REPLY Why don't organizations function better? (SD6622)

[SDMAIL Jim Hines]

Posted by  "Jim Hines" <jim at ventanasystems.com>

>>>> The organization has a problem: who is contributing

Tom's point is important.  In many organizations, hardly anyone makes a
contribution as an individual.  Rather the contribution is made as part of a
group.  

One way around this problem is a two-pronged approach:  (1) The organization
recognizes the results of a defined group, and (2) people switch groups
periodically.  Jody House and I experimented with simulations of this sort
of "selection".  The results were almost as good as simulations in which the
organization could discern individual performance (i.e. a simulation in
which the "pointing and pushing" mechanism (i.e. selection) acted at the
individual level rather than at the group level).

We also ran an experiment with people over a semester in a system dynamics
course we were teaching.  People got to choose whether they wanted to
participate in the experiment or not.  It turned out that half the class did
and half the class didn't.  Those who opted out of the experiment did
homework assignments the way we'd always had the class do them -- in a group
that lasted basically the entire semester.  The students who opted in had an
interesting experience:  For each assignment, we distributed the students
randomly to a number of groups.  Jody and I rank-ordered the handed-in
solutions and gave points in accordance with a group's rank, each member of
a group receiving the given number of points.  For the next assignment we
randomly assigned the students to new groups, etc (i.e. each student was in
a new group for each assignment).  Each student's cumulative number of
points was made public to his current group.  

These points constituted a "pointing and pushing mechanism" (i.e. selection
pressure): The points "pointed" to those who had accumulated many of them
and -- because points were good things -- "pushed" others in the group
toward a desire to imitate the high-point students
(imitation=recombination).  

There were three results (1) the students liked switching groups a lot, (2)
the people participating in the experiment did better in the course on
average than people who did not, and (3) the people with the most points at
the end of the term were the people who Jody and I thought probably were the
best students.  (Note there are some obvious caveats in interpreting these
results -- but this message is already long enough that I won't elaborate
unless someone asks).  

Jim
Posted by  "Jim Hines" <jim at ventanasystems.com>
posting date  Thu, 13 Sep 2007 12:29:43 -0500