Hot deck is a method broadly used in surveys for imputing the missing values. In its easiest form the method includes sampling with replacement m values from the sample respondents Ar to an item y, where m is number of non-respondents to the item and r is the number of the respondents. The sampled values are taken in use in place of the missing values. The precision of imputation is improved by first forming two or more imputation classes making use of control variables observed in all sample units, and then applying the process separately within each imputation class for each item with the missing values.