https://www.gooddata.com/ logo
Join Slack
Powered by
Hello GD team - need some help on the feature "sho...
# gd-beginners
r
Hello GD team - need some help on the feature "show missing values". I've been reading other posts here and documentations on this option and FOR EACH formula, but still something seem to work oddly when I use this in visualizations. We have a test account with 113 users. If I bring them into a table visualization with metric "count of users", I see all 113. The moment I add further dimensions such as metric "Count of Completions" and attribute "Content name", by checking the "show missing values" box on both user name and content name, I am expecting to see all the 113 users with a null value for any content they have not completed. But I am seeing just a limited amount of users. Is there any step missing you think? Thanks in advance
f
Hi Raffaella! That’s an interesting question. I’ve been working on my test environment and couldn’t quite replicate the scenario you’re describing. Would you mind adding a screenshot of the insight, so that I can see how you’re adding all of those metrics/attributes? Also, are all of these attributes/facts in the same dataset, or different ones? Thanks!
r
Hi Francisco - I've sent you the screenshot via DM
They are all in same dataset but different tables
Just to follow up on this thread - we have looked at the data examples with Francisco via DM, and realized if we bring the content attribute as a column, the "show missing value" option on users won't work. Solution was to add the content as a metric filter and remove attribute from filter and from column area of a table visualization However we still experience the below issues: • When null values are included, only one of the attributes (full name) is showing, other attributes in the same table or linked table becomes (empty value) even if the user table has data in those attributes. • When clicking "show missing values" on the other attributes in the table report, we have duplicate rows (see below) E.g. ID | Full name | Location | Completion status ID 9866 | empty value | empty value | NUll Empty value | John Smith | empty value | Null Empty value | empty value | London | Null (when John Smith's ID is indeed 9866 and his location is London) We need to be able to save a simple report with user attributes and their completion status. Is the above a bug or is there an issue in the data?
f
Hi Raffaella, thank you for updating the thread. I changed tacks a little bit after our previous discussion, and have found a more appropriate solution. I built a metric using the CASE conditional statement to ensure that every result (including Nulls) would be captured. In your case, it’d look something like this:
Copy code
SELECT CASE WHEN (SELECT COUNT(Content Name)) >=1 THEN 1 ELSE 0 END
When adding this metric to the insight, you’ll get only 1s and 0s - 1 representing a completion, and 0 (which include any Null values, in my tests) representing it was incomplete. Using this metric, I was able to slice the insight by various attributes in the Rows section, and also add an attribute filter to narrow down the Content, whilst still displaying the full list of users in the database. Please take a look at the article I linked and the example metric and let me know if it works out better 🙂
Note: the main attribute/fact being Counted inside the CASE metric will depend on your LDM, and what you are using to count these completions; I used 
content name
in my example to illustrate only.
r
Thanks Francisco. We will look into this solution. Much appreciated!!!#
🙌 1
Open in Slack
PreviousNext
Powered by