fixing unstable variables

[#1]

How do you correct an unstable variable in a data set?

[#2]

Can you please define "Unstable".......it has outliers?, There are typos in you data input? The distribution is not from a normal distribution?

[#3]

Its on the age range I created. The age variable shouldnt have any outliers in the orginal data set. If you look at the Age variable it is fairly normally distributed. Age as continuous is significant but Age Range shows bias. 

If you run a regression on the data set you'll see where the unstability is. 

[#4 · Accepted solution]

I ran your example using the Age Ranges and I was supprized how JMP handled the column.  I had expected JMP to collapse all of the age data into your 4 groupings.  Rather, it broke out all of the different ages into separate(Ordinal???) effects.  I am not sure that is what you wanted, and I am not sure what is really going on.  I assume what you really want to do is to group the data into the 4 groups, and to use those groups as input into the regression.  If that is the case, I suggest that you create a new column and then use a formula to create the 4 groups.

If( :age < 35, "Millennial", :age < 50, "Generation X", :age < 65, "Boomers", "Greatest" );

[#5]

YES!!! That fixed it. Thank you!!