Before starting to understand how the data can help answer the questions raised by the marketing vice president, Rick's first goal is to get a feel for what is actually in his data. Questions such as the following come to mind:
How many sales representatives are there and where do they operate?
How many practices are there?
How many physicians are in each practice?
In his initial review of the data, Rick also wants to look at the general quality of the data, since he has seen problems with this in the past. He is reminded of the old saying, "Garbage in, garbage out."
Rick decides to do some preliminary preparation of his data table. He will enter descriptions of the variables in each column and then group, exclude, and hide the ID columns. He will save his documented and reorganized data table in a new file called PharmaSales.jmp.
The steps quickly described in this section illustrate some useful features of JMP that can be used for documenting and simplifying your work environment. Although these steps may initially seem like unnecessary overhead, our view is that mistake proofing and a right-first-time approach is useful in any context. Often, the lean nature of these steps is not revealed until later. In this case, imagine a similar request to Rick the following year. But, for a quick analysis, these steps are unnecessary.
With that in mind, if you prefer to skip this section, feel free to proceed directly ...