What is the recommended approach when units are inconsistent across a dataset?

• A. Ignore unit differences
• B. Convert to a common unit
• C. Remove all records with any unit differences
• D. Use only the first unit and assume others are the same

Answer: (B)

When units are inconsistent, you need to convert all measurements to a common unit so comparisons and analyses are meaningful. This standardization makes every value represent the same scale, allowing accurate aggregation, modeling, and visualization. For example, if some lengths are in meters and others in centimeters, choose a target unit (like meters) and convert every value to that unit. This preserves the true quantities and eliminates misleading differences that come from different unit labels.

Ignoring unit differences leads to false comparisons, because a value in a larger or smaller unit counts differently even if the quantity is the same. Removing records with unit differences wastes data and can bias results by reducing sample size and representativeness. Using only the first unit assumes all records share that unit, which isn’t guaranteed and can introduce large errors.

So, converting to a common unit is the correct approach.