A fictional tractor building company looks to understand why sales have decreased in the past year. Comparing the performance of two sales guys based on control charts yields very different results than a traditional "year-over-year" percent change analysis.
15. and what if we applied
some brand new
techniques?
Walter Shewhart written in 1939
http://amzn.to/dOWKBw
16. to understand the
variation in the data:
“Special Cause” variation
UCL – Upper
Control Limit
“Common Cause” variation x-bar (or average)
LCL – Lower
Control Limit
“Special Cause” variation
31. …and the fate of our
sales guys’ careers hangs
in the balance…
Joe Larry
32. The Morals of the Story:
• DON’T just compare % change
• DON’T assume a % change is
statistically significant just because you
don’t like it.
• DO use a Control Chart to filter out
noise in a data set and identify signals
• NEVER adjust the y-axis to cross the
x-axis at anything other than 0 (did you
catch that trick? The column charts
grossly over-exagerate the changes)
33. For bar charts, always set
the y-axis to 0
y-axis starts at 157 y-axis starts at 0
Appears to
be 4X less