I use this Excel table. It has all the demo tables for the excel wizard along with some scenerios for life science and semiconductor.
Worksheets in the Workbook
1. Disorganized: a page of messy data with a graph.
Use this page to demo the Excel JMP tab (PC only).
Set preferences, first row for column names.
Select C11 to H19 and click the data table button in Excel (JMP add-in) and look at the table in JMP
Part 2: Copy the small table, D33 to F26. Select, copy, JMP, new table, paste with column names
What you see is what you get - note the excel data is rounded and the rounded values are pasted
Now use the JMP widget in excel to do the same thing, the unrounded data is sent to JMP
Warn, that this is a MS issue, MS copies the rounded numbers to the clipboard, but JMP grabs the actual numbers
numeric and character: Useful for talking about the effect of a character row in a continuous numeric column and what JMP does with it
Formatted: Use the JMPs excel import wizard to import this.
Specify where the column names are and use span missing cells
Raw: just like the Formatted table, except like you would expect to get from a database query
This is how we would like to start in JMP, with just the raw data
Stability Data: The same data set split by time point and split by lot number. Import the data, Stack it, check modeling and data types, run stability platform.
LIMS(1) to LIMS(8): simulated process data from a laboratory information management system. Use the excel import wizard, concatenate all 8 tabs, add a column with the tab name and import. The split the data by the method group by ID, keep all option checked to get date col. Then make control charts from the different methods.
Experiment: This is a time course experiment that was repeated with three independent samples.
For each time point, and each sample there are two measurements, the protein concentration and a control.
Think of this as a protein expression experiment,
We are looking at an induced protein vs. a constitutively expressed protein.
The expressed protein level is normalized to the consituitively expressed protein (its a ratio)
In Excel this is easy to do, the third table is a a set of formulas to calculate ratios
In JMP we need to have one column for the time point, one column for the sample ID, one col for protein, and one for the control
Import the protein and control tables separately, stack each one, and then join them. Then add the formula col for the ratio
In Graph builder, make all three plots.
Use Fit y by X, group by sample ID and fit the curves.
Use Fit model to see if the samples are different (make sure to fit a quadratic term.)
Power Converter: Data in a format that I've seen from semiconductor companies.
The data is just the semiconductor capability data set thats been mangled to fit this format.
Import the spec limits: This is weird but it works. With the excel import wizard, set the column names for row 4
Set the data to start on row 2, and end on row 3. (I know this is upside down but it works)
Set the second column as the start of the data and then import it and transpose or stack
Import the data, row 4 is the column name, the data is below
Open Capability and import the specs from the table, also save the specs to the column property later.
electric cars: This is a couple of years old, but its a list of electric cars registered on the SAS campus with their praking lot.
The first column needs to be parsed. Compare Text to Columns in JMP and Excel. Good for data visualization.
This is really just a backup data set
Process Worksheet: Use this in a quality demo to compare the difficulty of the workflow in Excel vs. JMP
Phone: Phone charges and use by month. another backup data set
Spanned Layout: Use the Excel import wizard to open this page. Demo hierarcheal column names.
This is from the new in JMP11 demo
Potato Production: This is from the new in JMP11 demo. Demo the import wixard with this
JAN to DEC: Demo concatinating worksheets. This is from the new in JMP11 demo
DOE Table: Small DOE worked out in Excel.
Main point: small tools for small experiments. Also, Even a small DOE is better than OFAT.
G-Pad: Some data that came from graph pad
