Percentage difference between two data sets

[walkerj7]

Hello,

I have two data sets that I have already summarized to achieve counts by a particular group.  Now I need to see the % difference for total, by particular group and subset of that group.  The table below represents the data structure from both datasets, and the counts are different and the fee totals are different.  After summarizing, how do I determine the % difference in counts to see who had more fees assigned (by title and title2), and then how do I see the % difference in fee totals?

Name	Title	Title 2	Fee
John Doe	Software	Software Leader	100
Santa Clause	Software	Software Leader	200
Jack Kent	Marketing	Marketing Analyst	100
Boo Douglas	Marketing	Marketing Analyst	150

Thanks,

JW

[cjaneczko]

Add additional Summarize Tools after this one. One tool would Group by Title and Sum Fee, the other Summarize Tool would Group by Title 2 and Sum Fee. From there you can do whatever calculations you'd like. 

[Raj]

Do want this or someting else

Name	Title	Title 2	Fee	Result
John Doe	Software	Software Leader	100	Lowest
Santa Clause	Software	Software Leader	200	100%
Jack Kent	Marketing	Marketing Analyst	100	Lowest
Boo Douglas	Marketing	Marketing Analyst	150	50%

[cjaneczko]

What determines the %? Why is 200 = to 100% and 150 only 50%?

[Raj]

@cjaneczko as it is 100% more than the lowest and similar with 50%

100+100%= 200

[cjaneczko]

Try this.

[Raj]

@cjaneczko 
this is to be confirmed by @walkerj7  is he looking for this or something else.

[walkerj7]

Thanks for all the advice on how to solve this.  Just for transparency my output should look like the following:

Title	Title Count within its dataset	Title % of Total (Based on it's dataset)	Title Count Difference between datasets	Average Fee % difference (fee total in dataset a vs fee total in dataset b)
Marketing (Dataset A)	20	37%	40% lower than Dataset B	$50 lower
Marketing (Dataset B)	35	48%	40% higher than Dataset A	$50 higher

Then I would do this for Title 2 and in the end my analysis would determine which titles have higher fee counts and which titles have a biggest difference in fee totals.

[Raj]

is this the complete data set for which you have shared the sample output

[cjaneczko]

Going to need a lot more sample data to get the info you need. A sample of Data A and Sample of Data B. But you would need to add quite a few summarize tools, some Unions, some Append tools with Join tools to get to where you need to go. Its all doable, but a larger sample size of both data sets would be good. Its difficult to tell how you got to the 40% and $50 fee without the math that shows it, or sample data that adds up to that 40% and $50. The original samples don't add up to your desired output. Are you comparing the fees and title counts between the two sets against all titles or just same titles?