4 Replies Latest reply on Nov 6, 2018 8:17 AM by morton hsiao

# Histogram | Comparing Distribution of Two Measures

and

and made my own overlapping histogram, but I have two problems with it.

My data looks like

select value, 'g' as source from table 1

union all

select value, 'a' as source from table 2

And I made my measures as

aerr := if [source] = 'a' then value end

gerr := if [source] = 'g' then value end

So when I do the double histogram, the % of total number of records takes the total number of records as the denominator, not just the number of records where source equals g or a.  I think I may have to use a lod expression, which is of course easier than dax, but before I try, is this the way to go?

• ###### 1. Re: Histogram | Comparing Distribution of Two Measures

You need to extract your data and then save it as a packaged format. Otherwise we cannot do anything since it's asking for your credentials to your database.

• ###### 2. Re: Histogram | Comparing Distribution of Two Measures

Of course...

It's tricky because you have to do two things.  You have to save as a packaged workbook and click the extract button.  isn't the whole point of the extracted workbook the fact that the data is in it.  otherwise, what's the difference between an extracted workbook with no data and a normal workbook?

• ###### 3. Re: Histogram | Comparing Distribution of Two Measures

I am not sure but are you expecting an output like this?

• ###### 4. Re: Histogram | Comparing Distribution of Two Measures

Not really.  My histograms are fine, they are just not normalized to their respective series.  So the sum of the bars for each color should sum to 1.  right now it sums to 50%.  That's the issue I'm having.