Forum Discussion
Correlation between multiple columns
- 6 years ago
Have a look at this:
https://xxlbi.com/blog/pearson-correlation-coefficient-in-dax/ - Anonymous6 years ago
Hey sturlaws Thanks for your help.
Now i am stuck in the visualisation part of it , i have 16 measures for different correlation with 16 different questions.
I just wanna show a simple bar chart but i guess its not possible to make measure as X-Axis name.
Have a look at this:
https://xxlbi.com/blog/pearson-correlation-coefficient-in-dax/
Hey sturlaws Thanks for your help.
Now i am stuck in the visualisation part of it , i have 16 measures for different correlation with 16 different questions.
I just wanna show a simple bar chart but i guess its not possible to make measure as X-Axis name.
- sturlaws6 years agoResident Rockstar
Not possible to put measures on x-axis directly, true, but there is a workaround.
Create a stand alone table, that is a table without any relationships. Say your 16 measures are named measure1, measure2,...,measure16.The stand alone table, let's call it tabMeasures, should have one column, let's call the column Measures. The values in this column should be text strings. For simplicity you can use the same names as your 16 measures, but it is not strictly necessary:
The values on the rows could/should corresponds to your measures, in your case you should have 16 rows. Add a measure to this table like this
xAxisMeasure = SWITCH ( TRUE (); SELECTEDVALUE ( Table[measureName] ) = "measure1"; [measure1]; SELECTEDVALUE ( Table[measureName] ) = "measure2"; [measure2];
.
.
.
SELECTEDVALUE ( Table[measureName] ) = "measure16"; [measure16]; BLANK () )Now add the column of the new table to the x-axis, and add this measure as the measure
- Anonymous6 years agoNot applicablesturlaws You are a genius man. Thank You so much.
- Anonymous4 years agoNot applicable
Could you show me how to make the variable Y dynamic?