Searching for similarities

[AA364]

I have a large data set - over 100 people, each with 5 strengths associated with them, there are 34 total strengths. I'm wondering what tools I could use to tell me which people have the most in common?

Does something like that exist? beyond just filtering and manually looking?

[jteeple]

A tree Map would visualize that nicely.  

[AA364]

 @jteeple I am using a Tree Map! I was hoping that when I selected multiple strengths the list of names would go down so I would only see who had those say 3, but instead the list is anyone who has any one of the 3.

I'm not quite at @Greg_Deckler 's level yet - and unfortunately I can't open the PBIX he sent due to version compatibility.

If you have suggestions on how to change the tree map so I could select multiple values and only see people with both - that would be great.

[jteeple]

Have you created measures before?  When I first started, it was all very Greek to me.  What you should do is 

1. Create 'New Table'

2. Get list of Distinct Names by typping Names = DISTINCT('Name of your table here'[Name])

That will create a table called names with a column called Name and fill it with the disctinct values from your original set of data.

3. Create 'New Measure'

4. Copy Greg's measure.  You can use his names which may help the following explanation

% Alike = name of Measure or calculated value)

Variable PersonTraits = distinct list of traits from original dataset

Variable CompareTo = Gets the max count of names list in original table aka if my name was listed 20 times but yours was listed 5 it would return 20

Variable CompareTraits gets the distinct count where the person is being compared against the Max count of names listed and counting the traits to return a value

Variable Table is filtering the result of the PersonTrait variable against the results of CompareTraits

Variable Result is dividing the results of Table by the number of possible traits (5)

If this was helpful, vote for Greg.  If not let me know and I can try again or maybe Greg can clarify if I messed up

[AA364]

@Greg_Deckler Here is a screen shot of the source data. I had to unpivot in PowerBi to get it to do what I want - but I think you get the general idea.

[Greg_Deckler]

@AA364 See if the attached PBIX below signature helps. Did this with a disconnected table and this measure:

% Alike = 
    VAR __PersonTraits = DISTINCT('Table'[Value])
    VAR __CompareTo = MAX('Names'[Name])
    VAR __CompareTraits = DISTINCT(SELECTCOLUMNS(FILTER(ALL('Table'), [Name] = __CompareTo),"__Traints",[Value]))
    VAR __Table = FILTER(__PersonTraits, [Value] IN __CompareTraits)
    VAR __Result = DIVIDE( COUNTROWS(__Table), 5 )
RETURN
    __Result

[Greg_Deckler]

@AA364 What does your data look like? Can you post a sample as text?