What happens to dataset columns when creating object types?

• A. Columns are dropped.
• B. Each column in the backing dataset becomes a property in the new object type.
• C. Columns become separate object types.
• D. Columns are stored as metadata only.

Answer: (B)

When you create an object type from a dataset, each column becomes a property of that object type. This maps the tabular schema directly into the object model, so every column name defines a corresponding field on the object type with an inferred data type (for example, strings, numbers, dates). Each row in the dataset can then be represented as an instance of that object type, carrying values for all those properties.

This is why the other options aren’t right: you don’t drop columns, you don’t split them into separate object types, and the values aren’t stored only as metadata—the actual data from each column becomes the value of the corresponding property in the object type.