spssforum.com

I have sales mix variables from 1 - 5 (Mix1-Mix5). They each represent a % of firm sales in each market channel:
Mix1 = retail
Mix2 = food service
Mix3 = institutional
Mix4 = industrial
Mix5 = other

I want to create a new variable that identifies which channel has the most sales (ex: MixMax = channel that accounts for most sales).

For example, if case #1 had the following:
Mix1 = 40
Mix2 = 10
Mix3 = 25
Mix4 = 25
Mix5 = 0

then MixMax = "Retail"

How do I do this in SPSS 16?
Any ideas on how to deal with ties (ex: Mix1 = 50 and Mix2 = 50 and all others =0)?

Under Compute there is a max function

Check it out and let us know if this is what you are looking for

_________________
NOTE: Please read the Posting Guidelines and always tell us your OS, the SPSS version and information about your study and data!


Statman
Statistical Services
IBM/SPSS Beta Site

The max function simply kicks back the maximum value. I want to note which of the variables has that value. So, for my original example I am trying to note that the largest sales channel for the respondent is the retail channel. Does that help clarify my question?

Okay, I finally figured out a way to get there. Here is what I did:

1. Created a MixMax variable by computing the max(mix1, mix2, mix3, mix4, mix5).
2. Created new variable MixMaj
3. Filtered data on condition mix1=MixMax
4. Recoded MixMaj - missing variable changed to 1
5. Repeated steps 3&4 for 2 - 5
6. Checked for equal splits by filtering all data on mix# = mix$

This worked. However, I am curious to know if there is another way to do this work. Any ideas?

Well done

Thread closed