Definition

In data science and marketing analytics, the practice of filling in missing data with educated guesses when reality refuses to cooperate with your spreadsheet. It's essentially statistical fortune-telling that lets you pretend your dataset is complete. Data scientists treat it as sophisticated methodology; everyone else calls it making stuff up with math.

Example Usage

We used multiple imputation to handle the 40% of survey responses where people just clicked through without reading, because deleting that data would make our sample size embarrassing.

Source: Data analytics and statistics terminology

Related Terms