Compute column based on sets of users

[#1]

Hello - Can someone help solve this problem?

The data:

- File 1 is user-level data that has one row for each product the user has bought

- File 2 is totals for the entire set of users

Desired output:

- I want to compute the % of users that bought all products based on a selection of products.  That is, if I limit product to "toys", I want to compute the % of those toy-buying users that also bought every other product

- I want to compare that toy-buying user set to a fixed total from the entire set of users.

The issue:

- this works fine when no selections are made - see image:

- But when I make a product selection, it naturally shows only the users that bought that product:

I need the table to show the buy rate for users that purchased groceries, instead of limiting all the data to just records associated with product = groceries.  I suspect this is related to p() in set analysis but can't quite get it to work.

The sample QVW file is attached.  All suggestions are appreciated!

[sunny_talwar]

This?

[sunny_talwar · Accepted solution]

Or this

[kaushiknsolanki]

What is the expected output?

Regards,

Kaushik Solanki

[#5]

Thanks so much - yes, this is what I needed.