Which statement is true about SCD in general?

Which statement is true about SCD in general?

• A. They never preserve history
• B. They keep track of change in dimension values and are used to report historical data at any given point in time
• C. They require no keying columns
• D. They always remove historical data after a period

Answer: (B)

Slowly changing dimensions are designed to handle changes to dimension attributes over time. The main idea is to capture how a dimension value changes and to keep that history so you can report what the data looked like at any point in time. This enables accurate point-in-time analysis and historical reporting, such as knowing a customer’s address or a product’s price as it existed on a specific date. For example, when a customer moves, you don’t simply overwrite the old address. Instead, you store a new version of the customer row with the new address and the period during which it was valid. This lets you reconstruct the state of the data as of any date. That’s why this statement is the best fit: it emphasizes tracking changes in dimension values and supporting historical reporting at any given time. The other options don’t fit because SCDs are about preserving history (not eliminating it), they typically rely on keys to identify and version rows, and while some specific approaches may purge data under certain policies, preserving history is a core reason for using SCDs.

Slowly changing dimensions are designed to handle changes to dimension attributes over time. The main idea is to capture how a dimension value changes and to keep that history so you can report what the data looked like at any point in time. This enables accurate point-in-time analysis and historical reporting, such as knowing a customer’s address or a product’s price as it existed on a specific date.

For example, when a customer moves, you don’t simply overwrite the old address. Instead, you store a new version of the customer row with the new address and the period during which it was valid. This lets you reconstruct the state of the data as of any date.

That’s why this statement is the best fit: it emphasizes tracking changes in dimension values and supporting historical reporting at any given time. The other options don’t fit because SCDs are about preserving history (not eliminating it), they typically rely on keys to identify and version rows, and while some specific approaches may purge data under certain policies, preserving history is a core reason for using SCDs.