Missing, null, or empty values in records

Keep the following considerations in mind when analyzing data in a dataset.

Null values

Missing data is represented as a null value in the record. Null values may appear in any column and should always be interpreted as no data. For example, you will see a null value when a question is skipped by a survey logic action.

Null value due to a survey logic action

The response below is an example of a null value due to a Survey Logic action. The condition was based the following Single Choice question:

Which country do you live in?

• Canada
• USA

If Canada is selected, the member will see this question:

If USA is selected, the member will see this question:

In the record below, the member selected USA, therefore value for the Postal Code question is null, and the value for the Zip Code questions is 94110.

...
{
    "columnId": "15a1d220-54bf-deef-2284-fa120c2b6450",
    "value": null
},
{
    "columnId": "33ea3db8-341c-853e-3e66-c2fe8c40e6e9",
    "value": "94110"
},

Empty values

A value may not be assigned to column in a record for a variety of reasons, such as the participant:

• Clicked Next without answering the question.
• Left the survey before they saw the question.

There are cases when a value is an empty string, an empty set, or a zero (0). This can be interpreted as a valid response as opposed to missing data. For example, a survey contains the following Multiple Choice question:

Which type of loyalty programs are you enrolled in? Select all that apply.

• Cash back
• Discount
• Frequency
• Points

Participants who do not belong to a loyalty program, will skip the question. Therefore the empty set value in the column is a valid response.

Parent topic: Datasets