Forum Discussion
Distinct values based on multiple columns
If it helps, it SUPER easy to do what you are asking in reverse... You can select all 3 columns, then click 'Group By' and create a new Count Rows column. Now the count will be 2 for 123456 and 1 for everyone else? (Meaning that grouping of 3 data columns had 2 occurences, where the rest only had 1 distinct occurance of hte 3 columns.)
Hope this helps a little...
FOrrest
If it helps, it SUPER easy to do what you are asking in reverse... You can select all 3 columns, then click 'Group By' and create a new Count Rows column. Now the count will be 2 for 123456 and 1 for everyone else? (Meaning that grouping of 3 data columns had 2 occurences, where the rest only had 1 distinct occurance of hte 3 columns.)
Hope this helps a little...
FOrrest
Sorry for the newbie question, but how do I get to the screen you are showing. I have gone to modeling where I would usually create a new table, but I can't select more than one column at a time. Are there videos or tutorials somewhere?
When you 'Get Data' select 'Edit' during the import process, or after import select 'Edit Queries' from the Home Tab of the Desktop Client.
I was able to do the group by and it counted right, but then none of my graphs worked. I will have to play with it some more.
Glad you are almost there! Remember your graphs will treat the Grouped Data as if it was the original data source, so take into account the Count column when looking for multiple enteires that are no longer there. Thank you, FOrrest