According to the material, which method is recommended for handling missing data?

• A. Delete records
• B. Mean imputation
• C. Mode imputation
• D. Prediction with regression

Answer: (C)

Missing data create a problem because you want to use the data you have without biasing results or losing too much information. When the variable is categorical, filling in missing values with the most frequent category—the mode—is a practical and straightforward approach. It preserves the distribution of that category and keeps the dataset intact, unlike deleting records which reduces sample size and can bias analyses. Using the mean to impute would be inappropriate for categorical data, since a numeric average doesn’t make sense for categories and would distort the category frequencies. Regression-based imputation can be more accurate in some situations, but it adds complexity and relies on assumptions about relationships that may not hold. So, the material recommends mode imputation for categorical variables because it is simple, preserves the existing distribution, and keeps the data usable without introducing unwarranted complexity.