When might you decide to impute missing data?

• A. When there is no missing data.
• B. When missing values could be plausibly estimated from other values.
• C. When you want to delete the dataset.
• D. When you need to simplify the data presentation.

Answer: (B)

Imputing missing data means replacing those missing values with plausible estimates so you can use the full dataset without dropping cases. You’d decide to impute when there’s enough information in the other data to make a reasonable guess about what the missing value should be. This helps keep sample size intact and can reduce bias that comes from simply removing incomplete records.

Think of it like using what you know from related information in the dataset to fill in gaps. For example, if a test score is missing, you might estimate it from other related variables or the overall pattern in the data. But imputation isn’t appropriate if there’s no basis to estimate the value, or if the reason a value is missing is tied to something you can’t observe—then imputing could mislead results.