Ask most analysts and they’ll have a very straightforward theory about how decisions are made.

  • They did up numbers.
  • They put them into context.
  • Decision makers make a decision based off those numbers and context.

Only that they don’t.

Enter March and Olsen. In 1972 they programmed a simulation in Fortran. It’s called a Garbage Can Model. Their idea was a solid 40 years ahead of its time.

To summarize the Garbage Can Model:

  • Institutions are organized anarchies.
  • Problems, solutions, participants and energy go into a Garbage Can and shaken all around.
  • Solutions really search for problems.

When you mix it with Arrow’s Impossibility Theorem, you get a much more complete picture of why groups of people don’t behave the way you think they do.

To summarize Arrow’s Impossibility Theorem:

  • Individuals may have very straight forward preferences (Beef>Chicken>Pork), but when combined all together, the groups preferences frequently become circular. (Beef>Chicken>Pork>Beef>Chicken….)
  • That circularity forces decision making algorithms to be sub-optimal.
  • There is no perfect voting system.

To synthesize:

  • An algorithm that enables group decision making to be representative of the all the facts, opinions, and preferences to be rational, can not exist.
  • People cause organized anarchies.
  • Choice, itself, is ambiguous.

The first bullet point absolutely gutted me in my sophomore year. The second bullet saved my senior year. The third bullet point is putting a lot more wind into my sails.

There’s a dangerous underlining assumption that if organized anarchies (institutions, companies, departments) had better information, and make sense of it, that they would instantly make better decisions.

Is it really true that better informed people make better decisions?

I have anecdotes. (I make no claim that this is evidence.)

  • I make better decisions in the gym when I have my notebook with a complete history of what I lifted, on what.
  • I make far better eating decisions when I put in the effort to track it.¬†
  • I make better financial decisions because I track spending.

¬†Those are individual decisions. I don’t consult anybody in making them. It’s a dictatorship of one.

Is it really true that better informed groups of people make better decisions?

I’m not so sure that that is always the case.

Given what you now know about group decision making – what do you think?


I’m Christopher Berry.
I tweet about analytics @cjpberry
I write at

2 thoughts on “Groups of people don’t make decisions the way analysts assume they do

  1. Great topic of discussion.

    My 2 cents- better informed groups of people that care about facts have a higher chance of making better decisions.

    Bottom line is if you don’t care much about facts, but are driven by gut, politics, what you had for breakfast, whoever can shout loudest, whoever is in charge etc. etc. etc., its actually much cheaper to not be better informed- you save on analysts and all that tedious fact collecting. Also you can make decisions faster.

    That’s depressing for those of us who are analysts, but often the case I think.

    IF the group really tries to look at facts when making its decisions it must be better to have facts to look at.

    Maybe its better to have groups of people provide information to a dictator and have them make a decision. If its a good dictator…

    Group think tends to dilute everything, and in many cases being bold, and fast is better than being exactly right on the first decision. A dictator (or very small group) can use facts and consult widely, but make three different attacks on a problem, testing them in the real world, set up to reduce risk and find something that works while a group of people will still be trying to set up their meeting with everyones schedule and “socialise” the idea with “stakeholders”.

  2. Hi James,

    Thank you for the comment.

    I agree.

    And I will build on your point:

    There are facts and then there are feelings about expectations.

    Variation exists within any group. Some people are very comfortable with facts. Other people are very comfortable with feelings. Some do both well.

    How a group decides what to do, upon what basis of understanding, is directly influenced by the tensions between both groups.

    You will always encounter, in your life, people who will reject a fact on the basis that they don’t want a future to happen. It’s convenient reasoning, and it’s sly.

    Groups are gonna groupthink.

    Finally, I’ll add that if we understand why groups decide the way they do, we can be more effective at helping them optimize.

    What do you think?

Comments are closed.