A data dictionary with missing values represented by -99. Which of the following ways of conceptualizing 'nothing' is the dataset using?

A data dictionary with missing values represented by -99. Which of the following ways of conceptualizing 'nothing' is the dataset using?

• A. An attribute has a particular state that indicates nothingness (null)
• B. An attribute has a particular value that indicates nothingness
• C. An attribute exists but has no value (empty)
• D. An attribute doesn’t exist (missing)

Answer: (B)

A sentinel value that encodes missing data. In this dataset, -99 is stored in a field to represent that the observation is not available, so the value is not a real measurement but a flag meaning “missing.” Because the field actually contains a number (−99), this isn’t a true null, an empty value, or an absence of the attribute. It’s a deliberate encoding: the attribute has a particular value that signifies nothingness. When processing, you’d treat −99 as missing and handle it accordingly, rather than treating it as a valid data point.

A sentinel value that encodes missing data. In this dataset, -99 is stored in a field to represent that the observation is not available, so the value is not a real measurement but a flag meaning “missing.” Because the field actually contains a number (−99), this isn’t a true null, an empty value, or an absence of the attribute. It’s a deliberate encoding: the attribute has a particular value that signifies nothingness. When processing, you’d treat −99 as missing and handle it accordingly, rather than treating it as a valid data point.