'Twas the night before field close and all through the internet,
not a respondent was responding, not even a Female head of house.
The questionnaires were posted on the net
with care in hopes that data soon would be there.
The clients were nestled all snug in their beds,
while visions of PowerPoint decks danced in their heads.
And the project team has just submitted the model request form.
When out in the office there arose such a clatter,
I sprang from my cube to see what was the matter.
Away to the PC, I flew like a flash.
Tore open the zip file
and through up the SPSS.
The mean PI of the new concept gave a glimmer of hope
to the client service folks below.
When, what to my wondering eyes should appear,
but a little multicollinearity and 8 respondents with missing data.
With the drivers unclear,
I knew in a moment it must be a Treenet.
More rapid than correlations the importance scores came.
And he whistled and shouted and called them by name...
Functional benefits, EMPACT and AffinID.
To the top of the stack, to the top of the pile,
now for importance we understand all.
As bivariate plots against dependent variables show
when they meet with a knot, they bend on the fly.
Non-linear obstacles we decry,
the relationship is clear, and now we know why.
With the click of a SEND, the report deck went out.
The clients were happy, the results made sense.
But I heard them exclaim ere they stopped for the night.
That’s just wave 1, wave 2 is coming so just sit tight.
Happy Holidays from all of us at CMB.