Forum Discussion

afaherty's avatar
afaherty
Icon for Helper V rankHelper V
2 years ago

Ignore blanks when calculating %

Hello All,

I have researched this for hours but have been unable to resolve it.  Any help would be appreciated.

I have survey data.  Of course, sometimes respondents do not answer every question. 

Here is an example of my data:

Respondent IDThe class was very helpful
1Strongly Agree
2Agree
3 
4Agree
5Strongly Disagree
6Agree
7Strongly Agree
8 

 

I have a bar graph displaying the % of respondents who answered with each response.  It is sorted in order (Strongly Agree, Agree, Disagree, Strongly Disagree)

 

Here is my dax:

% = DIVIDE([RespondentIDUniqueCount],CALCULATE([RespondentIDUniqueCount],ALL('Survey'[The class was very helpful],'Survey'[The class was very helpful Order Sort])))
 
The problem is that I don't want those who didn't answer the question to be included in the calculation or the bar graph, for example, like this where it shows (Blank):
 

 

I do not want to simply filter out the blanks in the visual because then the total percentages do not equal 100%, of course. 

I should also add that I do not want to remove the entire row, as the person may have answered other questions on the survey but left that particular question blank.

 

Thank you for any help you can offer!

2 Replies

  • Dangar332's avatar
    Dangar332
    Icon for Resident Rockstar rankResident Rockstar

    HI, afaherty 

     

    use Visual level filter like below image made from your data

    in advance filter option choose isnotempty for The class was very helpful column

     

    • afaherty's avatar
      afaherty
      Icon for Helper V rankHelper V

      Thank you, but as I mentioned, that means then that the bars in the graph do not add up to 100%.  I would prefer to handle this via DAX.