listeners who don’t really like you all that much.

Conversely, in every audience, 20 percent of listeners will really, really like you.

In most audiences, 60 percent could go either way.”

On December 16, 2015 Gerry Sandusky similarly blogged about The 20-60-20 rule from the front of the room. His post also appeared at Presentation Magazine.

It sounds plausible, and suggests that you would do well to ignore those inevitable frowners. Instead concentrate on the rest of the audience. You might persuade some of that 60 percent that they can learn from listening to your speech. That 20-60-20 is a simple bar chart, a very rough histogram for a bell shaped curve.

You could get slightly fancier and instead show a 10-40-40-10 bar chart. Here you would start out by having half the audience in your favor. 10 percent love you, and 40 percent like you. Then imagine changing the other 40 percent who don’t like you. Forget about the last 10 percent who just hate you.    

You could get even fancier and instead show a 10-20-40-20-10 bar chart. Here you start out by having almost a third of the audience in your favor. 10 percent love you, and 20 percent like you. Then imagine changing the other 40 percent who, as of yet, don’t care. Forget about the 20 percent who don’t like you, and last 10 percent who just hate you.    

A normal or Gaussian distribution is what we’d imagine to be hiding behind the 20-60-20 rule. Last year there was an article at ArXiv on The 20-60-20 Rule by two Polish mathematicians, Piotr Jaworski and Marcin Pitera. Their abstract says that if a random vector follows multivariate normal distribution and we split the whole population into three groups, then this fixed ratio leads to a global equilibrium state.

Alf Rehn also mentioned that you might misread the reaction of an audience member. One he thought was seething with anger turned out to have been so deeply engaged he’d memorized parts of Alf’s speech! Similarly, back in 2011 in a blog post about Learn to ignore these audience behaviors I mentioned a speaker who was disconcerted by seeing a woman (presumably a professor) who kept screwing up her face in disagreement and contempt. She just was a former graduate student with a very mobile face and a habit of attending lots of lectures.
 The 20-60-20 rule also shows up in management articles that claim it represents how people perform on the job - 20 percent are strong (go-getters), 60 percent are average (mediocre), and the last 20 percent are weak (slackers).

But there also is a nasty bunch of performance review claptrap claiming there instead is a mythical Vitality Curve justifying “rank and yank.” Here the three categories are 20 percent stars (the tigers), 70 percent average (the sheep), and 10 percent not up to expectations (the goats).

Back in July 2003 Andy Meisler discussed that in a Workforce article titled Dead Man's Curve. There isn’t a solid basis (like a Generalized extreme value distribution) for why 10 percent are scapegoats. It is just an arbitrary decision. It’s really not about a bell curve. It’s just about for whom the bell tolls. Perhaps the Machiavellian thought is that flushing more than ten percent down the toilet annually would completely decimate corporate performance. Still, if you repeat the yanking process about three times you likely will eliminate what once were some very solid performers.

On November 14, 2013 there was a commentary article in the Wall Street Journal titled Jack Welch: ‘Rank and Yank’? That’s Not How It’s Done and subtitled Using ‘differentiation‘ aligns employee performance with an organization’s mission and values. That reuse of the word differentiation belongs in the Guffipedia.    

The image of a normal distribution came from Wikimedia Commons. 🔔

No comments: