Summarize giving different data if I add extra groupBy_columnName

[jmab]

I have data that looks something like this, with other columns of data, too. 

UserID	SessionID	BC
5	1	0
5	2	0
5	3	0
6	4	1
6	5	0
7	6	0
7	7	0
7	8	1
8	9	0
8	10	0
9	11	0
10	12	1
10	13	1
11	14	0
11	15	1

Some of the data have blanks and/or incorrect data, as there are new categories being added and bugs fixed at each of our releases.

My goal is to get a count of the unique UserIDs that never have a BC count of 1, and then build a histogram to count the number of sessions each user has done.

I've used this to separate users into buckets:

UserIDUniqueness =

VAR varCurrentValue = myData[UserID]

VAR varInstances =

    COUNTROWS(

        FILTER(

            myData,

            myData[UserID] = varCurrentValue

        )

    )

var Result =

    IF(

        varInstances = 1,

        "Unique",

        IF(

            varInstances > 1 && varInstances <= 5,

            "Bucket1",

            IF(

                varInstances > 5 && varInstances <= 10,

                "Bucket2",

                IF(

                    varInstances > 10 && varInstances <= 15,

                    "Bucket3",

                    IF(

                        varInstances > 15 && varInstances <= 20,

                        "Bucket4",

                        IF(

                            varInstances > 20 && varInstances <= 25,

                            "Bucket5",

                            IF(

                                varInstances > 25 && varInstances <= 30,

                                "Bucket6",

                                "BucketMore"

                            )

                        )

                    )

                )

            )

        )

    )

RETURN

    Result

And this works. Now, when I create the histogram table, I am doing:

UserUniqueness = FILTER (

    SUMMARIZE(

        'myData',

        myData[ApplicationUserType],

        "UniqueTotal", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Unique",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"

        ),            

        "Bucket1Total", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Bucket1",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"    

        ),

        "Bucket2Total", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Bucket2",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"  

        ),

        "Bucket3Total", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Bucket3",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"  

        ),

        "Bucket4Total", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Bucket4",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"  

        ),

        "Bucket5Total", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Bucket5",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"    

        ),

        "Bucket6Total", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "Bucket6",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"    

        ),

        "BucketMoreTotal", CALCULATE(

            DISTINCTCOUNT('myData'[UserID]),

            myData[UserIDUniqueness] = "BucketMore",

            myData[ApplicationVersion] <> "0.0.0.0",

            myData[ApplicationVersion] <> "0.1.0.0",

            myData[ApplicationVersion] <> "20.3.0.0"

        )

    ),

    NOT(myData[ApplicationUserType] = "")

)

and this gives me a nice histogram.

BUT, if I add another groupBy_columnName

UserUniqueness = FILTER (

    SUMMARIZE(

        'myData',

        myData[ApplicationUserType],
        myData[ApplicationVersion],

        "UniqueTotal", CALCULATE(

        .....

then my histogram has totally different data counts, which I don't understand.

Any ideas?

[asitm]

@jmab  it's going to be simpler if you use one of the histograms from appsource - there are both free and paid ones.

https://appsource.microsoft.com/en-us/marketplace/apps?search=histogram&ocid=1158848_20231030_histog... 

[parry2k]

@jmab seems like you are over killing everything but the reason could be because this produce a unique row

        myData[ApplicationUserType],
        myData[ApplicationVersion],

and the unique count of users for the above combination which will be different if you are doing summaries only on usertype