Thinking, Fast and Slow Chapters 10 and 11


Chapter 10 is titled The Law of Small Numbers.  Kahneman poinbts out that even well trained researchers don’t always run large enough experiments.  He was disheartened that he reviewed one of his experiments and found it had a 50% chance not to show the hypothesis as true- even if it was- due to variation.

One example he included to make the point is that counties that have the highest rates of kidney cancer are rural counties in conservative states.  Counties that have the lowest rates of kidney cancer are also rural counties in conservative states.  These counties have small populations so you see the largest variance in rate in these counties.  The results have nothing to do with lifestyle in these counties- it’s just an artifact of using counties to determine population size when measuring kidney cancer.

He also worries about the Gates Foundation’s small schools initiative.  The highest pass rates are at schools that are small.  (But, small schools do not generally have the highest pass rates.)  This could just be variation and not an actual affect of small school size.  In fact it runs contrary to intuition at higher levels of schooling since larger schools can offer advantages like more variety in classes and programs.

Chapter 11 is about anchoring.  I have read about anchoring before.  Basically a number given at the start of a negotiation or before a person makes an estimate influences the outcome surprisingly much.  The estimates can differ by over 30% of the difference between the two anchors.

There are two possible reasons for this.  Both seem to play a part based on experimental results.  System 2 starts at the anchor and moves until it reaches an uncertain region when estimating.  If it reaches the uncertain region from above the estimate will be larger than if it enters it from below.  System 1 likes to find harmony in the world. So it may adjust world view slightly to make the anchor seem more reasonable.

I have two takeaways from this chapter related to my work at the college and my work with the union.  Almost none of our pilot student interventions reach 500 students in a single year so I should expect variability to be large and not proclaim or condemn a pilot early in its run unless there is a reason beyond the data itself.   When negotiating I should walk away from the table and force a reset if the other side starts from an unreasonable position.  Otherwise their anchor will dictate the negotiation.  Trying to set an equally obnoxious counter-anchor will create a difficult to bridge gulf.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s