What technique replaces missing values with the mean or median of existing data?

What technique replaces missing values with the mean or median of existing data?

• A. Data Scaling
• B. Informative Missingness
• C. Imputation
• D. EDA

Answer: (C)

Filling in missing values with the mean or median is a form of imputation. Imputation is the process of substituting missing data with estimated values so that analyses can proceed as if the dataset were complete. Using the mean or median leverages the central tendency of the observed values to stand in for the missing ones, which keeps the overall sample size intact and allows algorithms that require complete data to run. This approach is simple and commonly used, but it comes with trade-offs. It can reduce variability and potentially bias relationships between features, especially if data are not missing completely at random. More advanced methods, like multiple imputation or model-based imputation, aim to preserve uncertainty about the missing values and often yield more reliable results. Other options described do not match this technique: data scaling adjusts the range or distribution of features but doesn’t fill in gaps; informative missingness refers to missing values signaling something about the data rather than filling them in; and exploratory data analysis is about uncovering patterns in the data rather than imputing missing values.

Filling in missing values with the mean or median is a form of imputation. Imputation is the process of substituting missing data with estimated values so that analyses can proceed as if the dataset were complete. Using the mean or median leverages the central tendency of the observed values to stand in for the missing ones, which keeps the overall sample size intact and allows algorithms that require complete data to run.

This approach is simple and commonly used, but it comes with trade-offs. It can reduce variability and potentially bias relationships between features, especially if data are not missing completely at random. More advanced methods, like multiple imputation or model-based imputation, aim to preserve uncertainty about the missing values and often yield more reliable results.

Other options described do not match this technique: data scaling adjusts the range or distribution of features but doesn’t fill in gaps; informative missingness refers to missing values signaling something about the data rather than filling them in; and exploratory data analysis is about uncovering patterns in the data rather than imputing missing values.