In the simulation of Deming’s funnel experiment, the

@RISK outputs show how tampering leads to poor results, at least in terms of the mean and standard deviation of the distance of the final drop from the target. However, the results we presented don’t show how the tampering rules, particularly rules 3 and 4, go wrong. To get a better idea of this, create two scatter charts, one of the x-coordinate in column D versus the drop number in column A, and one of the ycoordinate in column E versus the x-coordinate in column D. (You could also create a third scatter chart, of the y-coordinate versus the drop number, but it would be about the same as the first.) Use the chart subtype that “connects the dots” for each scatter chart.

To go from one rule to another, enter a number from 1 to 4 in cell B3, not a formula. Then press the F9 key several times to see how the scatter charts change.

Describe how the drops seem to evolve over time according to the various rules.