SIMULATION: Structure, Performance and Ability
modeled on Edward Deming's Red Ball Experiment
©1999 Edward G. Rozycki
RETURN to Catalog
edited 4/2/12

To demonstrate that it is the structure of the task, not individual performance, that controls the range of normal variance of task outputs.

GAME 1 -- (materials needed: several sets of OV's showing cards, removable covers.)

1. Avoid bad card dealt face down: overhead reproductions of sets of cards with covers for overheads.

variation a. Use same set to permit memorization. Evaluation unfair to first person.

variation b. Change set to frustrate learning: Unfair to all

Discuss how structure of task affects estimations of ability

2. Keep statistics to determine + and - 2 sigma as limits.

Note: that odds are theoretically computable

GAME 2 --(needed: a. set of thumbtacks; b. set of different colored thumbtacks.)

1. Throw tacks.

A. Throw a handful of same colored tacks to get points down.

B. Mixed two colors of tacks, to throw one color points up.)

2. Compute statistics. Note that odds are not theoretically computable.