Measure type is converting NULL values in fields to the value "0". How can I show that it's NULL?

Question & Answer
, ,
1

I’m trying to create a dashboard to track an operational workflow.

I want to know if a question called is_supporting is: True / False / NULL.

Null would indicate to me that the question hasn’t been answered by a staff member. However - the null values are all showing up as “0”, make it difficult to discern where the question was answered as “FALSE” or whether the question hasn’t been answered.

  • Changing it from a measure to dimension type did not help
  • Creating a calculated field like - ifelse(isNull({IS_SUPPORTING), “No response”, {IS_SUPPORTING} = 1, “Yes”, “No”) also didn’t work; still seeing "No"s for everything.

How can I do this?

1 Like
2

Thank you for asking QuickSight question in this community. Could you tell me the data type in field of is_supporting? Or the data type specified for the calculated field. Thanks.

1 Like
3

@Xing is_supporting is a Boolean in our data - TRUE/FALSE/NULL.

In the quicksight dataset, being considered a “measure”.

4

@shohini Boolean data type is not natively supported in QuickSight. See reference below: Supported data types and values - Amazon QuickSight. The workaround is to change it to another data type.

5

hi @Xing, I am experiencing the same issue. The data on my DB side is TRUE/FALSE/NULL, when importing it into a Quicksight dataset, the fields get mapped to a numeric type {TRUE: 1, FALSE: 0, NULL: 0}. Is there away to set this mapping to preserve the NULL values instead of mapping them to 0 at the dataset level without having to change the underlying data coming from the DB?

6

FYI: I don’t mind if the boolean fields get mapped to a 1/0 numerical type if booleans are not natively supported, but NULLs getting mapped to 0 as well presents a big problem for my analyses.

7

Hi @slopez I have tagged your request as a feature request since there is no out of the box support yet.