A Review of Hot Deck Imputation for Survey Non‐response

Abstract

Summary Hot deck imputation is a method for handling missing data in which each missing value is replaced with an observed response from a “similar” unit. Despite being used extensively in practice, the theory is not as well developed as that of other imputation methods. We have found that no consensus exists as to the best way to apply the hot deck and obtain inferences from the completed data set. Here we review different forms of the hot deck and existing research on its statistical properties. We describe applications of the hot deck currently in use, including the U.S. Census Bureau’s hot deck for the Current Population Survey (CPS). We also provide an extended example of variations of the hot deck applied to the third National Health and Nutrition Examination Survey (NHANES III). Some potential areas for future research are highlighted.

Away-from-home eating: nutritional status and dietary intake among Brazilian adults

“…The methodology used for data imputation can be found in other publications (15,16) . Briefly, data were subjected to a system of review and automatic imputation known as Critique and Imputation for Quantitative Data (CIDAQ).…”

Methods for Addressing Missing Data with Applications from ACS Exams

“…5 Many of these methods require that data is MCAR or MAR for accurate estimates. [1][2][3][4][5][6][7][8]14,15 Two of the most basic single imputation techniques are mean and linear regression imputation. Mean imputation involves replacing missing values with the average of the observed data.…”

“…3 Many additional single imputation methods exist and have been discussed in detail elsewhere. [1][2][3][4][5][6][7][8]14,15 In addition to single imputation, there are a few model-based procedures that have been described as "modern" 5 or "state of the art" 3 in missing data techniques. One method is multiple imputation (MI), which works similarly to the regression-based techniques described previously, except that a residual term is added to help restore a loss in the variability of the estimated data.…”

“…In this method, missing responses are filled in with values from other "similar" participants in the observed data. [1][2][3]14,15 Hot-deck procedures are well-known for being used by the U.S. Census Bureau for the Current Population Survey (CPS). 14,15 Hot-deck methods have the benefits of being conceptually simple to understand and easy to implement without creating complex models, and (like other imputation methods) they can provide a common complete data set for use by multiple analysts.…”

“…14 Additionally, because the imputed values are derived from actual participants, the values will never be outside the possible range of responses; for example, a value of a 73 will never be imputed for a test where only 70 points are possible. 15 Finally, hot-deck works with categorical missing …”